Consumers’ Food Choice at a Restaurant

Depending on Nutritional Information and Nutritional Menu Context

Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy
in the Graduate School of The Ohio State University

By

Hae Jin Yoon, M.S.

College of Education and Human Ecology

The Ohio State University

2009

Dissertation Committee:

R. Thomas George, Advisor

Margaret Binkley

Curtis Haugtvedt

Kathryn Stafford

i
Copyright by

Hae Jin Yoon

2009

i
 Abstract

Americans are gaining weight. As such, the issue of obesity has received the

increased attention. One of the most recent concerns is how the food industry is

contributing to the problem of obesity and related health problems; increased

consumption of food away from home, food which often contains high levels of calories

and fats, may be worsening the problem of obesity. Thus, while restaurants make an

effort to develop and introduce healthy food, the government has also addressed the

problem by introducing The Menu Education and Labeling Act (MELA), which requires

restaurants to present nutritional information on the menu. Such legislation has been

proposed or passed in several cities and states. Accordingly, the importance of and the

interest in nutritional information disclosure has increased. A thorough examination of

the effects of nutritional information disclosure is required in order to achieve the benefits

of nutritional information disclosure for restaurateurs, consumers, and policy makers.

The purpose of this study was to provide a better understanding of how

consumers process nutritional information in a restaurant and which factors influence

consumers’ nutritional information processes and food choices. More specifically, this

study sought to determine the effect of providing nutritional information on a menu, as

well as the moderating effects of several factors such as nutritional menu context,

motivation to process nutritional information, and nutritional knowledge.

In order to achieve this purpose, an experimental design method was employed to

ii
collect data. An experimental design was developed using three key variables: nutritional

information level, nutritional level of target items, and nutritional level of menu context.

Using the results of two pilot studies, favorable menu items available in restaurants and

six important nutrient contents were selected; levels of nutritional values for the six

nutrient contents were selected for each menu item.

Multiple regression analysis was conducted to test the proposed hypotheses, and

significant results were found. First, the presence of nutritional information had a

negative effect on consumer food evaluation and decisions, regardless of the healthiness

of the item at hand. Additionally, the healthiness of other alternatives on the menu

significantly influenced consumer food decision processes as well as the effect of

nutritional information disclosure. Thus, it was concluded that the placement of menu

items on the menu is very important in promoting healthy item choices. It is important to

educate and encourage consumers to process nutritional information accurately because

motivation to process enhances the effects of nutritional information disclosure.

The results of this study provide useful information not only for restaurateurs,

who may be able to develop more efficient marketing strategies using nutritional

information, but also for consumers, who will be able to make better choices when eating

foods away from home.

Key terms: nutritional information, menu context, motivation, nutritional knowledge

iii
 Dedication

Dedicated to all my family:

With their love and support, I was able to reach this point in my life.

iv
 Acknowledgements

First of all, I would like to thank my advisor, Dr. Thomas George, whose advice,

support, and patience helped me complete this study. Without his encouragement and

guidance, I would not have been able to complete this dissertation or my Ph.D. I also

thank all my committee members. I am very grateful to Dr. Binkley who strongly

encouraged and supported me in completing this dissertation and in preparing my future

work. Another expression of thanks goes to Dr. Haugtvedt, who helped in conducting the

experiment and provided guidance for my research and dissertation. I also wish to thank

Dr. Stafford whose invaluable suggestions and feedback improved my dissertation and

encouraged me to be critical and analytical. I am also indebted to Dr. H.G. Parsa who

was my former advisor. His invaluable guidance and advice helped me to complete my

Ph.D. coursework successfully and contributed to my doctoral education.

I wish to express appreciation to my family, especially my parents, Hwan Byung

Yoon and Jung Rim Ryu, whose love, support, and patience have allowed me to achieve

this position. Very sincere thanks go to my sisters and their families whose emotional

support kept me going through all the difficult periods during my studies in Columbus.

Finally, I am very grateful to my friend, Jungkee Jang, who always supported and

encouraged me. I also thank my friends and colleagues. Without their support, my

graduate studies would not have been the same. Thanks to all.

v
 Vita

1998……………...…………………….……B.S. Business Administration, Dongduck

Women’s University, South Korea

2000………………………….……………...M.A. Business Administration, Ewha

Woman’s University, South Korea

2004……………………………………..…..M.S. Statistics, The Ohio State University

2005 to Present…………………………..…..Graduate Research and Teaching Associate,

The Ohio State University

Publications

1. Yoon, H., Nusair, K., Parsa, H.G., & Naipaul, S. (2010). Pricing Discounts and
Consumers Perceptions: A Comparison Between Hospitality and Non-
Hospitality Industries. Journal of Foodservice Business Research. 13(Dietary
Guidelines for Americans)

2. Yoon, H., Thompson, S.S. & Parsa, H.G. (2009). Bayesian Approach to Assess
Consumers’ Brand Selection Process and Identification of Brand Attributes in
Service Context. International Journal of Hospitality Management, 28 (1), 33-
41.

3. Nusair, K., Yoon, H., & Parsa, H.G. (2008). Effect of Utilitarian and Hedonic
Motivations on Consumer Satisfaction with Travel Websites. Journal of
Information Technology & Tourism, 10 (1), 75-89.

4. Serovich, J.M., Craft, S., & Yoon, H. (2007). Women’s HIV Disclosure to Immediate
Family. AIDS Patient Care and STDs, 21 (12), 970-980.

Fields of Study

Major Filed: Hospitality Management

vi
 Table of Contents

Abstract............. .................................................................................................................. ii 

Dedication.............. ............................................................................................................ iv 

Acknowledgements ..............................................................................................................v 

Vita.................. ................................................................................................................... vi 

List of Tables .......................................................................................................................x 

List of Figures ................................................................................................................... xii 

CHAPTER 1: INTRODUCTION ........................................................................................1 

1.1 Background ............................................................................................................1 

1.2 Statement of the Problem .......................................................................................7 
1.3 Purpose of the Research .......................................................................................10 
1.4 Significance of the Research ................................................................................ 11 
1.5 Organization of the Dissertation ..........................................................................12 
1.6 Definition of Terms ..............................................................................................13 
Nutritional Information ......................................................................................13 
Healthy Food ......................................................................................................13 
Motivation to Process Nutritional Information ..................................................14 
Nutritional Knowledge.......................................................................................14 
Nutritional Menu Context ..................................................................................14 
CHAPTER 2: LITERATURE REVIEW ...........................................................................16 

2.1 Theoretical Framework ........................................................................................16 

2.1.1 The Elaboration Likelihood Model (ELM)...............................................16 
2.1.2 Context Effect ...........................................................................................19 

vii
 2.2 Nutritional Information on the Menu ...................................................................21 
2.2.1 The Influence of Providing Nutritional Information on the Menu ...........21 
2.2.2 The Influence of Nutritional Menu Context .............................................26 
2.2.3 The Role of Motivation to Process Nutritional Information .....................29 
2.2.4 The Role of Nutritional Knowledge .........................................................33 
2.3 Summary of Literature Review and Hypotheses .................................................37 
CHAPTER 3: METHODOLOGY .....................................................................................41 

3.1 Experimental Design ............................................................................................41 

3.2 Measurement of Variables....................................................................................43 
3.2.1 Dependent Variables .................................................................................43 
3.2.2 Independent Variables ...............................................................................45 
3.2.3 Moderating Variables ................................................................................46 
3.2.4 Other Survey Questions ............................................................................49 
3.3 Pilot Study I .........................................................................................................49 
3.3.1 Procedure ..................................................................................................50 
3.3.2 Data Analysis and Results.........................................................................50 
3.4. Pilot Study II .......................................................................................................56 
3.4.1 Procedure ..................................................................................................56 
3.4.2 Data Analysis and Results.........................................................................57 
3.5 Primary Study ......................................................................................................62 
3.5.1 Sample.......................................................................................................62 
3.5.2 Sample Size ...............................................................................................62 
3.5.3 Study Procedure ........................................................................................64 
3.5.4 Data Analysis ............................................................................................65 
CHAPTER 4: RESULTS ...................................................................................................66 

4.1 Sample..................................................................................................................66 
4.2 Reliability and Internal Validity of Measures ......................................................69 
4.3 Manipulation Check of Menu Items ....................................................................71 
4.4 Hypotheses Test ...................................................................................................73 
4.4.1 The Influence of Providing Nutritional Information on the Menu ...........74 
viii
 4.4.2 The Influence of Nutritional Menu Context .............................................78 
4.4.3 The Role of Motivation to Process Nutritional Information .....................85 
4.4.4 The Role of Nutritional Knowledge .........................................................91 
4.5 Additional Analysis: Logistic Regression Analysis .............................................98 
CHAPTER 5: CONCLUSIONS ......................................................................................102 

5.1 Summary of Results and Discussion .................................................................102 

5.2 Conclusions and Implications ............................................................................109 
5.3 Limitations and Future Research ....................................................................... 112 
References.... ....................................................................................................................116 

Appendix A: Questionnaire of Pre-test I .........................................................................124 

Appendix B: Questionnaire of Pre-test II ........................................................................135 

Appendix C: Examples of Menu......................................................................................148 

Appendix D: Results of Additional Multiple Regression Analysis .................................155 

Appendix E: Objective Nutritional Knowledge Questions ..............................................158 

ix
 List of Tables

Table 2.1: Proposed Hypotheses.........................................................................................38

Table 2.2: Overview of Previous Studies............................................................................39

Table 3.1: Experimental Design..........................................................................................43

Table 3.2: Study Instruments..............................................................................................48

Table 3.3: Evaluation of Menu Items..................................................................................54

Table 3.4: Mean Values of Nutritional Perception of Menu Items......................................58

Table 3.5: Manipulated Nutritional Values of Menu Items..........................................…....61

Table 4.1: Sample Size........................................................................................................66

Table 4.2: Demographics of the Sample.............................................................................68

Table 4.3: Results of Exploratory Factor Analysis..............................................................70

Table 4.4: Results of Reliability Analysis...........................................................................71

Table 4.5: Mean Values of Nutritional Perception of the Menu Items without
Nutritional Information.......................................................................................72

Table 4.6: Results of Regression Analysis: Effect of Nutritional Information

Disclosure...........................................................................................................76

Table 4.7: Predicted Values of Dependent Variables depending on Nutritional

Information.........................................................................................................77

Table 4.8: Results of Regression Analysis: Effect of Nutritional Menu Context...............82

x
Table 4.9: Predicted Values of Dependent Variables without Interaction Terms b/w
the Nutritional Menu Context and the Nutritional Information Level...............85

Table 4.10: Results of Regression Analysis: Effect of Motivation to Process

Nutritional Information......................................................................................89

Table 4.11: Results of Regression Analysis: Effect of Objective Nutritional

Knowledge.................................................................................................... ...94

Table 4.12: Results of Regression Analysis: Effect of Subjective Nutritional

Knowledge........................................................................................................97

Table 4.13: Results of Hypotheses Testing.........................................................................98

Table 4.14: Results of Logistic Regression Analysis........................................................101

Table D.1: Results of Regression Analysis: Effect of Nutritional Knowledge on

Source Credibility............................................................................................156

Table D.2: Results of Regression Analysis: Interaction Effect b/w Nutritional

Knowledge and Motivation to Process...........................................................157

Table E.1: Frequency of Correct Answers of Objective Nutritional Knowledge

Questions.........................................................................................................159

xi
 List of Figures

Figure 2.1: Conceptual Framework of the Study................................................................37

Figure 4.1: Main Effect of Nutritional Information Disclosure....................................... ..78

Figure 4.2: Interaction Effect between the Presence of Nutritional Information and the
Nutritional Menu Context on Nutritional Perception.......................................83

Figure 4.3: Interaction Effect between the Presence of Nutritional Information and the
Nutritional Menu Context on Overall Food Attitude..................................... ..83

Figure 4.4: Interaction Effect between the Presence of Nutritional Information and the
Nutritional Menu Context on Purchase Intention.............................................84

Figure 4.5: Interaction Effect between the Presence of Nutritional Information and the
Motivation to Process on Nutritional Perception........................................... ..90

Figure 4.6: Interaction Effect between the Presence of Nutritional Information and the
Motivation to Process on Overall Food Attitude..............................................90

xii
 CHAPTER 1: INTRODUCTION

1.1 Background

Americans are gaining weight. The rate of obesity in the U.S. has unexpectedly

increased over the past 30 years. According to The National Health and Nutrition

Examination Survey (NHANES), almost half of American adults were overweight or

obese in the 1970s; by 2004, this proportion had increased to more than two thirds of

American adults. The growing rate of obesity has received significant attention in recent

years since it is one of the critical causes of several public health problems, including

cardiovascular disease, hypertension, stroke, and diabetes mellitus (Centers for Disease

and Control Prevention (CDC), 2007b). Consequently, the increasing rate of obesity has

led to an increase in medical expenditures. In 1997, the direct costs associated with being

overweight and obese amount to approximately 5.7% of total U.S. medical expenditures;

this figure continues to rise steeply (Wolf & Colditz, 1998). In 2002, the estimated

medical costs associated with obesity reached 92.6 billion dollars, which accounted for

9.1 % of total U.S. medical expenditures (CDC, 2007a). Obesity has therefore become

one of the fastest growing health issues; accordingly, research has been undertaken to

explain the causes of the obesity epidemic and to prevent the increasing rate of obesity.

From the perspective of economics, a variety of topics related to obesity have

been considered: the role that technological change plays in explaining trends in obesity,

1
how maternal employment affects obesity rates, the impact obesity has on wages and

health insurance premiums, and behavioral economic explanations of obesity . For

instance, the time and effort that women need to prepare food was reduced by a

technological revolution bringing about improvements, such as equipment for food

preparation, food packaging, delivery systems, as well as widespread microwave use

(Cutler, Glaeser, & Shapiro, 2003). By economizing time and effort, women spend fewer

calories on their work at home, thus creating an imbalance between calorie intake and

expenditure. This is considered to be one of the causes of obesity. In another study,

Chou, et al. (2004) identified several significant predictors of body mass index and the

probability of being obese: household income, the price of food at home, the price of

quick-service and full-service restaurants, the price of alcohol, as well as the price of

cigarettes.

Researchers in the area of social behavior have proposed lifestyle change as

another explanation for the obesity problem. Due to new technology, physical activities

undertaken both at the workplace and out-of-work have decreased (Philipson & Posner,

2003). People are also more likely to spend their time participating in watching

television or playing games rather than performing housework or other physical activities

(Cutler, et al., 2003). This sedentary lifestyle may contribute to obesity by disrupting the

balance between energy intake and expenditure. Many studies have provided evidence

for this lifestyle based explanation of obesity (Hill, Wyatt, Reed, & Peters, 2003; Hu, Li,

Colditz, Willett, & Manson, 2003).

More recently, researchers have become concerned that the food industry itself is

an important contributor to the problem of obesity and its related health problems, since

2
expenditures on food away from home are increasing. This includes all food purchased

and consumed away from home such as at restaurants, schools, or hospitals. According

to the U.S. Department of Agriculture (USDA, 2008), in 2007 money spent on food away

from home represented 49% of total household food expenditures, a figure which had

increased from 33% in 1970. It has been projected that expenditures on food away from

home will reach about 51% of total food expenditures by 2016; thus exceeding the

expenditure on food consumed at home. Additionally, Americans who are over 7 years of

age dined at restaurants on average 218 times per year; people with household incomes

over $75,000 ate at a restaurants more often (4.9 times per week) than did people with

incomes below $15,000 (3.2 times per week) (Ebbin, 2000). A number of studies have

indicated that this increase in food consumption away from home has a significant impact

on obesity rates (Binkley, Eales, & Jekanowski, 2000; Bowman & Vinyard, 2004; Chou,

et al., 2004; French, Story, Neumark-Sztainer, Fulkerson, & Hannan, 2001). This

phenomenon may be explained by the fact that many foods offered by full-service and

quick service restaurants – two main segments of the food away from home market – are

energy dense foods, high in both calories and fat (Glanz, Sallis, Saelens, & Frank, 2005;

Hill & Peters, 1998). For example, a king size Burger King Double Whopper with

cheese contains approximately 2,180 calories, more calories than the total recommended

daily allowance (2,000 calories). Furthermore, it has been shown that children, on

average, consumed more calories (770 calories) at a restaurant than they did when they

ate at home (420 calories) (Zoumas-Morse, Rock, Sobo, & Neuhouser, 2001).

Accordingly, more frequent food consumption at restaurants should tend to increase the

rate of obesity.

3
 As one of the suspected parties responsible for obesity, food companies have

been blamed and threatened with fines, restrictions, and legislation (Wansink & Chandon,

2006). Above all, quick service restaurants such as McDonalds and Burger King have

been blamed for the increase in obesity rates. As a result, they have received bad press

and have been sued by their customers. Food companies and legislators have come

together in order to overcome this problem: not only are food companies making an effort

to develop and provide more healthy options, but the U.S. government has also

encouraged consumers to purchase healthy food and has proposed the law referred to as

The and Nutrition Labeling Education Act (NLEA) for packaged foods and The Menu

Education and Labeling Act (MELA) for restaurants.

The NLEA was proposed and implemented by the U.S. Food and Drug

Administration (FDA). According to the NLEA, the labels of packaged food products

must list the nutritional information with daily reference values for the following

nutrients: total calories, total fat, saturated fat, trans fat, cholesterol, sodium, sugars,

carbohydrates, dietary fiber, protein, vitamin A, vitamin C, calcium and iron (FDA, 1994).

The purpose of this law is not only to increase consumers’ use of nutritional information,

but also to help consumers in choosing healthier foods based on the information provided

to them (Burton & Andrews, 1996). Byrd-Bredbenner (2000) found that 29% of

consumers always read and used nutritional labels, while 66% of consumers sometimes

read and used nutritional labels when purchasing food for home consumption. In

addition, in response to the NLEA, food product manufacturers evaluated and modified

existing packaged food and introduced nutritionally improved new products (Center for

Science in the Public Interest (CSPI), 2008).

4
 However, the NLEA applies only to packaged foods because it is easy to evaluate

nutritional values of standardized products. The legislation excludes food consumed at

the point of purchase, such as in restaurants or cafeterias, on airplanes, and from vending

machines (Kozup, Creyer, & Burton, 2003). Although some chain restaurants provide

nutritional information of menu items through brochures or on corporate websites,

displaying nutritional information is not required on menus in restaurants. It is difficult

to present nutritional information of all menu items due to limited space on menus, lack

of standardized foods, and various costs (Almanza, Nelson, & Chai, 1997). As the

expenditure for food away from home and the calorie intake from eating out have

continuously increased, the lack of nutritional information for foods sold or served in

restaurants has become increasingly controversial.

A separate law requiring nutritional information on restaurant menus has been

proposed in several states and is under review by Congress; this law is referred to as the

Menu Education and Labeling Act (MELA). The MELA law would require chain and

quick service restaurants to provide nutritional information such as the total amount of

calories, saturated fat, trans fat, carbohydrates, and sodium contained in menu items

(Cranage, Conklin, & Lambert, 2004). Recently, New York City enacted legislation

which requires restaurant chains with 15 or more establishments to list total calories of

items on the menu, while California enacted the California Menu Labeling Bill, requiring

that nutritional information be displayed for fast-food and chain restaurants with 20 or

more outlets. Philadelphia City Council also passed a menu labeling law which requires

restaurants with at least 15 outlets nationwide to provide nutritional information for four

nutrients in addition to total calories count (Nation’s Restaurant News (NRA), 2008).

5
Additionally, San Francisco and King County in Washington have enacted menu labeling

laws, and more than 20 states, cities and counties, including Montgomery County in

Maryland, Nassau and Westchester Counties in New York, and Multnomah County in

Oregon, were preparing menu labeling proposals in 2008 (CSPI, 2008).

As a result, restaurants are reducing calorie content in their recipes and portion

sizes. For instance, Starbucks is saving the calories by swapping whole milk for 2%, and

McDonalds reduced the calorie count of their french fries from 570 to 500 (Kim, 2008,

October 29). Many other restaurants such as Quiznos and Dunkin Donuts have

introduced smaller sized meals or less caloric food. In addition, consumers are given

access to nutritional information and can therefore choose healthier foods (Cranage, et al.,

2004). According to the report released by Center for Science in the Public Interest

(Wootan, 2003), almost 75% of consumers used the nutrition fact label while purchasing

packaged food, and about half of consumers picked healthier over unhealthier food based

on the nutritional information provided. Thus, the menu labeling law has inspired policy

changes nationwide, and restaurants and food companies have noticed the influence that

nutritional information disclosure on the menu can have on consumers.

With this increased interest in nutritional information, it is becoming necessary to

explore the impact that nutritional information has on consumer food purchase behavior

in restaurants. Several studies found that healthiness of food is one of the important

components consumers consider when making food choices, and that consumers are more

likely to purchase healthy foods with fewer calories or less fat when the nutritional

information is provided on the menu (Burton & Creyer, 2004; Edwards & Meiselman,

2005; Kozup, et al., 2003; Stubenitsky, Aaron, Catt, & Mela, 2007). However, it remains

6
unknown which factors encourage consumers to process the nutritional information, and

how nutritional information can best promote healthy food choices. Therefore, it is

essential to study the impact that nutritional information presented on the menu has on

food choices, and to identify the factors that influence how nutritional information is used

in food evaluation.

This study thus focused on the effect that nutritional information disclosure has on

consumers’ food evaluation, as well as the effects of other influential variables on

nutritional information processing and food purchase behavior in the restaurant context.

The results of this study will be useful not only for restaurateurs, who may be able to

develop more efficient marketing strategies using nutritional information, but also for

consumers, who will be able to make better choices when eating foods away from home.

1.2 Statement of the Problem

As the concern for health increases, so has the demand for healthy food. In order

to satisfy consumers’ wishes, a variety of healthy menu items have been introduced in

restaurants and the Menu Education and Labeling Act (MELA) for restaurants has been

proposed. A recent Harvard Poll found that 62% of Americans support this regulation

(Finkelstein, French, Variyam, & Haines, 2004).

However, many studies addressing nutritional information have not been

conducted in a restaurant setting. According to Dorms (2006), consumers use different

information in making decisions at restaurants than they use when deciding about what

food to consume at home. The information used differs because these two decisions

occur in different situational contexts; private or public places. In public places,

7
consumers contemplate their self-image or self-regulation if they are accompanied by

others, excluding family and friends, because it is important that consumers favorably

impress others by their choices. There was evidence that consumers tend to select

healthier menu items when dining with colleagues or business partners (Hochradel, 2007).

Thus, the first purpose of this study was to examine the extent to which the nutritional

information presented on menus influences consumer food evaluation and choice at a

restaurant.

Prior research has demonstrated that consumers are more likely to have a positive

food attitude and increased consumer purchase intention when provided with healthier

nutritional information on a menu. In contrast, food attitude and purchase intention

decrease when unfavorable nutritional information is provided (Burton & Creyer, 2004;

Cranage, et al., 2004; Kozup, et al., 2003). Additionally, consumers tend to

underestimate the number of calories or the amount of fat contained in food offered by

restaurants (Chandon & Wansink, 2007b). There is evidence that consumers’

expectations of the number of calories, amount of saturated fats, and sodium levels are

significantly different from the actual levels contained in restaurant food; this difference

may be even greater for less healthy foods than for healthier foods. However, when

nutritional information is presented on the menu, consumers tend to make healthier menu

item choices (Burton, Creyer, Kees, & Huggins, 2006). Thus, it is suggested that

providing nutritional information positively affects healthy eating behavior.

In contrast, other studies showed that providing nutritional information on menus

has no effect on the use of information, the evaluation of food, or consumer food choices

in a restaurant setting (Droms, 2006; Stubenitsky, et al., 2007). The lack of effect was

8
explained by suggesting that people consider eating out a special event where they are

allowed to eat any food, regardless of health (Stubenitsky, et al., 2007). Dorms (2006)

also suggested that, when making food choices, people consider other important and

influential components besides health benefits, such as the taste of the food or their food

preferences. Moreover, awareness and use of presented nutritional information may be

influenced by consumers’ individual characteristics, such as motivation to perform

healthy behavior, health consciousness, nutritional knowledge, and health status

(Moorman, 1990). Among these characteristics, the impacts of nutritional knowledge and

motivation to perform healthy behavior, especially motivation to process nutritional

information, have been explored in prior studies targeting packaged food products

(Burton, Garretson, & Velliquette, 1999; Keller, et al., 1997; Moorman & Matulich,

1993). The finding common to the studies was that consumers are more likely to have a

favorable attitude toward healthy food and purchase healthy food when they have high

nutritional knowledge and high motivation to process nutritional information. However,

despite the significant role nutritional knowledge and motivation play in food decisions,

subsequent studies have not investigated how these traits may affect consumer food

choices in restaurants. Accordingly, it should be determined how nutritional knowledge

and motivation to process nutritional information affect evaluations of menu items at

restaurants, depending on the presence of nutritional information.

Another influential factor in processing nutritional information is menu context.

Menu context is determined by the set of all available food items on a menu. Consumer

food choice may be affected by the menu context since consumer food preferences and

choices are dependent on all available options at the point of purchase. In evaluating a

9
specific menu item, consumers compare it with alternatives using information such as the

description or price of the menu items (Dhar & Simonson, 1992). For example, if new

price information is encountered or one of the prices in a product set is changed,

consumers usually compare the new price with the prices of all available products (Lynch,

Chakravarti, & Mitra, 1991). This is called the context effect; alternatives influence

consumer choice. Evidence of the context effect has been found in several previous

studies (Huber, Payne, & Puto, 1982; Huber & Puto, 1983; Simonson, 1989; Simonson &

Tversky, 1992). Similarly, the nutritional information associated with each available

alternative on the menu should relatively influence consumer food evaluation and choice

within the menu context. Nonetheless, the influence that menu context has in terms of

nutritional information processing has been examined in only a few studies (Burton &

Creyer, 2004; Kozup, et al., 2003); prior studies were further limited to include only

unhealthy target items or unhealthy contexts. Although the results from previous studies

showed a significant context effect on food attitude and purchase intention, these finding

must be further generalized: it is necessary to understand how consumers compare and

evaluate nutritional information of available options on the menu and also to determine

whether the context effect exists for both healthy and unhealthy items.

1.3 Purpose of the Research

The aim of this study was to investigate how the presence of nutritional

information on a menu affects consumers’ food evaluation and food purchase behavior in

a restaurant. Specifically, the objectives of this study were as follows:

(1) To investigate whether the presence of nutritional information on the menu

10
 will influence consumer food decision making in a restaurant context.

(2) To identify the influence of menu context in which a specific menu item is

evaluated when nutritional information is provided.

(3) To determine how the impact of nutritional information may be moderated

by consumers’ motivation to process nutritional information and nutritional

knowledge.

1.4 Significance of the Research

The problem addressed by this study is extremely important for a number of

reasons. First, the study adds to the limited amount of literature which considers the

influence of nutritional information presented on menus. In particular, only a few

research papers have investigated the effect of nutritional information disclosure in a

restaurant and have mainly focused on the effects of nutritional menu context depending

on the presence of nutritional information. Thus, the results of this study extend the

literature; they also provide support for menu labeling laws and help to create effective

strategies for the restaurant industry.

Second, the results of this study provide important insight into the way consumers

understand and apply nutritional information when purchasing food in a restaurant.

Restaurants could benefit by becoming aware of the potential effects nutritional

information have on consumer choices, and to promote the benefits. In addition,

motivation to process nutritional information is important factor in moderating the effect

of nutritional information, and nutritional knowledge directly affect consumer food

choices. Thus, it is critical to educate consumers about nutrition and the importance of

11
health.

This study also identified which menu items consumers choose among several

alternatives depending on the presented nutritional information. Thus, it helps

restaurateurs to undertake better marketing strategies which emphasize healthy foods

using nutrition labels and to develop menu items which balance between healthy and

unhealthy foods.

This study is ultimately expected to provide information on how to better

educate consumers such that they increase the use of nutritional information on menus

and control their food consumptions for health. It also aids restaurateurs in understanding

how consumers use nutritional information in the menu decision process, and allows

restaurant owners to use better marketing strategies to promote healthy foods.

1.5 Organization of the Dissertation

This dissertation is organized in five chapters: Chapter 1, the current chapter,

states the problem and the purpose addressed by the study. It outlines the significance of

the dissertation and discusses the potential contributions to both academic and practical

fields. Chapter 2 reviews the relevant literature to provide an understanding of the

impacts nutritional information and other factors have on decision making in a restaurant

setting. This chapter also includes the proposed hypotheses and conceptual model based

on previous literature. Chapter 3 addresses the methodology in developing the

experimental design and explains the statistical methods used. Chapter 4 presents the

results of data analysis and hypothesis testing. Chapter 5 discusses the findings of the

research and its limitations, and concludes the dissertation with theoretical and

12
managerial implications.

1.6 Definition of Terms

Nutritional Information

Nutritional information is defined as information concerning the nutrition content

of various foods as compared to the nutritional requirements necessary to support life.

Nutritional information includes the amount, in grams, of serving size, and the nutrients

included in that serving, like total calories, calories from fat, total fat, saturated fat, trans

fat, cholesterol, sodium, total carbohydrates, dietary fiber, sugar, protein, vitamin A,

vitamin C, calcium, and iron (Wikipedia, 2009).

Healthy Food

According to the criteria published by the Food and Drug Administration (FDA),

“healthy food” refers to “a food that is low in fat and saturated fat and that contains

limited amounts of cholesterol and sodium”. The recommended daily values of the above

four nutrients, based on a diet of 2,000 calories a day, are the following: less than 65g of

total fat, less than 20g of saturated fat, less than 300mg of cholesterol, and less than

2400mg of sodium. Furthermore, “if it is a single-item food, it must provide at least 10

percent of one or more of vitamins A or C, iron, calcium, protein or fiber. If it is a full

course meal including entrée and dessert, it must provide at least 10 percent of two or

three of vitamins A or C, iron, calcium, protein or fiber, and the content of sodium can not

exceed 480mg per serving” (FDA, 1999).

13
Motivation to Process Nutritional Information

According to Moorman and Matulich (1993, p.210), health motivation is defined

as “consumers’ goal-directed arousal to engage in preventive health behaviors”. That is,

health motivation refers to consumers’ willingness to perform, or interest in performing

healthy behaviors including nutrition label information acquisition from media, family, or

friends, life-balancing behaviors, positive diet addition, negative diet restriction, alcohol

moderation, and tobacco nonuse. However, this study focused on the motivation to

process given nutritional information with the goal of healthy eating behavior in a

restaurant setting.

Nutritional Knowledge

Nutritional knowledge refers to “the extent to which consumers have enduring

health related cognitive structures”. Namely, it is an ability to use the nutrition and health

information accurately (Moorman & Matulich, 1993, p.210).

Nutritional Menu Context

Bettman, et al. (1991) used the term “context” to refer to a choice set that includes

other characteristics of the choice environment, and Simon and Tversky (1992, p.281)

defined “context” as “the set of alternatives under consideration”. Based on these

definitions, menu context can be defined as a group of all available menu items that a

consumer evaluates when choosing items off the menu.

With respect to nutritional information, Kozup, et al. (2003, p.20) used the term

“nutritional context”; that is “the nutritional frame created by other menu items”. On the

14
other hand, Burton and Creyer (2004, p.122) employed the term “nutritional frame”,

which is “the context within which the menu item is presented”. Thus, in this study, the

term “nutritional menu context” was used to refer to a menu context containing the

nutritional information associated with each menu item.

15
 CHAPTER 2: LITERATURE REVIEW

In order to achieve the research objectives proposed in Chapter 1, prior studies

relating to restaurant menu labeling and healthy eating behavior have been reviewed.

The literature review provides the theoretical foundation for this study and the rationale

for the conceptual model and hypotheses. Specifically, this chapter reviews necessary

theories of how nutritional information in restaurants shapes consumer food choice and

outlines other significant factors that influence food choices.

2.1 Theoretical Framework

There are many theories which seek to explain consumer behavior in the market

place. Among them, two fundamental theories were drawn upon to defend the structure

and ideas of the present study: The Elaboration Likelihood Model (ELM) and Context

Effect. These models provide a method of examining the basic procedures consumers

use in information processing, and help identify the key influential factors involved in

information processing. As such, these models guided the conceptual model of this study.

2.1.1 The Elaboration Likelihood Model (ELM)

The Elaboration Likelihood Model (ELM), introduced by Petty and Cacioppo

(1981), was used in this study to account for the process that consumers use when they

16
become aware of nutritional information and use it in making food choices. The ELM is

generally employed to model how information or messages in persuasive communication

influence judgment or attitude change in both social and consumer psychology. The

ELM clarifies the process of how consumers build or change their attitudes through two

different routes: the central route and the peripheral route (Petty & Wegener, 1999).

Through the central route, consumers’ attitudes may be stronger or changed by careful

consideration of information, whereas attitude changes through the peripheral route are

based not on diligent consideration of information but rather on a simple inference from

the related cues in the persuasion context (Petty, Cacioppo, & Schumann, 1983).

Therefore, consumers generally make more effort to elaborate information or messages

through the central route than they do through the peripheral route.

In particular, motivation and ability to process information plays an important role

in accessing and elaborating information through the central route (Petty & Wegener,

1999). Motivation to process information or messages usually depends on individual or

situational factors such as the need for cognition, involvement, or personal relevance

(Bagozzi, Gurhan-Canli, & Priester, 2002). If the provided information increases a

consumers’ perceived involvement, they will be more likely to elaborate given

information (Petty & Cacioppo, 1984). For example, if nutritional information is

provided to consumers who are very health conscious, they should be more motivated to

use it in making food choices (Chandon & Wansink, 2007b). In addition to motivation,

the ability to process information in light of distraction, repetition, or knowledge is

another important factor in the information elaboration process. For instance, if

consumers are distracted, they will be less able to make a cognitive effort; on the other

17
hand, consumers who are knowledgeable in a subject should be more willing to spend

time and effort thinking about it (Petty & Brock, 1981) . When consumers have enough

motivation and ability to process the relevant information or message content, they are

more likely to elaborate all relevant information carefully.

Furthermore, arguments or cues play an important role in the information

elaboration process. The information available in the persuasion situation can be

evaluated on a variety of criteria, such as quality, source, message, and context

information (Petty & Wegener, 1999). According to the ELM, when consumers have

sufficient motivation and ability to process, they may focus on information itself and

evaluate the information using any one of these criteria. In particular, the quality of

argument, including such traits as the number of arguments and the source’s level of

expertise, is more important for consumer. On the other hand, consumers under the

peripheral route who lack motivation or ability to process the presented information use

only small amounts of information or source attractiveness as criteria for information

evaluation.

Consequently, consumers who process information or messages through the

central route are expected to engage in cognitive effort; the thoughts and attitudes

obtained through this process should be more reasonable, relatively persistent, and

resistant to opposite persuasion (Haugtvedt & Petty, 1992; Petty, Haugtvedt, & Smith,

1995). Additionally, attitudes created or changed by the elaborated process are more

likely to represent consumers’ behavioral intention and behavior than are attitudes which

result from less cognitive processes (Bagozzi, et al., 2002).

18
2.1.2 Context Effect

The ELM says that motivation and ability to process information, along with

quality of information provided, directly influence consumers’ information processes; the

resulting thoughts or attitudes may be favorable or unfavorable. Studies meant to

examine the ELM have introduced various significant variables that impact the

elaboration process: mood, the recipient of the message, the number of message sources,

and time pressure (Petty & Wegener, 1999). Although the influence of context, which is

defined by all available alternatives in a choice set, has not been examined in prior

studies relating to the ELM, it remains one of the prominent variables in persuading

consumers to purchase a product. When consumers process given information or

advertising messages with the intention of making a purchase, they must evaluate and

compare information for all available options within the choice context. Thus, attitude

toward a specific product or choice can be influenced by information for or evaluation of

other alternatives. In this case, it is very important which options are included in the

context.

Context effect, which is the impact of “the set of alternatives under consideration”,

was introduced by Simon and Tversky (1992). According to several researchers, context

effect is explained by the principle of tradeoff contrast. Tradeoff contrast describes the

phenomenon by which a product appears more attractive when it is surrounded by less

attractive alternatives than when it has a background containing more attractive options

(Lynch, et al., 1991; Simonson & Tversky, 1992). Thus, depending on the choice set in

which a specific product is presented, the preference for the product should differ. Simon

and Tversky (1992) divided the context effect into two types: background context effect

19
and local context effect. Background context effect describes the case when past options

influence a current choice; local context effect occurs when a target item is compared

with current alternatives in the context. In the example of food selection at a restaurant,

all available menu items should influence the choice of a particular menu item; this is an

example of the local context effect. Thus, if nutritional information, one attribute by

which menu items may be evaluated and judged, is provided on the menu, consumer food

attitude or choice should be influenced by the nutritional values of other alternatives on

the menu. On the other hand, past experiences of menu selection at other restaurants may

be the reference point by which menu items in the current context are evaluated and

judged. In this case, the information of menu items in past contexts influences the choice

of a menu item in the current context; this is called the background context effect.

Based on the ELM and context effect models this study examined the effects that

nutritional information disclosure on the menu and other significant factors may have on

consumer food evaluation and choice. In this study, three different levels of nutritional

information were provided and explored: (1) food description without nutritional

information; (2) food description with information for the total calorie content only; and

(3) food description with nutritional information including six key nutrition contents.

Depending on the provided nutritional information levels, consumers’ nutritional attitude

toward a menu item, their overall attitude toward a menu item, and their purchase

intention are evaluated. This study also examined the roles that motivation to process

nutritional information and nutritional knowledge play in consumer food evaluation.

Finally, how the context effect affects evaluation and selection of a menu item depending

on the level of nutritional information provided is investigated.

20
2.2 Nutritional Information on the Menu

2.2.1 The Influence of Providing Nutritional Information on the Menu

The effect that nutrition labels or nutrition claims have on consumers’ perceptions

and purchase behavior for packaged food products consumed at home has been

extensively demonstrated. Studies have explored this relationship from various

perspectives: level of nutritional information (Burton & Andrews, 1996; Garretson &

Burton, 2000), package design (Ford, Hastak, Mitra, & Ringold, 1996; Keller, et al.,

1997), nutrition fact format (Kozup, Burton, & Creyer, 2006), message type (Andrews,

Netemeyer, & Burton, 1998; Garretson & Burton, 2000), and demographics (Axelson &

Penfield, 1983; Nayga, Lipinski, & Savur, 1998). They provided evidence that

nutritional labels on packaged food products have significant associations with nutritional

knowledge, attitudes toward food products, and increased use of label information in food

choices (Burton & Andrews, 1996; Burton, et al., 1999; Garretson & Burton, 2000;

Marietta, Welshimer, & Anderson, 1999).

Studies targeting restaurant menu choices have been recently completed with

similar results found in the restaurant environment (Burton & Creyer, 2004; Cranage, et

al., 2004; Hwang & Lorenzen, 2008; Kozup, et al., 2003). Cranage, et al. (2004) studied

how providing nutritional information impacts customer satisfaction, purchase intention

and healthy food choice. To find the effect of presented nutritional information, they

chose hot entrée menu items in an on-campus café and analyzed the nutritional content of

menu items using the USDA database. They collected sales data for the items over a two-

week period during which nutritional information was not provided. They then provided

nutritional information for the items, including portion size, calories, fat, calories from fat,

21
cholesterol, sodium, and protein, and collected data for another two weeks. The results of

this study revealed that nutritional information disclosure at the point of purchase

increased customer satisfaction for food quality and future purchase intention for the food;

consumers may feel that they have been more responsible in their choice after being

exposed to nutritional information. This research also concluded that consumers were

willing to select more healthy foods when nutritional information was provided.

In order to examine consumer evaluation of menu items when providing health

claims and nutritional information on the menu, Kozup, et al. (2003) conducted three

different experiments. The first two experiments were designed to test the effects of

nutritional information for a packaged food (a frozen lasagna) and a restaurant menu item

(a lasagna entrée), respectively. Members of a consumer household research panel

participated in the two experiments. The researchers assessed consumer evaluation

measures (food and nutrition attitude, purchase intention, and source credibility) and

disease risk measures (heart disease and stroke) associated with a certain product,

depending on whether a heart healthy claim was presented on the product. Other

nutritional values were also manipulated, such as favorable (healthy) or unfavorable

(unhealthy) values, and the measures were reevaluated. For both the packaged food and

the menu item, favorable nutritional information positively influenced consumers’ food

evaluation and disease risk perception. In contrast, the heart health claim only affected

consumer attitude toward the nutrition of the packaged food and the perceived source

credibility for the menu item. Consumers’ food evaluation and disease risk perception for

the menu item were also significantly affected by an interaction between the heart healthy

claim and nutritional information provision. Thus, the provision of healthful nutritional

22
information with a heart healthy claim increased attitude toward nutrition, overall food

attitude, and purchase intention, whereas providing unhealthful nutritional information

with a heart healthy claim reduced consumers’ positive attitude toward a food and

purchase intention.

In the third experiment conducted by Kozup, et al. (2003), a nutritional context

variable was added. This additional variable measured the nutritional level of other

available menu items. The researchers used three different chicken menu items and

manipulated the nutritional values for one target item and the nutritional context, which

included the nutritional information for the other two items. For this study, participants

were recruited at a shopping mall. In contrast to the two prior experiments, in this

experiment providing nutritional information about the target item had a significant

impact only on attitude toward nutrition. Nutritional context, however, significantly

affected food attitude, purchase intention, and disease risk perception.

Burton and Creyer (2004) built upon the third experiment conducted by Kozup et

al. (2003). They focused on the effects of nutritional context by providing nutritional

information. Using four different menu items, they manipulated the nutritional values of

both healthy and unhealthy target items, while the nutritional values of all non-target

items were also manipulated such that they appeared unhealthier than the target item.

Consumers perceived an item as increasing their risk for heart disease and cancer in the

case when nutritional information was present on the menu; the evaluation of nutrients

such as fat, saturated fat, and cholesterol significantly differed depending on the presence

of nutritional information. However, providing nutritional information did not have

significant main effects on overall food attitude and purchase intention.

23
 In recent research, Burton et al. (2006) demonstrated that consumers were not

aware of the exact levels of nutrients in foods offered at a restaurant, and tended to

underestimate the content of nutrients such as calories, fat, saturated fat, and sodium. In

particular, the differences between the nutritional values expected by consumers and the

actual values were greater for unhealthy foods than for healthy foods. Based on this

result, the researchers conducted the experiment to explore how consumers change their

attitudes, purchase intention, and food choices when their expected nutritional values

differ from the provided actual values. For the test, they selected four different food

items, and measured the expected nutritional values for each of the four items.

Consistent with their survey, the expected nutritional values for two unhealthy items

(chef’s salad and hamburger with fires) tended to be inconsistent with the actual

nutritional values, while the expected values for the other two items (grilled chicken

breast with baked potato and turkey sandwich), which are relatively healthy items, tended

to be closer to the actual values. The results also showed that the presence of nutritional

information on the restaurant menu had a significant main effect on food attitude, and

consumers’ food choice was influenced by the level of nutritional information provided.

When only total calorie count was presented, consumers were more likely to select a

healthier menu item (turkey sandwich). On the other hand, the menu item type

influenced all dependent variables: food attitude, perception of associated risk of weight

gain and heart disease, and purchase intention. For the two foods with high calorie

content and high fat content, the hamburger and chef’s salad, consumer purchase

intention decreased when nutritional information was provided on the menu. Consumers

also changed their food choice if actual calorie and fat content was higher than expected.

24
Accordingly, the provision of nutritional information helped consumers evaluate nutrients

of menu items more accurately and led them to choose healthier foods at a restaurant.

Hwang and Lorenzen (2008) have also examined the effects of providing

nutritional information on restaurant menus. Firstly, the researchers checked if

consumers felt that different nutritional labeling on the menu was helpful in evaluating

food healthiness. They provided five different levels of nutritional information: no

information, only total calories, total calories plus macronutrients, total calories and

macronutrients plus fat, and total calories, macronutrients, and fat plus fiber. They then

measured how helpful consumers deemed the nutritional information presented on each

label, as well as the perceived source credibility. As more nutritional information was

added, consumers felt that the provided nutritional information was more helpful and

more credible. Furthermore, consumer preference for a given menu item increased as

additional nutritional information was provided. In the second study, researchers

compared the nutrition related attitude, overall food attitude, and willingness to pay for

an unhealthy menu item (southwest chicken sandwich) under three different conditions:

no nutritional information, provision of nutritional information, and provision of

nutritional information where the item was advertised as containing only 50% of the

original fat content (low fat). When nutritional information was provided, consumers had

a more unfavorable nutrition attitude, as well as a more negative overall food attitude.

But consumers perceived that when the menu item was presented under the low fat

condition with nutritional information, it was healthier than the item with original fat

content. Thus, consumers recognized the change in fat content. Consumers were also

likely to pay $2 more for the menu item under the low fat condition.

25
 In conclusion, previous studies have demonstrated that providing nutritional

information on the menu influences consumer food evaluation and choice. However, the

influence of nutritional information disclosure on overall food attitude and purchase

intention varied depending on other variables such as menu context and individual

characteristics. In addition, most previous studies employed an unhealthy target menu

item with unfavorable nutritional values. Thus, this study sought to determine the effect

that nutritional information has on consumers’ food decisions in restaurants when other

influential variables are held constant. Based on the literature reviewed herein, the

following hypothesis is suggested:

H1: The presence of nutritional information on the menu will significantly

influence consumers’ food decision making, especially consumers’ nutritional perception,

overall food evaluation, and purchase intention.

2.2.2 The Influence of Nutritional Menu Context

There is evidence that consumers’ preferences and product choices are affected by

choice context (Dhar, Nowlis, & Sherman, 1999; Simonson & Tversky, 1992). In

comparing the utilities or subjective values associated with each alternative, the consumer

tends to choose the alternative with the highest utility or subjective value. Thus, in

analyzing consumer choice, it is very important to determine not only which products are

in the choice set, but also which products act as comparisons. For instance, consumers

were more likely to choose a product in a context consisting of alternatives with unique

good features than in a context consisting of alternatives with unique bad features (Dhar,

et al., 1999). Also, if consumers perceived that the nutritional values of the available

26
options were similar, they were more likely to search for additional information to

distinguish the products; this search behavior influenced the product evaluation and

decision behavior (Balasubramanian & Cole, 2002). But, the information search intensity

decreased with increasing similarity of nutrition across brands after the Nutrition

Labeling and Education Act legislation had been enacted. As a result, the nature of the

context effect on food choices when nutritional information is provided needs to be

further investigated; similarly, only a few studies have been conducted in the restaurant

setting (Burton & Creyer, 2004; Kozup, et al., 2003). This study addresses these

deficiencies.

Kozup, et al. (2003) explored whether providing nutritional information for

alternatives in a choice set influenced consumer food evaluation, purchase intention, and

disease risk perception in the restaurant setting. To test the menu context effect, they

selected three items (slow-roasted chicken, chicken marsala, and grilled chicken fajitas);

of these three items, one was designated the target menu item and two were designated

alternatives. The following three levels of information were provided for the nutrition

levels of both the target item and the menu context: no information, healthy and

unhealthy. The researchers assumed that the nutritional values of the target item would

act as a reference point by which all items on the menu would be evaluated. Thus, item

evaluation and purchase intention would be influenced by the nutritional menu context of

alternatives on the menu. They found that the nutritional context of alternative menu

items had a significant effect on attitude toward a menu item and purchase intention.

Specifically, the nutritional context significantly moderated the effects that health claims

and nutritional information had on consumer attitudes toward a menu item. Therefore,

27
the effect that healthy information about a target item had on consumers’ attitude toward

that item and its nutrition was less pronounced when the nutrient contents of non-target

items were healthy rather than when they were unhealthy.

Burton and Creyer (2004) built upon the experiment conducted by Kozup et al.

(2003) in order to examine and identify the nutritional context effect. According to

Simonson and Tversky (1992), consumers’ decision processes were affected by other

options in a consideration set, which is referred to as the local context effect. Thus,

Burton and Creyer (2004) suggested that consumers’ food evaluations and purchase

behaviors at a restaurant should be affected by the nutritional context of alternatives on

the menu. To test their proposal, they chose one target menu item (pot roast) and three

non-target items (lasagna, chicken fajitas, and chicken chow mien). They provided three

different levels of nutritional information for the target item (control, healthy, and

unhealthy information), but only unhealthy information was presented for non-target

items so as to provide an unhealthy context. Consumers’ food evaluations and disease

risk perceptions were measured for the target item and for one of the non-target menu

items. Compared to non-target menu items, attitudes toward the target item and purchase

intention for the target item were positively evaluated when the nutritional information

was offered. It was demonstrated that the presence of nutritional information for the non-

target menu items significantly influenced the attitude and purchase intention for the

target menu item. There also existed significant interaction effects between nutritional

information and nutritional context on overall food attitude and purchase intention for the

target menu item. The study showed that the nutrition levels of other options on the

menu should be considered to be just as important as the provision of nutritional

28
information for the item itself.

Thus, the nutrition levels of alternatives in a menu context have significant effects

on food evaluation, and also moderate the effects of nutritional information for the item

itself. From the previous literature, the following hypotheses are identified:

H2: The nutritional menu context will significantly influence consumers’ food

decision making, especially consumers’ nutritional perception, overall food evaluation,

and purchase intention.

H3: The nutritional menu context will significantly moderate the effect that the

presence of nutritional information has on consumers’ nutritional perception, overall food

evaluation, and purchase intention.

2.2.3 The Role of Motivation to Process Nutritional Information

Consumers’ individual characteristics play an important role in the food purchase

decision process. For instance, Burton and Andrews (1996) found that consumers’ age

negatively influenced the ability to understand a nutrition label, and that education level

was positively related to healthy food consumption (Saegert & Young, 1983; Yung,

Contento, & Gussow, 1984). In addition to age and education level, various health

characteristics have been considered in examining healthy behaviors: motivation to

perform healthy behavior, health knowledge, health status, health locus of control,

healthy behavior control, and income (Moorman & Matulich, 1993). Moorman and

Matulich (1993) used these individual health characteristics in predicting consumer

healthy behaviors such as health or label information acquisition from media, family, or

friends, life-balancing behaviors, positive diet addition, negative diet restriction, alcohol

29
moderation, and tobacco nonuse. These healthy behaviors were undertaken as preventive

measures, meant to avoid diseases or unhealthy states, and were divided into two groups:

health information acquisition behaviors and health maintenance behaviors. Through a

survey, the researchers measured all individual health characteristic variables and healthy

behaviors, and identified the relationship between healthy behavior and individual

characteristics. All health characteristic variables (motivation to perform healthy

behavior, health knowledge, health status, health locus of control, healthy behavior

control, age, education and income) were significant predictors of healthy behaviors.

Motivation to perform healthy behaviors especially moderated the effects of all health

characteristics on healthy behavior performance.

Among these individual health characteristics, the motivation to perform healthy

behaviors and the nutritional knowledge variables have been considered in studies meant

to identify the effects of nutritional information on healthy food choices. The reason for

this is that knowledge and motivation to process information are key factors that affect

perception and evaluation of given information (Petty, et al., 1983). Previous research

studies have demonstrated the impact of nutritional knowledge and motivation to process

nutritional information on consumer food decisions (Burton, et al., 1999; Droms, 2006;

Keller, et al., 1997).

Motivation to perform healthy behaviors is one of the significant factors that

influence healthy behaviors (Moorman & Matulich, 1993); this study focused specifically

on nutritional information process behavior. Balasubramanian and Cole (2002) and

Moorman (1990) investigated the nutritional information process, and identified the role

that motivation to acquire and process nutrition related information plays in the process.

30
In both nutritional information process models, motivation to process nutritional

information influenced information search behaviors and played a critical role in

evaluations of the nutrition of a product. Thus, the role that motivation to process

nutritional information plays in making healthy food decisions should be studied in a

restaurant setting.

Keller, et al. (1997) found that nutritional claims, nutritional values, and

motivation to process information impacted consumer evaluation and purchase intention

for a packaged food product. The researchers developed four types of nutrition claims

(no claim, 99% fat free, low fat, and low calorie) and manipulated the levels of nutritional

values (poor, medium, and good) for a frozen chicken dinner. Utilizing members of a

statewide household research panel, they measured each participant’s motivation to

process information and several dependent variables including attitude toward nutrition,

overall food attitude, perceived credibility of the product marketer, and purchase

intention. The results indicated that consumers’ level of trust for the information provider

significantly differed depending on the type of nutritional claims. It was also found that

nutritional value had significant effects on all dependent variables: perceived credibility

of the product marketer, nutrition attitude, overall product attitude, and purchase intention.

However, there was no main effect of motivation to process information on consumer

evaluation and purchase intention; motivation to process nutritional information did,

however, moderated the effects of nutritional claims and nutritional value. Consumers

with high motivation to process information had overall more favorable attitudes toward

nutrition as well as toward the product, and were more likely to purchase a product when

nutritional information was provided on the package. Therefore, motivation to process

31
nutritional information increases the accessibility of nutritional information and the

probability that the information will be processed. It also moderates the relationship

between nutritional information and consumers’ food evaluations.

A recent study investigated the tradeoff effect between carbohydrates and fat on

disease risk perception, purchase intention, and nutrition perception when nutritional

claims and nutritional information were provided (Kemp, Burton, Creyer, & Suter, 2007).

Researchers again found that the motivation to acquire and process nutritional

information had a moderating effect on the tradeoff effect. For this study, an experiment

was developed with two manipulated variables, each with three levels; nutritional fact

panel (no, low carbohydrate-high fat, and high carbohydrate-low fat), and nutritional

claim (no, low fat, low carbohydrate). Participants, who were members of a household

research mail panel for a southern state, were presented with one of the nine possible

scenarios created by combining the variables above, and responded to questions meant to

gauge motivation to process nutritional information, disease risk perception, nutrition

perception, and purchase intention. Nutritional facts, including the nutritional

information concerning the carbohydrate and fat contents of the product, impacted risk

perception of heart disease and stroke and nutrition perception, while the nutritional

claim variable only had a main effect on heart disease risk perception. The researchers

also found that motivation to process nutritional information had a significant influence

on the perceived risk of disease, but not on nutrition perception and purchase intention.

In contrast, motivation to process nutritional information moderated the effect that the

nutritional fact panel variable had on purchase intention. Thus, consumers with high

motivation to process nutritional information were likely to have more positive purchase

32
intentions when nutritional information was provided.

Accordingly, this review suggests that consumers’ motivation to process

nutritional information may moderate the main effects that nutrition level may have on

nutrition perception.

H4: The motivation to process nutritional information will significantly moderate

the effect that the presence of nutritional information has on consumers’ nutritional

perception, overall food evaluation, and purchase intention.

2.2.4 The Role of Nutritional Knowledge

Nutritional knowledge is another critical factor that has been examined in a

number of studies focusing on healthy eating behavior and food choice (Moorman, Diehl,

Brinberg, & Kidwell, 2004; Wardle, Parmenter, & Waller, 2000). Nutritional knowledge

refers to the ability to use nutrition and health information accurately (Moorman, 1993).

In particular, there is evidence that nutritional knowledge plays an important role in how

consumers use nutritional labels (Drichoutis, Lazaridis, & Nayga, 2005; Guthrie, Fox,

Cleveland, & Welsh, 1995; Levy & Fein, 1998) as well as in food evaluations and

choices (Andrews, et al., 1998; Droms, 2006; Kozup, et al., 2006).

Andrews, et al. (1998) investigated how the effects of nutritional ad claims were

changed by nutritional knowledge in advertising. The purpose of this study was to

identify the effects of nutritional claims depending on nutritional knowledge and

disclosure type (none, absolute amount of fat in grams, relative to the daily value,

evaluative), and ad claim type (no claim, general claim, and specific claim). From the

pretest, researchers developed new objective measures of nutritional knowledge, and then

33
manipulated disclosure type and ad claim type (“delicious” for no claim, “healthy” for

general claim and “no cholesterol” for specific claim). They recruited participants in a

shopping mall and showed a target advertisement after displaying two clutter

advertisements. The results indicated that ad claim type influenced how respondents

evaluated fat content and healthiness, whereas disclosure type had an impact on how they

evaluated fat content, healthiness and perception of heart disease and cancer risk.

Nutritional knowledge had a significant main effect on how respondents evaluated fat

content, but did not affect the other variables. In the advertisement showing the amount

of fat in grams (absolute type), consumers with high nutritional knowledge more

positively evaluated the fat content, and perceived the risk of cancer accordingly, than did

consumers with low nutritional knowledge. Thus, the moderating effect of nutritional

knowledge was limited to the perception of fat content.

Andrews, et al. (2000) conducted a similar study to examine the direct effects of

nutritional knowledge on consumers’ nutrition evaluations of products and the

moderating effects on the relationship between disclosure type and evaluation of nutrients.

Similar to the previous study, respondents were presented with one of two different ad

types, general expression (“Now Healthier than Ever”) and specific phrase (“Now 1/3

Less Salt”), and one of four different nutrition interventions (none, absolute, relative,

evaluative) in an advertisement for Campbell’s soup. The researchers collected data from

mall shoppers, and measured their evaluations of sodium content and healthiness, the

perception of high blood pressure and coronary heart disease risks, claim believability,

and attitudes toward the ad and the brand. Nutritional knowledge had a significant main

effect only on the evaluation of healthiness of the food, and moderated the effect that

34
nutrient disclosure type had on the evaluation of sodium content and the risk perception

of high blood pressure. Consumers with high nutritional knowledge more precisely

evaluated the sodium content in evaluative disclosure type ads than they did in relative

and absolute disclosure type ads. In contrast, nutritional knowledge did not have any

impact on claim believability or on attitudes toward the ad and the brand.

The effect of nutritional knowledge was also investigated by Kozup, et al. (2006).

The purpose of this study was to examine how nutritional knowledge and the presence of

trans fat information affected consumer evaluations and perceptions of disease risk. The

researchers focused on trans fat information since many foods such as crackers, snacks,

fried foods, and fast foods contain trans fat that causes long term adverse health effects.

To manipulate trans fat knowledge, the researchers artificially increased nutritional

knowledge by providing respondents from a high knowledge group with an article which

included trans fat information. They also manipulated the level of trans fat (no, low (1g),

and high (4g)) as well as the format of the nutrition fact (a per serving basis, both a per

serving and a per package basis). For this study, a box of snack crackers was selected as

the target item, and university students served as subjects. Induced knowledge directly

affected how respondents perceived the product in terms heart disease, stroke, and blood

pressure risk, and also moderated the effect that the actual level of trans fat had on those

three diseases risk perceptions. On the other hand, the level of trans fat had a significant

effect on heart disease and stroke risk perception, while nutrition fact format influenced

only blood pressure risk perception. Although induced knowledge and the level of trans

fat did not directly affect perceived importance of trans fat and nutrition perception, the

interaction effects between induced knowledge and level of trans fat were significant.

35
That is, consumers who were exposed to trans fat information considered trans fat as an

important component of food and perceived food as being less nutritious when trans fat

information was provided.

Most studies to date which focus on nutritional knowledge have been targeted

toward packaged food products. However, Dorms (2006) has examined the role

nutritional knowledge plays in dietary decision making in a restaurant setting. The first

study, which sought to discover the effects nutritional information has on consumer food

choices, did not obtain significant result. In order to identify other influential factors

relating to the nutritional information process, the researcher measured various variables

including self-esteem, self-objectification, and nutritional knowledge. Dorms (2006) also

manipulated nutritional information of menu items offered by national chain restaurants;

dietary restraint in food choices served as a dependent variable. The results indicated that

self-esteem and self-objectification were significant predictors of dietary restraint and

nutritional knowledge had a significant effect on food choice. That is, consumers high in

nutritional knowledge intended to select the healthier menu item at a restaurant.

As a result, nutritional knowledge seems to play an important role in making food

decisions. Based on the literature reviewed herein, the following hypothesis is offered:

H5: Nutritional knowledge will significantly moderate the effect that the presence

of nutritional information has on consumers’ nutritional perception, overall food

evaluation, and purchase intention.

36
 Based on the proposed hypotheses, the following conceptual model of this study

was developed.

Nutritional H2
Menu Context

H3 Consumers’ Food Decision Making

• Nutritional Perception
Nutritional H1
Information • Overall Attitude toward a Menu Item
• Purchase Intention

H4 H5

Motivation to Process Nutritional Knowledge

Nutritional Information

Figure 2.1: Conceptual Framework of the Study

2.3 Summary of Literature Review and Hypotheses

Previous literature provides enough evidence to show that consumers’ healthy

eating behavior in a restaurant could be influenced by external factors such as nutritional

information disclosure, as well as by consumers’ internal factors such as motivation to

process nutritional information and nutritional knowledge. When nutritional information

is offered at the point of purchase in a restaurant, consumers high in motivation to

process and nutritional knowledge tend to process the given information and tend to

compare with other alternatives on the menu using their knowledge ability. As a result,

the final evaluation of menu items through this procedure should lead consumers to

37
purchase healthier items in a restaurant. Based on prior studies, following hypotheses are

elicited, and the specific content of previous literature related to nutritional information is

summarized in Table 2.2.

Hypotheses
H1. The presence of nutritional information on the menu will significantly influence
consumer food decision making, especially consumers’ nutritional perception,
overall food evaluation, and purchase intention.
H2. The nutritional menu context will significantly influence consumers’ food decision
making, especially consumers’ nutritional perception, overall food evaluation, and
purchase intention.
H3. The nutritional menu context will significantly moderate the effect that the presence
of nutritional information has on consumers’ nutritional perception, overall food
evaluation, and purchase intention.
H4. The motivation to process nutritional information will significantly moderate the
effect that the presence of nutritional information has on consumers’ nutritional
perception, overall food evaluation, and purchase intention.
H5. Nutritional knowledge will significantly moderate the effect that the presence of
nutritional information has on consumers’ nutritional perception, overall food
evaluation, and purchase intention.

Table 2.1: Proposed Hypotheses

38
 Studies Sample Target Food Key Variables Findings
a b
Burton & Consumer Frozen chicken Level of Nutritional Value NA (+), PI (+), Nutrient Accuracy (+), Nutrition
Andrews research panel dinner product (NV) Information Awareness (0)
(1996) Nutrition Label Format NA (0), PI (0), Nutrient Accuracy (+), Nutrition
*Level of NV Information Awareness (0)
Age*Level of NV NA (+), PI (+), Nutrient Accuracy (0), Nutrition
information awareness (0)
Keller, et al. Statewide Frozen chicken Nutrition Content Claim NA (0), FAc (0), Credibility (+), PI (0)
(1997) household dinner product Level of Nutrition Value NA (+), FA (+), Credibility (+), PI (+)
research panel Motivation*Claim NA (+), FA (+), Credibility (0), PI (+)
Andrews, et General Margarine Ad claim type Fat content (+), Healthiness (+)
al. (1998) consumers product Disclosure type Fat content (+), Healthiness (+)
in 3 cities Nutritional knowledge Fat content (+), Healthiness (0)
Kozup, et al. Study 1 Study 1 Study 1
(2003) Consumer Frozen lasagna Heart HCd NA(+), FA(0), Credibility(0), PI(0), Heart (+), stroke(+)
household dinner NIe NA(+), FA(+), Credibility(+),PI(+), Heart(+), stroke(+)
research panel
39

Study 2 Study 2 Study 2

Consumer Lasagna entrée Heart HC NA(0), FA(0), Credibility(+), PI(0), Heart(+), Stroke(+)
household NI NA(+), FA(+), Credibility(+), PI(+), Heart(+), Stroke(+)
research panel
Study 3 Study 3 Study 3
Shoppers in a Restaurant Menu Heart HC NA(+), FA(0), Credibility(0), PI(+), Heart(+), Stroke(+)
local mall items NI NA(+), FA(0), Credibility(0), PI(0), Heart(+), Stroke(+)
Nutritional context NA(+), FA(+),Credibility(0), PI(+), Heart(+), Stroke(+)
Note: (+): significant effect, (0): no effect
a. Nutrition Attitude; b. Purchase Intention; c. Food Attitude; d. Healthy Claim; e. Nutritional Information

Continued

Table 2.2: Overview of Previous Studies

39
 Table 2.2 continued

Studies Sample Target Food Key Variables Findings

Cranage, et University Food items in on NI Satisfaction with Food Quality (+), Intention to
al. (2004) Students campus café Repurchase (+), Healthy Food Choice (+)
Burton & Consumer Restaurant Menu NI NA (+), FA (0), PI (0)
Creyer household items HC (Cancer & Heart) NA (+), FA (0), PI (0)
(2004) research panel Nutritional Frame (NF) NA (+), FA (+), PI (+)
NI* NF NA (0), FA (+), PI (+)
Burton, et Residents in a Restaurant Menu NI (None, Calorie, All) FA (+), PI (0), Food Choice (+), Heart (+)
al. (2006) single south- items
central state Menu Item Type FA (+), PI (+), Food Choice (+), Heart (+)
Dorms Students and Restaurant Menu Study 1
(2006) participants in a items NI Stage of behavior change (0), Food Choice (0)
local community Study 2
NI*Healthy knowledge Dietary restraint (+), Food Choice (+)
40

Kozup, et al. University Snack crackers Level of Trans fat Disease Risk (+), Perceived importance of trans fat (+)
(2006) students Trans fat knowledge Disease Risk (+), Perceived importance of trans fat (0)
Level*Knowledge Disease Risk (+), Perceived importance of trans fat (+)
Kemp, et al. Household Frozen chicken Level of Nutrition NA (+), PI (0), Heart (+), Stroke (+), Weight gain (0)
(2007) research mail dinner Type of HC NA (0), PI (0), Heart (+), Stroke (0), Weight Gain (0)
panel Motivation*Level NA (0), PI (+), Heart (0), Stroke (+), Weight Gain (0)
Hwang & Grocery shoppers Menu item Level of NI NA (+), FA (+), Willingness to pay (+)
Lorenzen
(2008)

40
 CHAPTER 3: METHODOLOGY

In the previous chapter, the conceptual model and five hypotheses were developed

based on prior research studies. Their purpose was to examine the impact of nutritional

information disclosure and the role of influential factors in processing nutritional

information. To test the model and hypotheses an experimental design method was

employed. An experimental method is one way to determine the causal relationship

between variables by controlling for other confounding variables. In order to identify the

effect of a certain variable, the treatment variable, it is very important to indentify the

expected sources of variability and also to design an experiment which reduces these

sources of variability (Dean & Voss, 1999). The randomized experimental design method,

in particular, is used to reduce some possible biases such as placebo effects. Thus, this

study employed the randomized design method in order to obtain rigorous internal

validity. This chapter describes the development of the experimental design of this study,

the instruments associated with key variables, and the survey procedure used in this study.

3.1 Experimental Design

To verify the proposed hypotheses, a between-subject experimental design was

applied with three levels of nutritional information (absence of nutritional information,

total calories only, and nutritional information of five important nutrients in addition to

41
total calories), two nutritional levels of a target menu item (healthy and unhealthy), and

two nutritional levels of menu context (healthier than the target item and unhealthier than

the target item).

To test the effect of nutritional information on the menu and find the extent to

which the level of nutritional information provided has an impact on consumer food

evaluation, three different levels of nutritional information were provided: no nutritional

information, which is considered a control group (CG), total calories only (TC), and

nutritional information including six important nutrient contents (NI). According to

Cranage et al. (2004), consumers use only a fraction of the nutritional information

provided. In addition, the nutrition labeling regulations passed by several cities and

states demonstrate the variety of nutritional information that can be included on menus.

New York City, for instance, passed a law requiring that fast food outlets include only

total calorie counts for menu items, whereas California requires total calories, saturated

fat, carbohydrates, and sodium information to be included on menus. Thus, it is

necessary to examine how different levels of nutritional information impact consumer

choice rather than simply testing for the effects of including any information at all. For

this study, the specific nutrition content for the NI Group was based on the definition of

healthy food, passed regulations, and consumer evaluation of important nutrients through

the pretest.

Two levels of nutritional value of the target menu item, healthy or unhealthy, were

manipulated to examine the effect of nutritional information disclosure depending on the

healthiness of a menu item. In previous studies, selected menu items investigated were

mostly unhealthy items with high calories, fats, and sodium. However, nutritional

42
information may have a different effect depending on the healthiness of a menu item.

Thus, this study tested the effects of provision of nutritional information for items of two

different levels of nutrition, healthy and unhealthy items.

The nutritional level of a menu context was also manipulated by altering the

nutritional values of alternatives in a menu context compared to the nutritional level of

the target menu item. For this manipulation, two different nutritional levels of the target

menu item and two different nutritional levels of a menu context were used. A total of 12

different scenarios from a 3 (nutritional information level) × 2 (nutritional level of a

target item) × 2 (nutritional level of menu context) design were developed. The specific

experimental design is outlined in Table 3.1.

No Nutritional Total Calorie only Nutritional

Information Group Information Group Information Group
(Control Group) (TC Group) (NI Group)
Menu Context Menu Context Menu Context
Target Item Healthy Unhealthy Healthy Unhealthy Healthy Unhealthy
Healthy 1 3 5 7 9 11
Unhealthy 2 4 6 8 10 12

Table 3.1: Experimental Design

3.2 Measurement of Variables

3.2.1 Dependent Variables

To determine consumer food decision processes, three dependent variables were

measured in this study: consumers perceived nutrition, overall attitude toward a menu

43
item, and purchase intention. The measures of these three variables were adopted from

previous studies, but were modified for this study. All responses were measured using a

five-point Likert-scale.

Nutritional Perception

Consumers’ nutritional perception was measured using three items which have

been used in several previous studies (Burton & Creyer, 2004; Garretson & Burton, 2000;

Kozup, et al., 2003). The three items were: “Based on the information provided on the

menu, how important do you think a menu item would be as part of a healthy diet?” rated

by a five-point Likert scale defined as 1= not important at all to 5=very important, “What

level of nutrition do you think is suggested by the information regarding the menu item

on the menu?” measured by a five-point Likert scale defined as 1=not nutritious at all to

5=very nutritious, and “ How healthy do you think the menu item on the menu would be?”

rated by a five-point Likert scale defined as 1= very unhealthy to 5=very healthy.

Overall Attitude toward a Menu Item

The questions for overall attitude towards a menu item were chosen based on

previous research (Burton & Andrews, 1996; Burton & Creyer, 2004; Garretson &

Burton, 2000; Kozup, et al., 2003). Three questions were asked concerning the

consumer’s overall attitude toward a given menu item. The specific questions were:

“Overall, how much are you favorable toward a menu item?” measured by a five-point

Likert scale defined as follows; 1=very unfavorable to 5=very favorable, “Overall, how

much do you prefer a menu item?” rated by a scale defined by 1=very dislike to 5=very

44
like, “What is your overall feeling toward a menu item?” measured by a five-point scale

defined by 1=very negative and 5=very positive.

Purchase Intention

Three items used in prior studies (Burton & Andrews, 1996; Burton & Creyer,

2004; Garretson & Burton, 2000; Kozup, et al., 2003) were utilized for assessing

purchase intention of a menu item. The three items included measurements of the

respondent’s interest in purchasing, willingness to purchase, and likelihood of purchasing

a menu item, and were rated by a five-point Likert scale defined as follows: 1=not

probable at all to 5=very probable for the first question, 1=not willingness at all to 5=very

willingness for the second question, and 1=very unlikely to 5= very likely for the third

question.

3.2.2 Independent Variables

Nutritional Information

In order to find the effect of nutritional information disclosure, three different

levels of nutritional information were included as independent variables. One of three

levels of nutritional information was provided on the menu: absence of nutritional

information, total calories only, and nutritional information of six nutrients. These three

categorical groups can be expressed using two dummy variables in data analysis. Thus,

this information was recorded as two indicator variables to distinguish the level of

nutritional information. For example, one dummy variable was assigned a value of 1 if

total calories count was the only information provided; otherwise it was coded as 0. A

45
second dummy variable took on a value of 1 if nutritional information for the six nutrient

was offered on the menu; otherwise it took on a value of 0. The control group, where no

nutritional information was offered on the menu was indentified by assigning values of 0

to both dummy variables.

Nutritional Menu Context

According to the experimental design, two different menu contexts were tested: a

healthy context and an unhealthy context, depending upon the nutritional values of

alternatives offered on the menu. To test the main effect of menu context, nutritional

menu context was included as an independent variable, but the effects of menu context

were controlled for when other moderating variables such as motivation to process and

nutritional knowledge were tested. Nutritional menu context was also recorded as an

indicator variable which took on a value of 1 if the provided nutritional menu context was

healthy and 0 if it was unhealthy.

3.2.3 Moderating Variables

Measures of the following moderating variables were used in this study. They

were adopted from existing scales used in prior studies. Minor modifications were made

such that the scales were appropriate for a restaurant context. Except for the questions

about objective nutritional knowledge, which were measured using a multiple choice

question scale, the moderating variables were evaluated using a five-point Likert scale.

46
Motivation to Process Nutritional Information

Motivation to process nutritional information referred to consumers’ willingness

to acquire and process nutritional information in the context of a restaurant (Moorman,

1993). Motivation to process nutritional information was measured using five items, all

of which had been used in previous studies (Chandon & Wansink, 2007b; Kemp, et al.,

2007; Moorman, 1990). The following five statements were presented, and respondents

were asked to rate their level of accord with each statement using a five-point Likert scale

defined as 1=strongly disagree to 5=strongly agree: “In general, I am interested in

looking for nutritional information of menu items in a restaurant”, “In general, I pay

attention to nutritional information while choosing a menu item in a restaurant”, “I would

like to receive additional nutritional information about menu items in a restaurant”, “ It is

important to me that nutritional information is available on the menu”, and “I usually

ignore nutritional information while choosing a menu item in a restaurant”.

Nutritional Knowledge

Nutritional knowledge refers to the ability to use nutritional and health

information accurately (Moorman, 1993). To measure nutritional knowledge, both

objective and subjective knowledge related questions were asked. Subjective nutritional

knowledge was assessed by how participants rated their present level of nutritional

knowledge. Three items developed by Moorman, et al. (2004) were used, and responses

were measured using a five-point Likert scale defined as 1=strongly disagree to

5=strongly agree for the following statements: “Compared to most people, I am quite

knowledgeable about nutritional information”, “Compared to most people, I am more

47
confident in using nutritional information to make a choice”, and “I feel confident about

my ability to comprehend nutrition information on the menu”.

In measuring objective nutritional knowledge, a total of fifteen multiple choice

questions were selected from a previous study (Andrews, et al., 1998), and from the ten-

question nutrition quiz offered by the North Carolina Department of Agriculture and

Consumer Services (NCDA&CS). Fifteen questions were selected from the two sources.

Each is associated with important nutrients identified by previous studies and labeling

laws; the difficulty level of each question was tested in the two pilot tests.

Table 3.2 displays the summary of study instruments for measuring the suggested

variables.

Variables No. of Items Sources

Burton & Creyer (2004); Garretson &
Perceived Nutrition 3
Burton (2000); Kozup, et al. (2003);
Burton & Andrew (1996); Burton &
Overall Attitude toward a
3 Creyer (2004); Garretson & Burton
menu item
(2000); Kozup, et al. (2003)
Burton & Creyer (2004); Garretson &
Purchase Intention 3 Burton (2000); Burton & Andrew
(1996); Kozup, et al (2003)
Motivation to Process Chandon & Wansink (2007b); Kemp,
5
Nutritional Information et al. (2007); Moorman (1990)
Subjective Nutritional
3 Moorman, et al. (2004)
Knowledge
Andrew, et al. (1998);
Objective Nutritional
15 North Carolina Department of
Knowledge
Agriculture and Consumer Services

Table 3.2: Study Instruments

48
3.2.4 Other Survey Questions

Besides the three main dependent variables, respondents were also asked to report

which menu item they would choose to order. For each menu item, the expected taste

and the credibility of provided nutritional information were assessed using one- and two-

item Likert scales, respectively; the items were drawn from previous studies (Butron &

Creyer, 2004; Kozup, et al., 2003). At the end of the survey, general questions about

restaurant visits, sources of nutritional information, and health status were added.

Demographic questions which asked respondents about age, gender, ethnicity,

employment status, monthly income and monthly expenditure for food away from home

were also included.

3.3 Pilot Study I

The goal of the first pilot test was to identify popular menu items and important

nutrients that consumers consider when choosing a menu item in restaurants. Consumers’

perceived healthiness of menu items was also evaluated when the actual nutritional

values were provided. Based on these results, nutritional values of menu items would be

manipulated for the main survey.

The first pilot survey included undergraduate students at a large mid-western

university as respondents. Students were asked to fill out the pilot study using a paper

and pencil method. A total of 158 students voluntarily participated in the first pilot test,

which took 15-20 minutes to complete.

49
3.3.1 Procedure

In the first pilot study, eight different menu items available at casual restaurants

were provided on the menu, each with a description: Chicken Caesar Salad, Classic

Sirloin, Lemon-Herb Roasted Chicken, Chicken Fajitas, Chicken Marsala, Grilled

Salmon, Grilled Pork Chops, and California Turkey Club. After reading eight menu

items with descriptions, participants were asked to list their three most preferred items

from the list and also to select their top choice.

In the second section, participants answered sixteen questions meant to measure

objective nutritional knowledge. They evaluated the importance of eleven nutrient

contents that consumers usually use in making food choices in restaurants using a five-

point Likert scale defined as 1=Not at all important to 5= Very important. Participants

were also asked to rank the eleven nutrient contents in order of importance.

Finally, actual nutritional information of eight menu items was provided, along

with a description of each item. The perceived nutritional level of each menu item was

evaluated to check the level of provided nutritional values. Additionally, demographic

information was collected. The questionnaire used for the first pilot study is included in

Appendix A.

3.3.2 Data Analysis and Results

The collected data were analyzed using SPSS v.17 using various methods of

analysis. Sixty-four percent of the participants were female with the majority of them

between the ages of 19 and 23 (81%). Most were Caucasian (77%). Sixty-three percent

of participants reported their monthly income to be less than $1,000, and half of them

50
(51%) spent between $100 and $200 on monthly food expenditures.

Selection of Important Nutrition Content

The first objective of this survey was to identify six important nutrients that

should be presented on the menu for the nutritional information (NI) group. According to

the average importance scores for the eleven nutrient contents, total calories (mean=3.54

out of 5) was the most important contributor when choosing a menu item in a restaurant,

then total fat (mean=3.44), protein (mean=3.29), and trans fat (mean=3.28). Similar

results were found in the ranking analysis. Forty-three percent of respondents ranked

total calories as most important, while total fat was placed first or second by thirty-four

percent of respondents. Also, calories from fat, trans fat, protein, and saturated fat were

highly ranked. The top six nutrient contents were therefore: total calories, total fat, trans

fat, calories from fat, saturated fat, and protein. However, several of the six nutrient

contents such as calories from fat and protein are not included in the definition of healthy

food by the Food and Drug Administration (FDA, 1999) and are not included in any

enacted menu labeling laws. According to the definition of healthy food, the amount of

total fat, saturated fat, cholesterol, and sodium are important when deciding if food is

healthy or not. California requires a listing of total calories, saturated fat, carbohydrates,

and sodium. Philadelphia added trans fat to those requirements. In short, total calories,

total fat, saturated fat, and sodium are critical nutrient contents that consumers need to

consider when selecting a menu item. Similarly, cholesterol is also included in the

definition of healthy food. Thus, for this study, the following six nutrient contents were

selected and were placed on the menu for the (NI) group: total calories, total fat,

51
saturated fat, cholesterol, protein, and sodium.

Decision of Nutritional Values of Menu Items

In order to achieve the second objective of pilot testing, the reported perceived

nutritional values for the eight menu items were analyzed. The goal of this analysis was

to identify the levels of healthiness of menu items so that the appropriate nutritional

values used in the main experiment were manipulated. Analysis of variance (ANOVA)

revealed that the perceived nutritional values of the eight menu items were significantly

different from one another, and the results of the Post Hoc test indicated that the eight

items could be categorized into three groups depending upon their perceived nutritional

values (F=97.5, d.f.=(7, 1250), p≤0.001). Among the eight items, the Grilled Salmon

(mean=4.10 out of 5) and the Lemon Herb Roasted Chicken (mean=4.03) were perceived

as healthier items, which included 15% to 20% of the daily recommended amounts of fat,

saturated fat, and sodium based on a 2,000 calories diet. On the other hand, the Chicken

Caesar Salad (mean=2.46), the Chicken Marsala (mean=2.62), and the Chicken Fajitas

(mean=2.64) were considered relatively unhealthy items. Unhealthy menu items

contained 70% to 80% of the recommended daily amounts of fat, saturated fat, and

sodium. These results agreed with the manipulated levels used in previous studies

(Burton, et al., 2006; Hwang & Lorenzen, 2008; Kozup, et al., 2003) where the

nutritional values of healthy food were set at 15% to 20% of the daily recommend values,

and the nutritional values of unhealthy foods were set at 50% to 60% of the daily

recommended values. Thus, the nutritional values for the healthy target item were set

such that the food accounted for 20% of the recommended daily values of fat, saturated

52
fat, cholesterol, and sodium. For items in the healthy menu context, the nutritional values

of the two alternatives items were manipulated such that they accounted for 12% and 10%

of recommended daily values, respectively. The unhealthy target item was made to

account for around 60% of recommended daily values, and the unhealthy menu context

included two alternative options containing 65% and 70% of the recommended daily

values, respectively. The amount of protein for each item was set such that it was

inversely related to the amounts of fat, cholesterol and sodium contained in the item. For

instance, healthy items include 60% to 70% of daily recommended values of protein.

Total calories information for the six menu items was set based on the actual total calorie

counts of each menu item, but some modifications were made depending on the

manipulated level of daily values for other provided nutrients. The manipulated

nutritional values of menu items were similar to their actual values so that it was possible

to control for the effects of actual healthiness of menu items. Table 3.3 shows the

perceived nutrition level of menu items.

53
 Nutritional Perception a Preference b Choice
Item
(Mean, S.D.) (Frequency, %) (Frequency, %)
Chicken Caesar Salad 2.46 (0.80) 77 (16.3%) 17 (10.8%)
Grilled Salmon 4.10 (0.72) 72 (15.3%) 35 (22.2%)
Lemon Herb Roasted
4.03 (0.65) 42 (8.9%) 8 (5.1%)
Chicken
Chicken Marsala 2.62 (0.84) 42 (8.9%) 10 (6.3%)
Chicken Fajitas 2.64 (0.71) 76 (16.1%) 28 (17.7%)
Classic Sirloin Steak 3.05 (0.81) 71 (15.1%) 34 (21.5%)
Grilled Pork Chop 3.27 (1.06) 29 (6.2%) 5 (3.2%)
Italian Turkey
3.30 (0.70) 62 (13.2%) 19 (12.0%)
Sandwich
Note: a. The mean value of three items when the actual nutritional values were provided,
which were measured using the a five-point Likert scale, and 5 is the highest
value representing very healthy
b. Multiple Responses (Three out of Eight items)

Table 3.3: Evaluation of Menu Items

Check of Objective Nutritional knowledge

The third objective of the pilot study was to construct the objective nutritional

knowledge questions. The sixteen selected nutritional knowledge questions were

analyzed to confirm the measures. The mean score for the nutritional knowledge

questions was 9.8 out of 16 points with standard deviation 2.26 (the median score was

10). There were three difficult questions for which fewer than 22% of respondents

provided correct answers and five relatively easy questions with more than 80% of

answers correct. The rates of correct answers for the other 8 questions were between 50%

and 80%. To better discriminate between levels of objective nutritional knowledge, two

questions related to fat were deleted because for both questions, 53% of answers were

54
correct and 27% of answers were “don’t know”. Instead, one difficult question related to

fat was added that 27% of respondents had correctly answered in a previous study

(Andrews, et al., 1998). All other questions were kept in the second pilot test.

Selection of Menu Items

Regarding menu items, Chicken Caesar Salad (16.3%), Chicken Fajitas (16.1%),

Grilled Salmon (15.3%), and Classic Sirloin Steak (15.1%) tended to be included in

respondents top-three items; the top choice among the three preferred items was the

Gilled Salmon (22.2%). But, the Classic Sirloin Steak (21.5%) and the Chicken Fajitas

(17.7%) were also chosen with similar frequencies. Because of this, Grilled Salmon and

Chicken Fajitas were selected as healthy and unhealthy target items (See Table 3.3).

However, to reduce the effects of food preference, the base ingredient of all selected

menu items was changed to chicken. According to Farm Animal Statistics estimated by

The Human Society of the U.S. (HSUS), the most commonly consumed meat in 2004

was chicken (85 pounds per person), which was followed by beef (65 pounds per person)

and pork (51 pounds per person). The amount of chicken consumed in the U.S. has been

continuously increasing, reflecting that Americans enjoy eating more chicken and that it

is becoming a popular food source. Because of this, the Grilled Salmon and the

California Turkey Club Sandwich were replaced with the Grilled Chicken and the Italian

Chicken Sandwich. The Classic Sirloin Steak, one of the popular items, was excluded

from the menu because it was not easy to find a substitutable chicken item. To create a

menu context including two alternative chicken items, the Lemon Herb Roasted Chicken

and the Italian Chicken Sandwich were selected to create the healthy menu context, while

55
the Chicken Caesar Salad and the Chicken Marsala were provided for an unhealthy menu

context. Thus, the following six menu items were selected to be used for the main

experiment: Lemon Herb Roasted Chicken, Grilled Chicken, Italian Chicken Sandwich,

Chicken Fajitas, Chicken Marsala, and Chicken Caesar Salad.

To check the perceived healthiness of the six selected menu items and the

manipulation of nutritional values of each, a second pilot test was conducted. In addition,

the reliability and construct validity of all measures were tested.

3.4. Pilot Study II

The aim of the second pilot study was to check the consumer perceived

healthiness of the selected menu items without nutritional information, as well as the

manipulation of nutritional values of menu items. The reliability of all measures was also

tested. A total of 94 undergraduate students volunteered to take part in the second pilot

study; each received extra course credit as an inducement to participate.

3.4.1 Procedure

The questionnaire asked respondents to evaluate six menu items, and to answer

objective nutritional knowledge questions, individual characteristic questions, and

demographic information questions. The six menu items were presented in a random

order, and the order in which the menu items were presented was also rotated to reduce

the order effect. To test the effect of the level of nutritional information, three different

levels of nutritional information were provided with the description of each menu item:

absence of nutritional information, total calories only, and nutritional information of six

56
nutrient contents. Participants were randomly assigned to receive one of the three levels

of nutritional information.

Participants responded to questions that measured how they perceived the

nutrition of each item, their overall attitude toward a menu item, and their purchase

intention for each of six menu items; they were also asked to choose one of six menu

items.

In the second section, participants were asked to answer fifteen nutritional

knowledge questions, while the third section asked about consumers’ individual

characteristics such as subjective nutritional knowledge, motivation to process nutritional

information, and future behavior intention related to the use of nutritional information.

Lastly, general questions about restaurant visits and demographic information such as

gender, age, income and ethnicity were asked. The questionnaire is attached in Appendix

B.

3.4.2 Data Analysis and Results

Sixty-three percent of respondents were female; eighty-eight percent were

Caucasian. The majority of respondents (92%) were between the ages of 18 and 23.

Most of the respondents (77%) reported that their monthly income was less than $1,000;

fifty percent of respondents estimated that their monthly expenditures for food away from

home were between $100 and $200. Sixty-two percent of respondents visited restaurants

once or twice during a week, and the major nutritional information sources used were

restaurant websites (37%) or menus (29%).

The perceived nutritional level of the six menu items without nutritional

57
information was checked to manipulate the control group. The results of analysis of

variance (ANOVA) showed that there were significant differences in nutritional

perception among six menu items without nutritional information (F=20.8, d.f.=(5, 234),

p≤0.001). In particular, the Post Hoc test revealed that consumers perceived that the

nutritional levels of the Chicken Fajitas (mean=2.67 out of 5), the Chicken Marsala

(mean=3.04), and the Chicken Caesar Salad (mean=3.10) significantly differed from the

nutritional levels of the Italian Chicken Sandwich (mean= 3.40), the Lemon Herb

Roasted Chicken (mean=3.69), and the Grilled Chicken (mean=4.24). As would be

expected, the first three menu items, the Chicken Fajitas, the Chicken Marsala, and the

Chicken Caesar Salad, were selected as unhealthier menu items, while the Italian Chicken

Sandwich, the Lemon Herb Roasted Chicken, and the Grilled Chicken were chosen as

healthier items. Table 3.4 displays the mean values of perceived nutritional levels for the

six menu items when nutritional information was not presented.

Healthy Item Mean (St.dev.) Unhealthy Item Mean (St.dev.)

Italian Chicken
3.40 (0.72) b Chicken Caesar Salad 3.10 (0.91) c
Sandwich
Lemon Herb Roasted
3.69 (0.68) b Chicken Fajitas 2.67 (0.73) c
Chicken
Grilled Chicken 4.24 (0.75) a Chicken Marsala 3.04 (0.77) c
F=20.8 (d.f.=5, 234), p≤0.001
Note: a, b, and c are significantly different at the level of 0.05.

Table 3.4: Mean Values of Nutritional Perception of Menu Items

Similar to the results of the first pilot test, the Chicken Fajitas (23%) and the
58
Grilled Chicken (22%) were the most popular menu items. The Chicken Fajitas was

selected as the unhealthy target item, and the Chicken Marsala and the Chicken Caesar

Salad were chosen as the unhealthy alternatives to manipulate the unhealthy menu

context. For the healthy target item, the Grilled Chicken was used since it was the most

popular item among three healthier items; however, the nutritional perception of the

Grilled Chicken became more negative as nutritional information was added (mean=4.24

without nutritional information; mean=3.97 with only total calories; mean=3.78 with

nutritional information). It is possible that while consumers perceived the Grilled

Chicken to be very healthy, once they had been provided with the healthy nutritional

information of the Grilled Chicken, they realized it was not as healthy as expected. As a

result, the Lemon Herb Roasted Chicken, which was chosen by eighteen percent of

respondents, was selected as the healthy target item since the Lemon Herb Roasted

Chicken did not have a large fluctuation in the perceived healthiness. The Grilled

Chicken and the Italian Chicken Sandwich were provided on the menu as healthier

options for the healthy menu context. The fact that the Grilled Chicken appeared less

healthy when nutritional information was provided also influenced the manipulated

nutritional values for both the Grilled Chicken and the Lemon Herb Roasted Chicken. In

order to maintain similar levels (12%) of nutritional values for the Lemon Herb Roasted

Chicken as those presented in the second pilot test, the nutritional values of the healthy

target item were slightly adjusted such that they accounted for 15% of daily

recommended values. This adjustment was necessary because the nutritional values of

the healthy target item had to be higher than the levels of the other healthy alternatives.

On the other hand, the nutritional values of the Grilled Chicken were changed to account

59
for 8% of daily recommended values because it was perceived as the healthiest item

among six menu items and the actual nutritional values (20%) made this item appear too

unhealthy. The nutritional levels of the Italian Chicken Sandwich were maintained at

their prior level in the second pilot test (10% of recommended values). The manipulated

nutritional values of the six menu items were very close or even healthier than the actual

nutritional values, and they are shown on the Table 3.5.

The second pilot study also provided an opportunity to check the modified

objective nutritional knowledge questions. The mean score was 8.7 out of 15 questions

(median=9). The results of the second pilot test revealed that there were four difficult

questions and five easy questions. The reliabilities of all measures were tested and

Cronbach’s alpha of all instruments met the acceptable level; all Cronbach’s alpha values

were larger than 0.85. Thus, without any modifications, all measures were used in the

main survey.

60
 Healthy Menu Item Unhealthy Menu Item
Target Item Alternative 1 Alternative 2 Target Item Alternative 1 Alternative 2
Nutrition Lemon Herb Italian Chicken Chicken Caesar Chicken
Grilled Chicken Chicken Fajitas
Content Roasted Chicken Sandwich Salad Marsala
Total calories 420 380 340 830 900 920
9g 6.5g 5.2g 38g 43g 46g
Total Fat
(14% DV) (10% DV) (8% DV) (58% DV) (66% DV) (71% DV)
3g 2g 1.4g 12g 13g 15g
Saturated Fat
(15% DV) (10% DV) (7% DV) (60% DV) (65% DV) (75% DV)
45mg 36mg 24mg 186mg 201mg 216mg
Cholesterol (15% DV) (12% DV) (8% DV) (62% DV) (67% DV) (72% DV)
384mg 264mg 216mg 1464mg 1560mg 1680mg
61

Sodium
(21% DV) (11% DV) (9% DV) (61% DV) (65% DV) (70% DV)
31g 34g 36g 15g 13g 10g
Protein
(62% DV) (68% DV) (72% DV) (30% DV) (26% DV) (20% DV)
Manipulated
15% DV 10% DV 8% DV 60% DV 65% DV 70% DV
Level
Note: The percent daily values are based on a 2,000 calorie diet.

Table 3.5: Manipulated Nutritional Values of Menu Items

61
3.5 Primary Study

3.5.1 Sample

The sample selected for this study consisted of undergraduate and graduate

students at a large mid-western university. The study was announced to undergraduate

students who were enrolled in marketing or hospitality management courses; any student

enrolled in these courses could sign up to participate. In order to recruit graduate

students, an email was sent to all graduate students in the home department. Potential

participants were offered incentives to participate in the study. Namely, undergraduate

students were offered extra points in selected courses, while graduate students were

offered $5 gift cards.

3.5.2 Sample Size

To assure reasonable power for the various analyses, the sample size was

calculated using the formula suggested by Dean and Voss (1999). Using the Tukey

multiple pair-wise comparisons of main factors, the sample size was calculated based on

a set of 95% simultaneous confidence intervals for identifying significant pair-wise

differences between the levels of nutritional information. The formula for the confidence

interval is:

⎛ 2 ⎞
Mean difference between two levels ± qa ,v ( x −1),0.05 / 2 × msE ⎜ ⎟
144444244444 ⎝ bcx3⎠
msd

Where a= level of nutritional information = 3

b= level of healthiness of a target item = 2

c = level of healthiness of a menu context =2

62
 v = number of condition = 3 × 2 × 2 = 12

x = sample size in a condition

A reasonable estimate of the mean square error (MSE) was obtained from the

data collected in the second pilot study; estimates ranged from 0.47 to 0.88 depending on

the menu item at hand. The largest value, 0.88, was used as the estimate of the MSE in

the sample size calculation. Assuming an equal sample size for each condition, the

acceptable minimum significant difference (msd) was set at 0.2, a value which was based

on the result of the second pilot test. Thus, the following formula was developed:

⎛ 2 ⎞
q3,12( x −1),0.05 / 2 × (0.88)2 ⎜ ⎟ ≤ 0.2 Æ 2
q3,12( x −1),0.05 ≤ 1.037 x
⎝ 4x ⎠

The value of qa,v(x-1),0.05 was chosen from the Tukey’s method table (Dean & Voss,

1999); the appropriate sample size per condition (x) was identified by setting both sides

of the equation equal to one another.

When x=10, q23,108,0.05 = 3.682 = 13.54 > 10.03 = 1.037×10 Æ increase x

When x=13, q23,144,0.05 = 3.632 = 13.17 ≅ 13.39 = 1.037×13

Thus, each condition required at least 13 participants, which made the total

required number of participants at least 156 (=13×12).

In addition, the necessary sample size for this study was estimated using the

sample size calculator software, PASS 2008. The software’s method was based on a

multiple regression model that was employed for testing proposed hypotheses. By

providing the desired power, significance level and number of independent variables, the

software is then able to calculate the necessary sample size. It was expected that a power

level of 0.8 was sufficiently strong to detect the size of the effects of the independent

63
variables (Frazier, Tix, & Barron, 2004), and the sample size was generated to assure

estimates significant at the 0.05 significance level. Additionally, the model with the most

possible independent variables was specified. This model would contain five

independent variables, including the two indicator variables indicating the three levels of

nutritional information, the nutritional knowledge variable, and two interaction terms

between nutritional knowledge and nutritional information. It would also contain the

information of control variables: the two dummy variables meant to control for the effects

of menu context and the healthiness of the target item, were also included. This

information with expected overall R2 values was entered into the software. The software

produced a sample size estimate of 250, which would achieve 80% power to detect 0.05

R2 using five independent variables with two control variables.

Thus, the sample size objective was 250 participants, with at least 13 participants

per condition.

3.5.3 Study Procedure

To collect the data, a self-administered survey was conducted in a computer lab

using the Media Lab computer program. The data were collected over the course of three

weeks, from April 13 to April 30. Participants were randomly assigned to one of the

twelve scenarios and allowed 30 minutes to complete the survey.

An introduction page that explained the study and the procedure was presented,

and then participants were shown a menu that included three different items with or

without nutritional information. The menu used in the experiment is attached in

Appendix C. Questions were presented one by one on the screen, and participants were

64
able to see the designed menu while they answered all questions so that they responded to

each question by comparing menu items without being required to guess or remember the

provided information.

Participants were asked to answer questions about the expected taste of a menu

item, nutritional perception, overall attitude toward a menu item, purchase intention, and

source credibility. Nutritional knowledge questions, the measure of motivation to process

nutritional information questions, health related questions, and demographic questions

followed in sequence.

3.5.4 Data Analysis

The data were analyzed using SPSS version 17.0 Windows. For testing the

reliability and internal validity of measures, reliability analysis using Cronbach’s alpha

and exploratory factor analysis (EFA) were utilized. Additionally, analysis of variance

(ANOVA) was executed for checking the manipulation of the nutritional values of the

menu items. For the hypothesis tests, multiple regression analysis was conducted.

Furthermore, to test the moderating effects of nutritional menu context, motivation to

process, and nutritional knowledge, a hierarchical multiple regression method was used.

65
 CHAPTER 4: RESULTS

4.1 Sample

A total of 319 observations were collected and included in the analysis.

Approximately 27 participants were assigned to each of the twelve conditions (See Table

4.1).

No Nutritional Total Calorie only Nutritional

Information Group Information Group Information Group
(Control Group) (TC Group) (NI Group)
Menu Context Menu Context Menu Context
Target Item Healthy Unhealthy Healthy Unhealthy Healthy Unhealthy
Healthy 27 27 26 27 26 27
Unhealthy 27 26 27 26 26 27

Table 4.1: Sample Size

The sample contained equal numbers of male and female participants, and the

majority of respondents (70%) were Caucasian. Ninety-two percent of respondents were

undergraduate students and were an average age of 22 years. The average monthly

income was $743.02 and ranged between $0 and $5,400; twenty-eight percent of

respondents reported their monthly income as $0, a fact which might be due to the

66
student sample. Respondents spent an average $212.72 per month on food away from

home. Ninety-seven percent of respondents were non-vegetarian, and many reported that

their favorite food was beef (34%) or chicken (29%), as was expected. On average,

respondents ate at restaurants three times during a week; the number of weekly restaurant

visits ranged between 0 and 12. Further, participants reported eating at fast food

restaurants on average twice per week. The major sources from which respondents

obtained nutritional information included menus (25%), restaurant websites (24%), or

other websites (22%) like Yahoo or Google. Other information sources included word of

mouth from family or friends and publications, such as books or magazines. Most

respondents (97%) reported that their current health status was good (mean=3.9 out of 5),

and that they were mainly concerned with diets for health reasons (mean=4.24 out of 5)

rather than for the weight loss reasons (mean=3.66). Using the reported respondents’

weight in pounds and height in inches, Body Mass Index (BMI) was calculated using the

following formula: [weight ÷ (height)2] × 703 (CDC, 2009). Most of the respondents

(70%) were normal weight with BMIs between 18.5 and 24.9, while twenty-four percent

of respondents having BMIs greater than 25 were overweight or obese. Table 4.2

displays the specific demographic information of respondents in this study.

67
 Frequency Percent (%)
Gender
Male 160 50.2%
Female 159 49.8%
Race
Caucasian 224 70.2%
African American 13 4.1%
Hispanic 2 0.6%
Asian 78 24.5%
Native American 1 0.3%
Others 1 0.3%
Status
Undergraduate 293 91.8%
Graduate 23 7.5%
Staff 2 0.6%
Vegetarian
Yes 9 2.8%
No 310 97.2%
Type of Favorite Food
Beef 109 34.2%
Pork 6 1.9%
Chicken 93 29.2%
Lamb 7 2.2%
Turkey 14 4.4%
Seafood 64 20.1%
Others 26 8.2%
Nutritional Information Source a
TV 93 11.7%
Restaurant Website 191 24.1%
Other Website 173 21.8%
Newspaper 35 4.4%
Menu 197 24.8%
Ask to the server 58 7.3%
Others 46 5.8%
Total 793 100%

Continued

Table 4.2: Demographics of the Sample

68
Table 4.2 continued

Mean (S.D.) Median

Age (years) 22.34 ( 4.33) 21
Monthly Income ($) 743.02 (872.16) 500.00
Monthly Expenditure on FAFH ($) 212.72 (163.85) 200.00
Restaurant Visit (#) 2.99 (2.26) 2
Fast-food Restaurant Visit (#) 1.94 (1.85) 1
b
Current Health Status 3.97(0.77) 4
Health Status in past 5 yrs b 4.56 (0.81) 5
c
Diet for Health 4.24 (0.85) 4
c
Diet for Weight Loss 3.66 (1.26) 4
Body Mass Index 23.20 (3.69) 22.77
Note: a. Multiple Responses (at most three)
b. Measured by 5 point Likert Scale ranged from 1=very poor to 5=very good
c. Measured by 5 point Liker Scale ranged from 1=not interested at all to 5=very
interested

4.2 Reliability and Internal Validity of Measures

Exploratory factor analysis (EFA) was conducted to check the internal validity of

continuous variables including the three dependent variables: motivation to process,

subjective nutritional knowledge, and source credibility. A total of 19 items for 6

variables were included in the analysis using a Principle Component method with a

Varimax rotation. According to the result of the EFA, five components were identified

which had eigenvalues greater than 1.0 and which explained 82% of total variance. The

result revealed that measures of overall food attitude and purchase intention were

identified as one dimension, but that the other variables were clearly identified. Table 4.3

summarizes the factor loadings with total explained variances for each factor.

69
 Scale Items Factor Loading
Overall Attitude toward a Item Q1 0.88
Overall Attitude toward a Item Q2 0.91
Overall Attitude toward a Item Q3 0.90
Purchase Intention Q1 0.95
Purchase Intention Q2 0.94
Purchase Intention Q3 0.93
Motivation to Process Q1 0.88
Motivation to Process Q2 0.87
Motivation to Process Q3 0.85
Motivation to Process Q4 0.84
Motivation to Process Q5 (R) a 0.80
Nutritional Perception Q1 0.86
Nutritional Perception Q2 0.87
Nutritional Perception Q3 0.89
Subjective Nutritional Knowledge Q1 0.87
Subjective Nutritional Knowledge Q2 0.76
Subjective Nutritional Knowledge Q3 0.82
Source Credibility Q1 0.90
Source Credibility Q2 0.92
Eigenvalue 5.19 3.99 2.46 2.22 1.72
Percent of Total Variance 27.30 20.97 12.94 11.69 9.08
Cumulative Percent of Total Variance 27.30 48.27 61.22 72.92 81.99
Note: a. Reverse Code

Table 4.3: Results of Exploratory Factor Analysis

For testing the internal consistency of items, the reliability analysis was conducted

utilizing Cronbach’s alpha. For all continuous variables, the reliability analysis was

implemented using the total sample. The values of Cronbach’s alpha for all measures

ranged from 0.82 to 0.96, which indicated an acceptable level of reliability, α=0.70, for

70
all measures. Thus, it was concluded that the instruments utilized for this study had

internally consistent reliability in measuring the defined variables. Table 4.4 displays the

values of Cronbach’s alpha for each instrument used in this study.

Measures No. of Items Cronbach’s Alpha (α)

Nutritional Perception 3 0.88
Source Credibility 2 0.82
Overall Attitude toward an item 3 0.92
Purchase Intention 3 0.96
Motivation to Process Nutritional Information 5 0.92
Subjective Nutritional Knowledge 3 0.86

Table 4.4: Results of Reliability Analysis

4.3 Manipulation Check of Menu Items

Prior to hypothesis testing, the manipulation check was completed to ensure that

the six menu items were placed in their proper nutritional menu contexts – healthy or

unhealthy – when nutritional information was not provided. For this manipulation check,

analysis of variance (ANOVA) was conducted using the respondents’ assessed values of

nutritional perception without nutritional information. The results of ANOVA showed

that there were significant differences between the perceived nutritional levels of the six

menu items (F=26.89, df=(5, 357), p≤0.001). Specifically, according to the tukey post

hoc test, the perceived nutritional levels of the Chicken Fajitas (mean=2.64 out of 5),

Chicken Caesar Salad (mean=3.00), and Chicken Marsala (mean=3.06) significantly

71
differed from the perceived nutritional levels of the Italian Chicken Sandwich

(mean=3.43), Lemon Herb Roasted Chicken (mean=3.67), and Grilled Chicken

(mean=4.09). Among the six menu items, the Grilled Chicken was perceived as the

healthiest item, while the Chicken Fajitas was perceived as the least healthy item. The

average perceived nutrition levels of the menu items when nutritional information was

not provided are presented in Table 4.5.

Healthy Item Mean (St.dev.) Unhealthy Item Mean (St.dev.)

Italian Chicken Sandwich 3.43 (0.68) b Chicken Caesar Salad 3.00 (0.62) c
Lemon Herb Roasted
3.67 (0.73) b Chicken Fajitas 2.64 (0.84) c,d
Chicken
Grilled Chicken 4.09 (0.78) a Chicken Marsala 3.06 (0.76) d
F=26.9 (d.f.=5, 357), p≤0.001
Note: a, b, c, and d are significantly different at the level of 0.05.

Table 4.5: Mean Values of Nutritional Perception of the Menu Items without Nutritional
Information

Regarding the choice of a menu item, respondents most often selected the

Chicken Fajitas (n=61, 24.4%) or the Lemon Herb Roasted Chicken (n=54, 21.6%); this

result confirms that popular items were selected as the healthy and unhealthy target items.

Thus, the six menu items were correctly manipulated for this study since the healthy

target item was the Lemon Herb Roasted Chicken, and the unhealthy target item was the

Chicken Fajitas. For the healthy menu context, the Italian Chicken Sandwich and the

Grilled Chicken were placed on the menu; the Chicken Caesar Salad and the Chicken

Marsala were provided as unhealthy options in the unhealthy menu context.

72
 The accuracy of the provided nutritional information was assessed using two

questions. For the six menu items, the mean values of source credibility ranged from

3.33 (Chicken Caesar Salad) to 3.95 (Grilled Chicken) out of a possible 5 points. The

scale was defined as 1=very inaccurate to 5=very accurate. There were significant

differences in the source credibility scores, depending on the menu item at hand (F=10.17,

df=(5, 946), p<0.01). The tukey post hoc test revealed that the mean value of the source

credibility score for the Grilled Chicken significantly differed from the mean values for

the other five items. This implies that consumers were more likely to trust the nutritional

information that was provided for the Grilled Chicken than they were to trust the

information provided for the other items; this may be because the Grilled Chicken had the

healthiest nutritional values among the items. However, source credibility scores did not

significantly differ with the level of nutritional information for any of the six menu items.

4.4 Hypotheses Test

The proposed hypotheses were tested using multiple regression analysis.

Focusing on the target items, multiple regression analyses were conducted separately for

each of three dependent variables: nutritional perception, overall attitude toward the

menu item, and purchase intention for the target item. To begin, the assumptions

necessary for multiple regression analysis were checked. There are four required

assumptions in multiple regression analysis, and tests were conducted to check each

assumption: the normality test, the test of independence of error terms, the check of

consistent error variance, and the test for linearity of the regression function (Neter,

Kutner, Nachtsheim, & Wasserman, 1996). For all estimated regression models, there

73
were no violations of either the linearity assumption or the normality assumption; these

were checked using the residual plot and the normal probability plot, respectively. The

scatter plots of residuals versus the order of respondents showed that the error terms were

independent of the order in which the data were collected for all models. Plots of

residuals versus the fitted values did not show any particular patterns for any of the

estimated models, indicating that the error variances were consistent.

Several demographic variables were included in all regression models as control

variables: age, gender, and Body Mass Index (BMI), and income. However, no

demographic variables were significant in the model, they were therefore excluded.

4.4.1 The Influence of Providing Nutritional Information on the Menu

Multiple regression analysis was used to test the effect that nutritional information

disclosure had on consumers’ food decision making process as measured by three

dependent variables: nutritional perception, overall food attitude, and purchase intention.

Two dichotomous variables for three different nutritional information levels were

included in the model as independent variables, along with an additional independent

variable which distinguished a healthy target item from an unhealthy item; thus, the effect

of the nutritional level of target items could be controlled. The following regression

model was developed:

DVi = β0 + β1· TI + β2·TC + β3·NI + e

where DV=Dependent Variable with i=1, 2, 3;

DV= Nutritional Perception if i=1; DV= Overall Food Attitude if i=2;

DV = Purchase Intention if i=3

74
 TI = Target Item with 1 = healthy target item or 0 = unhealthy target item

TC = Total Calories only Information with 1 = provided or 0 = not provided

NI = Nutritional Information of 6 contents with 1 = provided or 0 = not provided

e = error term

The results showed significant evidence that the presence of nutritional

information on the menu negatively influenced consumers’ food decision making when

the health of the target item was controlled for. The model fit for each of the three

equations was deemed significant at the 0.05 level by F-tests (Nutritional Perception:

F=36.93, d.f.=(3, 315), p<0.001; Overall food attitude: F=3.83, d.f.=(3, 315), p<0.01;

Purchase Intention: F=3.41, d.f.=(3, 315), p<0.05). Additionally, the presence of

nutritional information when the health of the target item variable was included explained

25% of variance of nutritional perception, but accounted for only 3% and 2% of variance

of overall food attitude and purchase intention, respectively. Therefore, the level of

healthiness of menu items and nutritional information disclosure were more associated

with perceived nutrition of the menu item than they were with overall food attitude or

purchase intention.

Nutritional perception of the menu item decreased by 0.20 units when nutritional

information for the 6 nutrient contents was provided (Nutritional Perception: b3=-0.20,

S.E.=0.12). The presence of total calories information was also negatively associated

with customers’ nutritional perception, but the relationship was not significant. In the

models explaining overall attitude towards a menu item and purchase intention,

nutritional information levels, total calories only and nutritional information for the 6

nutrient contents, had significant negative effects (Overall food attitude: b2=-0.32

75
(S.E.=0.14), b3=-0.44 (S.E.=0.14); Purchase intention b2=-0.38 (S.E.=0.16), b3=-0.49

(S.E.=0.16)). This result implies that consumers negatively evaluated a menu item and

became less likely to purchase a menu item when nutritional information was provided

on the menu. This result supports hypothesis 1, which explained the influence of

nutritional information disclosure on consumer food decision making.

Additionally, a healthy target item was perceived as much healthier than an

unhealthy item (b1=0.99, S.E=0.09). However, there was no significant difference

between healthy and unhealthy target items in terms of either overall food evaluation or

purchase intention. Thus, overall attitude and purchase intention were not influenced by

the healthiness of the menu item. Table 4.6 outlines the results of the multiple regression

analyses.

Dependent Variables
Nutritional Purchase
Food Attitude
Perception Intention
Independent
B (S.E.) B (S.E.) B (S.E.)
Variables
Constant 2.66 (0.09)*** 3.95 (0.11)*** 3.89 (0.13)***
Target Item 0.99 (0.09)*** 0.12 (0.11) 0.05 (0.13)
Total Calories Info -0.06 (0.12) -0.32 (0.14)** -0.38 (0.16)**
Nutritional Info
-0.20 (0.12)* -0.44 (0.14)*** -0.49 (0.16)***
of 6 contents
Adj. R2 0.25 0.03 0.02
Model fit (F-Value) 36.93*** 3.83*** 3.41**
Note: *Significant at p<0.1; **Significant at p<0.05; ***Significant at p<0.01.
Standard errors are in the parentheses.

Table 4.6: Results of Regression Analysis: Effect of Nutritional Information Disclosure

76
 Based on the results of multiple regression analysis, the following predicted

models were obtained, which allowed the predicted values of the three dependent

variables for each menu item to be calculated. Table 4.7 and Figure 4.1 display the

predicted values of the three dependent variables.

Predicted Nutritional Perception = 2.66 + 0.99TI – 0.06TC – 0.20NI

Predicted Overall Food Attitude = 3.95 + 0.12TI – 0.32TC – 0.20 NI

Predicted Purchase Intention = 3.89 + 0.05TI – 0.38TC – 0.49NI

Healthy Target Item Unhealthy Target Item

Total NI of 6 Total NI of 6
Control Control
Calories content Calories content
Nutritional Perception 3.65 3.59 3.45 2.66 2.60 2.46
Overall Food Attitude 4.07 3.75 3.63 3.95 3.63 3.51
Purchase Intention 3.94 3.56 3.45 3.89 3.51 3.40

Table 4.7: Predicted Values of Dependent Variables depending on Nutritional Information

77
 5

4
Nutritional Percaption
Overall Food Attitude
3 Purchase Intention

2
No Info Calories only NI of 6 contents

Figure 4.1: Main Effect of Nutritional Information Disclosure

4.4.2 The Influence of Nutritional Menu Context

To test the main and moderating effects of nutritional menu context, multiple

regression analysis with two stages was implemented. In the first stage of analysis, a

model which included three independent variables was estimated. The three independent

variables included the indicator variables representing the three nutritional information

levels and the variable representing nutritional menu context. In the second stage, two

interaction terms were added to the model; the interaction variables were meant to

capture the interactions between the nutritional context and the nutritional information

levels. The interaction terms were calculated by determining the product of the

nutritional menu context variable and each of the two indicator variables representing the

three levels of nutritional information. The dummy variable which identified the target

item was also retained in the model as a control variable. Thus, the full regression model

was specified as follows:

78
 DVi = β0 + β1·TI + β2·TC +β3·NI + β4·NC + β5(TC×NC) + β6(NI×NC) +e

where DV=Dependent Variable with i=1, 2, 3; DV= Nutritional Perception if i=1;

DV= Overall Food Attitude if i=2; DV=Purchase Intention if i=3

TI = Target Item with 1 = healthy target item or 0 = unhealthy target item

TC = Total Calories only Information with 1 = provided or 0 = not provided

NI = Nutritional Information of 6 contents with 1 = provided or 0 = not provided

NC = Nutritional Context with 1 = healthy context or 0 = unhealthy context

TC×NC = Interaction term between TC and NC

NI×NC = Interaction term between NI and NC

e = error term

Main Effect of Nutritional Menu Context

In the first stage of analysis, the models for all three dependent variables were

significant at the 0.01 level (Nutritional Perception: F=48.77, d.f.=(4, 314), p<0.001;

Overall food attitude: F=7.11, d.f.=(4, 314), p<0.001; Purchase Intention: F=7.50, d.f.=(4,

314), p<0.001). The nutritional information level and the nutritional menu context

explained 37% of total variation in consumers’ nutritional perception, and explained 7%

and 8% of the variation in overall food attitude and purchase intention, respectively.

The effect of nutritional menu context on all three dependent variables was

significant at the 0.01 level. When other variables were held constant, consumers

perceived an item to be 0.69 units less nutritious in the healthy context than they did in

the unhealthy context (b4=-0.69, S.E.= 0.09). At the same time, consumers’ overall

attitude toward an item and likelihood of purchasing an item decreased by 0.46 and 0.57

79
units, respectively, when an item was placed with healthy items on the menu (Overall

food attitude: b4=-0.46, S.E.= 0.11; Purchase intention: b4=-0.57, S.E.= 0.13). Therefore,

the proposed hypothesis 2, that “the nutritional menu context will significantly influence

consumers’ food decision making, especially consumers’ nutritional perception, overall

food evaluation, and purchase intention,” was supported by this result.

Additionally, the nutritional level of the target item had a significant effect on

nutritional perception. Consumers’ perceived nutritional level for a healthy item

significantly differed by 0.98 units from their perceived nutritional level of an unhealthy

item. The significant effect of presented nutritional information was maintained after

adding the nutritional menu context variable.

Moderating Effect of Nutritional Menu Context

When the interactions were included in the second stage of analysis, the adjusted

R2 values significantly increased by 2% for the nutritional perception model and by 3%

for both the overall food attitude and purchase intention models. In addition, for all three

dependent variables, the changes in F-value due to the addition of the interaction terms

were significant at the 0.01 level.

When the level of nutritional information was interacted with the menu context,

the variable had a significant effect on all three of the dependent variables (Nutritional

perception; b5=-0.55 (S.E.=0.21), b6=-0.67 (S.E.=0.21); Overall food attitude: b5=-0.79

(S.E.= 0.27), b6=-0.75 (S.E.=0.27); Purchase intention: b5=-0.99 (S.E.= 0.31), b6=-0.72

(S.E.=0.31)). This indicates that consumers’ nutritional perception of a menu item when

the item was placed in the healthy context decreased by 0.55 units when they were

80
exposed to total calories information and by 0.67 units when they were exposed to

nutritional information for the 6 nutrient contents, when compared to the no nutritional

information treatment. Similarly, overall attitude toward a menu item when the item was

placed in the healthy context was 0.79 units less when total calories were disclosed and

0.75 units less when the nutritional information of the 6 nutrient contents was disclosed

than when no nutritional information was provided. In addition, consumers were less

likely to purchase an item placed in the healthy context when total calories only or

nutritional information for the 6 nutrient contents was offered than when no information

was provided. Thus, hypothesis 3, that “the nutritional menu context will significantly

moderate the effect of the presence of nutritional information on consumers’ nutritional

perception, overall food evaluation, and purchase intention,” was supported.

With inclusion of the interaction terms, the main effect of nutritional information

disclosure was eliminated for all three dependent variables, but the main effect of the

nutritional context remained significant in only the nutritional perception model.

The results of these multiple regression analyses can be found in Table 4.8; the

interaction effects between the nutritional information levels and the nutritional context

on the three dependent variables are depicted in Figures 4.2, 4.3, and 4.4.

81
 Model I Model II
Nutritional Attitude Purchase Nutritional Attitude Purchase
Independent Perception toward food Intention Perception toward food Intention
Variables
B (S.E.) B (S.E.) B (S.E.) B (S.E.) B (S.E.) B (S.E.)
Constant 3.02 (0.10)*** 4.18 (0.13)*** 4.18 (0.15)*** 2.81 (0.11)*** 3.92 (0.15)*** 3.89 (0.17)***
Target Item 0.98 (0.09)*** 0.12 (0.11) 0.04 (0.13) 0.99 (0.08)*** 0.12 (0.11) 0.04 (0.13)
Total Calories -0.06 (0.11) -0.32 (0.14)** -0.39 (0.16)** 0.21 (0.15) 0.07 (0.19) 0.11 (0.22)
Nutritional Info
-0.21 (0.11)** -0.44 (0.14)*** -0.50 (0.16)*** 0.12 (0.15) -0.07 (0.19) -0.14 (0.22)
of 6 contents
Context -0.69 (0.09)*** -0.46 (0.11)*** -0.57 (0.13)*** -0.29 (0.15)** 0.06 (0.19) 0.002 (0.22)
Cal*Context -0.55 (0.21)*** -0.79 (0.27)*** -0.99 (0.31)***
82

NI*Context -0.67 (0.21)*** -0.75 (0.27)*** -0.72 (0.31)**

Adj. R2 0.37 0.07 0.07 0.39 0.10 0.10
Model fit (F-Value) 48.77*** 7.11*** 7.50*** 35.42*** 6.67*** 6.93***
R2 Change 0.02*** 0.03*** 0.03***
F Change 5.75*** 5.39*** 5.38***
Note: *Significant at p<0.1; **Significant at p<0.05; ***Significant at p<0.01.
Standard errors are in the parentheses.

Table 4.8: Results of Regression Analysis: Effect of Nutritional Menu Context

82
 5

Nutritional  Perception
4

Healthy Context
Unhealthy Context
3

2
No Info Calories only NI of 6 contents 

Figure 4.2: Interaction Effect between the Presence of Nutritional Information and the
Nutritional Menu Context on Nutritional Perception

5
Overall Food Attitude

Healthy Context
Unhealthy Context
3

2
No Info Calories only NI of 6 contents 

Figure 4.3: Interaction Effect between the Presence of Nutritional Information and the
Nutritional Menu Context on Overall Food Attitude

83
 5

4
Purchase Intention

Healthy Context

3 Unhealthy Context

2
No Info Calories only NI of 6 contents 

Figure 4.4: Interaction Effect between the Presence of Nutritional Information and the
Nutritional Menu Context on Purchase Intention

Based on the estimated models for the three dependent variables, predicted values

were obtained and are displayed in Table 4.9. Without nutritional information, consumers’

perceived nutritional level for a healthy item was expected to be 3.31 out of 5 points in

the healthy context and 4.00 out of 5 points in the unhealthy context. On the other hand,

consumers’ nutritional perception of an unhealthy item was expected to be 2.33 in the

healthy context and 3.02 in the unhealthy context.

84
 Healthy Item
Healthy Context Unhealthy Context
Total NI of 6 Total NI of 6
Control Control
Calories contents Calories contents
Nutritional
3.31 3.25 3.10 4.00 3.94 3.79
Perception
Attitude
3.84 3.52 3.40 4.30 3.98 3.86
toward food
Purchase
3.65 3.26 3.15 4.22 3.83 3.72
Intention
Unhealthy Item
Healthy Context Unhealthy Context
Total NI of 6 Total NI of 6
Control Control
Calories contents Calories contents
Nutritional
2.33 2.27 2.13 3.02 2.96 2.81
Perception
Attitude
3.72 3.40 3.28 4.18 3.86 3.74
toward food
Purchase
3.61 3.22 3.11 4.18 3.79 3.68
Intention

Table 4.9: Predicted Values of Dependent Variables without Interaction Terms b/w the
Nutritional Menu Context and the Nutritional Information Level

4.4.3 The Role of Motivation to Process Nutritional Information

In order to test the moderating effect of motivation to process nutritional

information (MOT), multiple regression analysis with two stages was conducted. In the

first stage of analysis, two indicator variables for the three nutritional information levels

and a continuous variable for the motivation to process nutritional information were

included as independent variables in the model. In the second stage, two interaction

terms between the motivation to process information and each of the nutritional

information levels expressed by the two indicator variables were added to the model.

85
Due to a multi-collinearity problem among independent variables, a centered variable for

motivation to process information was added to the model. The centered variable was

calculated by subtracting the mean value (mean=3.22) from the original value of

motivation to process. Then, the interaction terms were obtained by multiplying the

centered motivation value and the indicator variables of nutritional information levels.

The effects of the healthiness of the target item and the nutritional context were also kept

in the model. The following regression model was used to test the moderating effect of

motivation to process nutritional information.

DVi = β0 + β1·TI + β2·NC + β3·TC + β4·NI + β5·CMOT + β6(TC×CMOT)

+ β7(NI×CMOT) + e

where DV=Dependent Variable with i=1, 2, 3

DV= Nutritional Perception if i=1; DV= Overall Food Attitude if i=2,

DV=Purchase Intention if i=3

TI = Target Item with 1 = healthy target item or 0 = unhealthy target item

NC = Nutritional Context with 1 = healthy context or 0 = unhealthy context

TC = Total Calories only Information with 1 = provided or 0 = not provided

NI = Nutritional Information of 6 contents with 1 = provided or 0 = not provided

CMOT = Centered Motivation to Process nutritional information with mean=0

TC×CMOT = Interaction term between TC and CMOT

NI×CMOT = Interaction term between NI and CMOT

e = error term

When interaction terms were not included, the suggested regression model was

significant at the 0.01 level for all three of the dependent variables (Nutritional

86
Perception: F=38.94, d.f.=(5, 313), p<0.001; Overall food attitude: F=5.68, d.f.=(5, 313),

p<0.001; Purchase Intention: F=5.98, d.f.=(5, 313), p<0.001). Thirty-seven percent of

total variance in consumers’ nutritional perception was explained by the independent

variables with two control variables. The variables also accounted for 7% of total

variance of both overall food attitude and purchase intention.

There was no significant main effect of consumers’ motivation to process

nutritional information when other variables were held constant. But, the main effects of

both nutritional context and the presence of nutritional information were still significant

for all three dependent variables.

After adding variables capturing the interaction between nutritional information

levels and the motivation to process to the nutritional perception model, adjusted R2

increased by 2 percentage points. Also, after including the interaction terms, the F-value

change was only significant in the nutritional perception model (F=4.11, df=(2, 311),

p<0.05). The changes in adjusted R2 were not significant in the using overall food

attitude model or in the purchase intention model.

There was, however, a significant effect of the interaction between the nutritional

information levels and the motivation to process nutritional information on nutritional

perception and overall food attitude. In particular, the interaction between the presence

of nutritional information for the 6 nutrient contents and motivation to process was found

to be significant in the nutritional perception model (b7=-0.18, S.E.=0.09). This indicates

that consumers’ perceived nutritional level of a menu item decreased by 0.18 units as

their motivation to process increased by one unit, only if nutritional information for the 6

nutrient contents was offered. In contrast, as motivation to process increased, consumers

87
had more favorable food attitudes for menu items only when total calorie information

was provided (b6=0.23, S.E.=0.13). Thus, overall food attitude increased by 0.23 units

when consumers’ motivation to process increased by 1 unit, only when total calorie

information was disclosed. This result partially supports hypothesis 4 that predicted that

motivation to process would have a significant moderating effect on consumer food

decision making. Table 4.10 displays the results of this multiple regression analysis.

Figure 4.5 and Figure 4.6 show the interaction plots that represent the significant

interaction effects on nutritional perception and overall food attitude. For the interaction

plots of nutritional information levels with motivation to process, the three different

levels of motivation to process were calculated by adding and subtracting two times the

standard deviation from zero, which was the mean of the centered motivation to process

variable (S.D.=1.14). The interaction plots show that consumers with high motivation to

process perceived the nutritional level of a menu item differently than do those with low

or medium levels of motivation to process. The negative association between the high

motivation to process and the nutritional information level is clear (See Figure 4.5). In

terms of overall attitude toward a menu item, consumers with moderate levels of

motivation to process had more favorable attitudes toward a menu item when nutritional

information was provided. However, consumers who had low or high motivation levels

showed a more negative attitude toward menu items when nutritional information was

present (See Figure 4.6).

88
 Model I Model II
Nutritional Attitude Purchase Nutritional Attitude Purchase
Independent Perception toward food Intention Perception toward food Intention
Variables
B (S.E.) B (S.E.) B (S.E.) B (S.E.) B (S.E.) B (S.E.)
Constant 3.02 (0.10)*** 4.18 (0.13)*** 4.18 (0.15)*** 3.02 (0.10)*** 4.17 (0.13)*** 4.19 (0.15)***
Target Item 0.98 (0.09)*** 0.12 (0.11) 0.04 (0.13) 0.99 (0.09)*** 0.13 (0.11) 0.05 (0.13)
Context -0.69 (0.09)*** -0.46 (0.11)*** -0.57 (0.13)*** -0.68 (0.09)*** -0.46 (0.11)*** -0.57 (0.13)***
Total Calories -0.07 (0.11) -0.32 (0.14)** -0.39 (0.16)** -0.09 (0.11) -0.32 (0.14)** -0.39 (0.16)**
Nutritional Info
-0.22 (0.11)** -0.44 (0.14)*** -0.50 (0.16)*** -0.22 (0.11)** -0.43 (0.14)*** -0.49 (0.16)***
of 6 contents
Motivation (CMOT) 0.01 (0.04) -0.009 (0.05) 0.000 (0.06) 0.06 (0.07) -0.08 (0.09) -0.02 (0.10)
89

Cal*CMOT 0.07 (0.10) 0.23 (0.13)* 0.15 (0.15)

NI*CMOT -0.18 (0.09)** 0.02 (0.12) -0.07 (0.14)
Adj. R2 0.37 0.07 0.07 0.39 0.08 0.07
Model fit (F-Value) 38.94*** 5.68*** 5.98*** 29.54*** 4.69*** 4.65***
R2 Change 0.02** 0.01 0.008
F Change 4.11** 2.13 1.29
Note: *Significant at p<0.1; **Significant at p<0.05; ***Significant at p<0.01.
Standard errors are in the parentheses.

Table 4.10: Results of Regression Analysis: Effect of Motivation to Process Nutritional Information

89
 5

Nutritional perception
4
Low MOT
Medium MOT
High MOT
3

2
No Info Calories only NI of 6 contents

Figure 4.5: Interaction Effect between the Presence of Nutritional Information and the
Motivation to Process on Nutritional Perception

5
Overall Food Attitude

4
Low MOT
Medium MOT
High MOT
3

2
No Info Calories only NI of 6 contents

Figure 4.6: Interaction Effect between the Presence of Nutritional Information and the
Motivation to Process on Overall Food Attitude

90
4.4.4 The Role of Nutritional Knowledge

The moderating effect of nutritional knowledge was analyzed using multiple

regression analysis. Again, two stages of analyses were completed. Like in the previous

model, two indicator variables representing the three nutritional information levels and a

continuous variable for nutritional knowledge were included in the first stage model as

independent variables. Nutritional knowledge was measured using two different

instruments: objective nutritional knowledge was measured using fifteen multiple choice

questions and subjective nutritional knowledge was measured using a five-point Likert-

scale. Thus, two different models – one for objective nutritional knowledge and one for

subjective nutritional knowledge – were developed.

DVi = β0 + β1·TI + β2·NC +β3·TC + β4·NI + β5·CONK + β6(TC×CONK)

+ β7(NI×CONK) + e

DVi = β0 + β1·TI + β2·NC +β3·TC + β4·NI + β5·CSNK + β6(TC×CSNK)

+ β7(NI×CSNK) + e

where DV=Dependent Variable with i=1, 2, 3

DV= Nutritional Perception if i=1; DV= Overall Food Attitude if i=2,

DV=Purchase Intention if i=3

TI = Target Item with 1 = healthy target item or 0 = unhealthy target item

NC = Nutritional Context with 1 = healthy context or 0 = unhealthy context

TC = Total Calories only Information with 1 = provided or 0 = not provided

NI = Nutritional Information of 6 contents with 1 = provided or 0 = not provided

CONK = Centered Objective Nutritional Knowledge with mean=0

TC×CONK = Interaction term between TC and CONK

91
 NI×CONK = Interaction term between NI and CONK

CSNK = Centered Subjective Nutritional Knowledge with mean=0

TC×CSNK = Interaction term between TC and CSNK

NI×CSNK = Interaction term between NI and CSNK

e = error term

In the second stage, the model included interaction terms between objective

nutritional knowledge and each of the two nutritional information level variables, as well

as between subjective nutritional knowledge and each of the two nutritional information

level variables. The interaction terms were obtained by calculating the product of the

nutritional information indicator variable and the knowledge score, for both the

subjective and objective knowledge scores. The interaction terms again created multi-

collinearity problems in the models, so the centered objective and subjective nutritional

knowledge variables were used instead of the original values of the nutritional knowledge

variables. The centered objective and subjective nutritional knowledge values were

calculated as the differences between the original values and the mean values (mean of

objective nutritional knowledge=8.17 out of 15; mean of subjective nutritional

knowledge=3.20 out of 5). Using the centered variables, the interaction terms were

recalculated and included in the model. The effects of the target item and the nutritional

context were controlled for in all models.

The Effect of Objective Nutritional Knowledge

For the models which included the objective nutritional knowledge variable, the

model fits for each of the three dependent variables were significant at the 0.01 level

92
(Nutritional Perception: F=38.90, d.f.=(5, 313), p<0.001; Overall food attitude: F=5.71,

d.f.=(5, 313), p<0.001; Purchase Intention: F=6.02, d.f.=(5, 313), p<0.001). Nutritional

information levels and objective nutritional knowledge accounted for 37% of total

variance in consumers’ nutritional perception. This value increased by 2 percentage

points when the interaction terms were included. The independent variables alone

explained only 7% of total variance of both overall food attitude and purchase intention.

However, for these two independent variables, inclusion of the interaction terms between

nutritional information levels and objective nutritional knowledge did not significantly

increase the explained variation. Including interaction terms did not significantly change

F-values for any of the three dependent variables (See Table 4.11).

The results of multiple regression revealed that objective nutritional knowledge

did not carry a significant main effect when other variables were kept constant.

Additionally, a significant interaction effect between the presence of nutritional

information and objective nutritional knowledge was not found for any of the three

dependent variables: nutritional perception, overall food attitude, and purchase intention.

However, the other independent variables such as the nutritional context and the presence

of nutritional information did influence the three dependent variables. This implies that

consumers’ objective nutritional knowledge did not significantly moderate the effect of

the presence of nutritional information. Table 4.11 outlines the result of multiple

regression analysis.

93
 Model I Model II
Nutritional Attitude Purchase Nutritional Attitude Purchase
Independent Perception toward food Intention Perception toward food Intention
Variables
B (S.E.) B (S.E.) B (S.E.) B (S.E.) B (S.E.) B (S.E.)
Constant 3.02 (0.10)*** 4.19 (0.13)*** 4.19 (0.15)*** 3.01 (0.10)*** 4.19 (0.13)*** 4.21 (0.15)***
Target Item 0.98 (0.09)*** 0.12 (0.11) 0.04 (0.13) 0.99 (0.09)*** 0.11 (0.11) 0.03 (0.13)
Context -0.69 (0.09)*** -0.46 (0.11)*** -0.57 (0.13)*** -0.70 (0.09)*** -0.45 (0.11)*** -0.56 (0.13)***
Total Calories -0.07 (0.11) -0.33 (0.14)** -0.40 (0.16)** -0.06 (0.11) -0.35 (0.14)** -0.41 (0.16)**
Nutritional Info
-0.22 (0.11)** -0.45 (0.14)*** -0.51 (0.16)*** -0.21 (0.11)* -0.45 (0.14)*** -0.52 (0.16)***
of 6 contents
Objective NK
0.003 (0.01) 0.01 (0.02) 0.01 (0.03) -0.02 (0.03) 0.02 (0.04) 0.05 (0.16)
(CONK)
94

Cal*CONK 0.05 (0.05) 0.01 (0.06) -0.05 (0.07)

NI*CONK 0.04 (0.04) -0.05 (0.06) -0.07 (0.06)
Adj. R2 0.37 0.07 0.07 0.37 0.07 0.07
Model fit (F-Value) 38.90*** 5.71*** 6.02*** 27.86*** 4.21*** 4.45***
R2 Change 0.002 0.003 0.003
F Change 0.55 0.49 0.57
Note: *Significant at p<0.1; **Significant at p<0.05; ***Significant at p<0.01.
Standard errors are in the parentheses.

Table 4.11: Results of Regression Analysis: Effect of Objective Nutritional Knowledge

94
The Effect of Subjective Nutritional Knowledge

In order to test the role of subjective nutritional knowledge as a moderator,

multiple regression models containing two categorical variables for nutritional

information levels and a centered subjective nutritional knowledge variable were

estimated. The fit of all three models was significant at the 0.01 level (Nutritional

Perception: F=38.89, d.f.=(5, 313), p<0.001; Overall food attitude: F=5.78, d.f.=(5, 313),

p<0.001; Purchase Intention: F=6.21, d.f.=(5, 313), p<0.001). Thirty-seven percent of

the total variation in nutritional perception scores was explained by the independent

variables and control variables, while only 7% of the variation of both overall food

attitude and purchase intention was explained by the independent variables and control

variables. In the second stage, interaction terms between nutritional information levels

and subjective nutritional knowledge were added, but they did not significantly increase

the adjusted R2. Also, the changes in F-value due to the addition of the interaction terms

were not large enough to be significant for any of the three dependent variables:

nutritional perception, overall food attitude, and purchase intention.

According to the results of multiple regression analysis, subjective nutritional

knowledge did not carry a significant main effect for any of the three dependent variables

(Nutritional perception; b5=0.001, S.E.=0.05; Overall food attitude: b5=-0.049, S.E.= 0.06;

Purchase intention: b5=0.07, S.E.= 0.07). Additionally, the interaction terms between

nutritional information levels and subjective nutritional knowledge did not influence

consumers’ food evaluation processes: specifically nutritional perception, overall food

attitude, and purchase intention were unaffected by the interaction terms. However,

similar to the results of the prior analyses, the nutritional context and the presence of

95
nutritional information were significantly associated with all three dependent variables.

Consequently, the level of consumers’ subjective nutritional knowledge did not moderate

the effect that nutritional information disclosure had on consumers’ food decision making

process. Table 4.12 shows the results of the multiple regression analysis related to

subjective nutritional knowledge.

Additionally, it was investigated whether a term meant to capture the interaction

between objective and subjective nutritional knowledge would have any effect on the

three dependent variables. After adding the interaction term to both the objective and

subjective nutritional knowledge regression models, no significant interaction effect

between objective and subjective nutritional knowledge on consumer food evaluation and

purchase intention appeared. Further, a three-way interaction effect between both

nutritional knowledge variables and nutritional information disclosure proved

insignificant.

Consequently, neither objective nor subjective nutritional knowledge significantly

moderated the effect of nutritional information disclosure on consumer food decision

making, and thus hypothesis 5 that proposed that nutritional knowledge could act as a

moderator in the effect of nutritional information disclosure was not supported in this

study.

96
 Model I Model II
Nutritional Attitude Purchase Nutritional Attitude Purchase
Independent Perception toward food Intention Perception toward food Intention
Variables
B (S.E.) B (S.E.) B (S.E.) B (S.E.) B (S.E.) B (S.E.)
Constant 3.02 (0.10)*** 4.19 (0.13)*** 4.19 (0.15)*** 3.02 (0.10)*** 4.19 (0.13)*** 4.20 (0.15)***
Target Item 0.98 (0.09)*** 0.12 (0.11) 0.05 (0.13) 0.99 (0.09)*** 0.12 (0.11) 0.05 (0.13)
Context -0.69 (0.09)*** -0.46 (0.11)*** -0.58 (0.13)*** -0.70 (0.09)*** -0.47 (0.11)*** -0.58 (0.13)***
Total Calories -0.06 (0.11) -0.33 (0.14)** -0.40 (0.16)** -0.07 (0.11) -0.34 (0.14)** -0.41 (0.16)**
Nutritional Info
-0.21 (0.11)** -0.45 (0.14)*** -0.51 (0.16)*** -0.21 (0.11)** -0.45 (0.14)*** -0.51 (0.16)***
of 6 contents
Subjective NK
0.001 (0.05) 0.04 (0.06) 0.07 (0.07) -0.02 (0.08) 0.06 (0.11) 0.09 (0.12)
(CSNK)
97

Cal*CSNK 0.12 (0.12) 0.09 (0.15) 0.07 (0.18)

NI*CSNK -0.05 (0.12) -0.13 (0.15) -0.11 (0.17)
Adj. R2 0.37 0.07 0.07 0.37 0.07 0.07
Model fit (F-Value) 38.89*** 5.78*** 6.01*** 28.08*** 4.44*** 4.59***
R2 Change 0.004 0.006 0.003
F Change 1.03 1.09 0.58
Note: *Significant at p<0.1; **Significant at p<0.05; ***Significant at p<0.01.
Standard errors are in the parentheses.

Table 4.12: Results of Regression Analysis: Effect of Subjective Nutritional Knowledge

97
 Based on the results of multiple regression analysis, the results of the hypothesis

tests are summarized in Table 4.13.

Hypotheses Result

H1. The presence of nutritional information on the menu will

significantly influence consumer food decision making, Supported
especially consumers’ nutritional perception, overall food
evaluation, and purchase intention.

H2. The nutritional menu context will significantly influence

consumers’ food decision making, especially consumers’ Supported
nutritional perception, overall food evaluation, and
purchase intention.

H3. The nutritional menu context will significantly moderate the

effect of the presence of nutritional information on Supported
consumers’ nutritional perception, overall food evaluation,
and purchase intention.

H4. The motivation to process nutritional information will

significantly moderate the effect of the presence of
Partially Supported
nutritional information on consumers’ nutritional
perception, overall food evaluation, and purchase intention.

H5. The nutritional knowledge will significantly moderate the

effect of the presence of nutritional information on Not Supported
consumers’ nutritional perception, overall food evaluation,
and purchase intention.

Table 4.13: Results of Hypotheses Testing

4.5 Additional Analysis: Logistic Regression Analysis

Although a hypothesis about the actual choice of a menu item was not pre-

specified, additional analysis was conducted to investigate what factors may be relevant

predictors of item choice. Respondents were asked to select one item among the three
98
provided items on the menu; their choice was recorded as an indicator variable, which

equaled 1 if the target item was selected and equaled 0 otherwise. This variable acted as

the dependent variable in a logistic regression analysis, an analysis method used to find

the relationship between explanatory variables and a binary dependent variable.

All variables related to the presence of nutritional information were included in

the first logistic regression model; nutritional level of the target item, level of nutritional

information, nutritional menu context, motivation to process nutritional information, and

nutritional knowledge. The second model also contained other factors which could

influence the choice of menu item, such as the perceived healthiness of an item, the

expectation of taste and diet concerns for health or weight loss. In addition, while this

study employed three dependent variables to determine the effects of nutritional

information disclosure, in past studies, only consumers’ perceived nutrition had been

considered relevant in food choices (Lee & Cranage, 2007; Yoon, Sydny-Busso, & Parsa,

2009). Overall food attitude and purchase intention, however, have been mainly used as

proxy measures of the actual food choice. Thus, overall food attitude and purchase

intention were included in the final model with other variables. Demographic variables

including age, gender, and BMI were also excluded in the model due to non-significance.

Using Hosmer and Lemeshow (H-L)’ goodness of fit test, the models’ fit was

tested. Results indicated that the second model was the only well fitted model since it

had a p-value greater than the 0.05 level.

According to the results of the logistic regression, the presence of total calorie

information had a significant effect on the choice of target menu items (Exp (BTC)=0.62).

The odds of a target item over non-target item choice decreased by 0.62 when total

99
calorie information was provided. Further, when objective nutritional knowledge

increased by 1 score, the odds of a target item menu choice decreased by 0.91. However,

consumers selected the target items almost 3 times more often when it was placed on the

menu with unhealthy menu items as compared to when it was placed in the healthy menu

context (Exp (BNC)=2.99).

In the second model, it was found that a choice of a target item became 1.44 times

more likely with every 1 unit increase in health concern, but that the odds of the target

item choice decreased by 0.79 when interest in weight loss increased by 1 unit. In

addition, consumers selected target items 2.86 times more often when the expectation of

taste increased by 1 unit and 1.35 times more often when the perceived nutritional value

of the target item increased by 1 unit.

According to results of the third model, overall food attitude and purchase

intention were significantly associated with the choice of target items. A one unit

increase in overall food attitude and purchase intention made consumers choose target

items 2.27 times more often and 6.25 units more often, respectively. However, nutritional

perception and taste were not significant in this model.

The relationship between nutritional information disclosure and item choice was

not significantly moderated by motivation to process or nutritional knowledge.

Consequently, consumers were more likely to use the total calories information than they

were to use the nutritional information of 6 nutrient contents in the choice of a menu item.

Nutritional menu context was also an influential factor in choosing a menu item. As

expected, overall attitude toward a menu item and purchase intention were good

predictors of consumer choice behavior. The results of logistic regression are presented

100
in Table 4.14.

Dependent Variable = Choice of a target item

Model I Model II Model III
Independent Variables a
Exp(B) (S.E.) Exp(B) (S.E.) Exp(B) (S.E.)
Constant 0.63 (0.26)* 0.002 (1.26)*** <0.001 (1.77)***
Target Item b 1.22 (0.24) 1.24 (0.31) 1.29 (0.39)
Total Calories c 0.62 (0.29)* 0.54 (0.32)** 0.76 (0.41)
Nutritional Info.
0.65 (0.29) 0.67 (0.32) 1.21 (0.41)
of 6 contents c
Unhealthy Context 2.99 (0.24)*** 2.81 (0.28)*** 3.43 (0.37)***
Motivation 0.99 (0.12) 0.98 (0.16) 1.03 (0.19)
Objective NK 0.91 (0.05)* 0.90 (0.06)* 0.82 (0.08)***
Subjective NK 1.05 (0.16) 0.93 (0.18) 0.79 (0.23)
Diet for Health d 1.44 (0.19)** 1.47 (0.25)
Diet for Weight Loss d 0.79 (0.13)* 0.86 (0.16)
e
Taste 2.86 (0.18)*** 0.85 (0.27)
Nutritional Perception 1.35 (0.17)* 0.68 (0.23)
Overall attitude 2.27 (0.35)**
Purchase Intention 6.25 (0.47)***
-2Log Likelihood χ2 409.42 354.05 235.63
H-M χ2 17.65** 10.18 17.75**
Note: *Significant at p<0.1; **Significant at p<0.05; ***Significant at p<0.01.
Standard errors are in the parentheses.
a. Exponential value of a coefficient
b. Indicator variable with reference group = healthy target item
c. No nutritional information condition plays as a reference group.
d. Measured by five-point Likert scale ranged from 1=not interested at all to
5=very interest
e. Measured by five-point Likert scale ranged from 1=very poor to 5=very good

Table 4.14: Results of Logistic Regression Analysis

101
 CHAPTER 5: CONCLUSIONS

The purpose of this study was to examine the effect of nutritional information

disclosure on consumer food decision making in a restaurant. Other influential factors in

the nutritional information process were also investigated. In detail, this study

determined the effect of three different levels of nutritional information and three

variables that influence consumer food decision making: menu context, motivation to

process nutritional information, and nutritional knowledge.

By means of an experiment, data were collected from undergraduate and graduate

students at a large mid-western university and analyzed using several methods of analysis

such as analysis of variance, multiple regression, and logistic regression. The results of

this study are summarized in this chapter. This chapter also includes conclusions,

implications for research and industry areas, and limitations.

5.1 Summary of Results and Discussion

Influence of Providing Nutritional Information

As expected, there was evidence that the disclosure of nutritional information on a

menu significantly influenced consumer food attitude and purchase behavior. The

presence of nutritional information, both total calories and nutritional information

concerning the six nutrient contents, lead consumers to have less favorable attitudes

102
toward a menu item and less purchase intention for that item. Despite providing

nutritional values similar to, or even healthier than, the actual nutritional levels of menu

items, consumers evaluated both healthy and unhealthy menu items more negatively.

This may be due to the fact that the provided nutritional level of an item on the menu is

worse than consumers’ nutritional expectations. Consumers were likely to be

disappointed with the manipulated nutritional level of an item, and thus their attitude

toward the menu item and purchase intention would be less favorable. This finding

supports the results of previous studies which showed that consumers tended to

underestimate calories or fat, and that the differences between the expected and the actual

nutritional values tend to be greater for unhealthy foods (Burton, et al., 2006; Chandon &

Wansink, 2007a, 2007b).

Another interesting finding was that the perceived nutritional evaluation of a

menu item was only influenced when the nutritional information provided included the

total calories plus five nutrient contents of the item. This implies that consumers did not

change their nutritional perception of a menu when total calories alone were provided.

On the other hand, the result of logistic regression revealed that total calorie information

significantly influenced menu choice. Consumers chose a target item less frequently

when exposed to total calories. Rather than processing nutritional information of all six

nutrient contents, consumers used only total calorie information in selecting an item,

perhaps because calorie information is easy and simple to use in the choice. Thus,

consumers are likely to use simple information at the point of purchase, but nutritional

information for the six nutrient contents did influence their perception or attitude

development processes, which were significant predictors of consumer choice behavior.

103
Consequently, nutritional information for the six nutrient contents directly influenced

food evaluation and purchase intention but only indirectly influenced the actual choice

behavior, while total calorie information had a direct effect on overall food attitude,

purchase intention, as well as the choice itself. Thus, the presence of total calorie

information could encourage consumers to use the presented nutritional information on

the menu and to choose the healthy item in a restaurant. However, from a long-term

health perspective, it is more important to provide nutritional information for more

nutrients and to educate consumers in using broader nutritional information in their

choices.

Influence of Nutritional Menu Context

In this study, there were several interesting findings which relate to the nutritional

menu context. Consistent with prior studies, the nutritional menu context significantly

influenced consumer food decision making processes, as well as food choice (Kozup, et

al., 2003; Burton & Creyer, 2004). Comparing the provided nutritional information of all

alternatives on the menu influenced how consumers perceived the nutritional levels of a

target menu item: consumers tended to evaluate an item relative to others. Comparing

effects across dependent variables, nutritional menu context had the greatest effect on

consumers’ nutritional perception (See Table 4.8). Thus, a menu item accompanied by

less healthy items was perceived as a more healthy food and evaluated more favorably

than had it been accompanied by healthy items. Furthermore, menu items in the

unhealthy context were three times more likely to be chosen than items in the healthy

context. Consequently, the context effect existed in evaluating menu items using the

104
provided nutritional information, and thus nutritional information of alternatives on the

menu might be considered as the reference point from which a target item is evaluated.

This finding is consistent with prior studies (Burton & Creyer, 2004; Kozup, et al., 2003).

Along with the significant main effect of nutritional menu context, the interaction

of nutritional information disclosure with nutritional menu context also had a significant

effect on consumer food decision making. The healthy menu context further worsened

the negative effect of nutritional information exposure on nutritional perception, food

evaluation and purchase intention. This result is consistent with previous studies wherein

menu context influenced the effect of nutritional information on consumers food

evaluation (Burton & Creyer, 2004; Kozup, et al., 2003). In particular, the presence of

total calories in the healthy context had the greatest negative effect on consumer purchase

intention (See Table 4.8) among the three dependent variables, while nutritional

information for total calories plus five nutrients in the healthy context had an especially

negative effect on overall food attitude.

The Role of Motivation to Process Nutritional Information

Although motivation to process nutritional information did not have a significant

main effect on consumer food decision making, it played a role as a significant moderator

in evaluating a menu item with nutritional information. Nutritional information

disclosure interacted with consumer’s motivation to process had a significant effect on

perceived nutrition and overall attitude toward a menu item. More specifically, highly

motivated consumers perceived the nutritional level of a menu item more negatively only

if nutritional information for the six nutrient contents was presented, whereas for

105
consumers with low or medium levels of motivation to process, exposure to nutritional

information for the six nutrient contents lead to improved nutritional perception. Thus,

the change in nutritional perception due to the presence of total calories plus five nutrient

contents information was greater when the motivation to process nutritional information

was high.

Further, when total calorie information was present, the level of consumers’

motivation to process significantly influenced overall food evaluation. Consumers who

had medium levels of motivation to process positively evaluated a menu item, while

consumers low or high in motivation showed the more negative attitude toward a menu

item. However, there was no significant interaction effect between motivation and

nutritional information disclosure on purchase intention or on the actual choice of a menu

item. Consequently, the motivation to process nutritional information directly moderated

the effect of nutritional information disclosure on nutritional perception and overall

attitude, but not on purchasing behavior.

The Role of Nutritional Knowledge

Levels of nutritional knowledge were not significantly associated with

consumer’s food decision making process; further, knowledge scores did not moderate

the main effect of nutritional information disclosure on consumer food evaluation and

purchase intention.

However, both objective and subjective nutritional knowledge had positive

associations with source credibility (bobj=0.05, p<0.05; bsbj=0.09, p<0.10). This implies

that consumers high in objective or subjective nutritional knowledge believed that the

106
provided nutritional information was more accurate (See Table D.1 in Appendix D).

Although the presented nutritional values of menu items were manipulated, they were

close to the actual nutritional values. Thus, consumers high in nutritional knowledge had

more ability to process the provided nutritional information accurately. This result is

consistent with previous studies, which indicated that nutritional knowledge had a

significant effect on nutrition label use and accuracy of nutritional information use

(Drichoutis, et al., 2005; Droms, 2006; Moorman, 1993).

In addition, according to the Elaboration Likelihood Model (Petty & Cacioppo,

1986), the effect of ability to process information on attitude or perception is more

effective when consumers have a high level of motivation to process information. This

means that there should be a significant effect of nutritional knowledge on consumer food

decision making if consumers have enough motivation to process nutritional information.

Thus, the effect of nutritional knowledge interacted with motivation to process was

further investigated. According to the results of multiple regression, the interaction effect

between subjective nutritional knowledge and motivation to process was significant on

overall food attitude and purchase intention (overall food attitude: b=-0.12, p<0.05;

purchase intention: b=-0.13, p<0.05) (See Table D.2 in Appendix D). This result

illustrates a very interesting point about nutritional knowledge. Consumers high in both

subjective nutritional knowledge and motivation to process, which are associated with

consumers’ psychological state, evaluated a menu item more unfavorably and were less

likely to purchase a menu item. The effect of objective nutritional knowledge on

consumer food evaluation, however, was not influenced by the level of motivation to

process; further, the results of the logistic regression showed that objective nutritional

107
knowledge directly influenced the choice of a menu item (See Table 4.14). When other

variables were held constant, the odds ratio indicated that the choice of a target item over

a non-target item became significantly less likely when consumers’ objective nutritional

knowledge score increased.

In conclusion, consumers’ nutritional knowledge level did not significantly

influence consumers’ food evaluation or purchase intention of a menu item at a restaurant.

However, regardless of the presence of nutritional information, consumers’ nutritional

knowledge was significantly associated with how accurately they used nutritional

information. Also, objective nutritional knowledge itself affected consumer food choice

directly, but subjective nutritional knowledge influenced food evaluation and purchase

intention only when consumers had enough motivation to process nutritional information.

In particular, psychological encouragement is very important for consumers to engage in

healthy behaviors.

Influence of Other variables

Due to the homogeneous student sample, significant differences in food

evaluation, purchase intention and choice itself depending on demographic variables such

as age, income, and gender were not found. However, other attributes such as taste

expectation or interest in weight loss or health were significantly related to food choice.

According to the results of logistic regression model II (Table 4.14), increased nutritional

perception or taste expectation lead consumers to select a target item more frequently. In

particular, the expected taste of an item (Exp(B)=2.86, p<0.01) had a greater effect on the

likelihood of the item choice than did the perceived nutrition of the item (Exp(B)=1.35,

108
p<0.10). This result is consistent with other previous studies (Lee & Cranage, 2007;

Palmer & Leontos, 1995). Moreover, diet concerns for health or weight loss were

influential factors; if consumers were more interested in diet for health reasons, the

likelihood of the choice of a target item increased 1.44 times; if they were concerned with

diet for weight-loss reasons, the likelihood of the choice of a target item decreased. Thus,

the structure of consumer food choice behavior in a restaurant is complicated and other

influential components should be investigated in the future.

5.2 Conclusions and Implications

As menu labeling legislation requiring nutritional information on menus at

restaurants becomes increasingly prevalent, the discussion about whether labeling laws

will help improve both consumer healthy eating and the negative image of the restaurant

industry has become controversial. Thus, this study examined the effect of nutritional

information disclosure on menus and several factors affecting the use of nutritional

information in order to provide more effective strategies for policy makers and

restaurateurs. Using the experimental design method, several interesting findings were

obtained; significant implications for both practical and research areas arose.

Firstly, the results of this study indicate that the availability of nutritional

information on the menu influenced consumer food decision making in a restaurant

directly or indirectly. At the point of selecting an item in a restaurant, only provision of

total calorie information had a significant effect. The law requiring only total calories on

the menu, which was enacted by New York City, seems to be more efficient for both

consumers in selecting an item and restaurants in designing the menu. However,

109
nutritional information of other nutrients also played a significant role in changing

consumers’ menu choice indirectly. Thus, the influence that providing nutritional

information of other nutrients should not be passed over in enacting menu labeling laws.

In the case of King County, it is required to post calorie, fat, sodium, and carbohydrate

information on menus. But, restaurants using menu boards display only calorie

information with price on the board, and nutritional information for other nutrients is

offered in a plainly visible format (Conlin, July 27, 2007). The King County model

allows consumers to use more nutritional information in selecting food and policy makers

to develop more efficient and practical menu labeling regulations.

In addition, both levels of nutritional information – total calories and nutritional

information for the six nutrient contents – had significant negative effects on consumer

food decision making. When nutritional information was presented on the menu,

consumers evaluated the nutritional levels of menu items and overall food more

unfavorably and were less likely to purchase items. This implies that consumers

expected that menu items were much healthier than they were in reality. Thus, it is

necessary to inform consumers of the exact nutritional information of menu items and

educate them to process nutritional information accurately.

Consumer nutritional knowledge, both objective and subjective, was associated

with the accuracy of the use of nutritional information, and there were significant effects

of objective nutritional knowledge on food choice itself. In particular, consumers’

psychological variables, motivation and confidence in using nutritional information

(subjective nutritional knowledge) had significant interactional effects on food attitude

and purchase intention. Thus, it is important not only to provide more opportunities to

110
learn about health and nutrition but also to increase consumers’ interest or concern with

health and nutritional information.

In the restaurant industry, one way to reduce the negative effect of nutritional

information disclosure is to design the menu more effectively. It was found that the

nutritional menu context not only influenced consumer food decision making but

moderated the effect of nutritional information disclosure. When a menu item was

accompanied by healthier items, it was evaluated more unfavorably than when it was

accompanied by less healthy items; this effect became stronger when nutritional

information was provided. Thus, how items are grouped together on a menu becomes a

critical issue. For example, it is possible to design a menu page for only healthy items or

a menu emphasizing a few healthy items in an unhealthy context to reduce the context

effect. It is beneficial to design a menu providing an unhealthy menu context containing

healthy choices, thereby encouraging healthy food choices. Thus, to emphasize a

particular item in terms of healthiness, restaurateurs can utilize the menu context effect.

Finally, this study contributes to the existing literature in several ways. Although

a number of studies have been conducted meant to determine the effect of nutritional

information disclosure, they have been associated with The Nutrition Labeling and

Education Act (NLEA) and therefore, limited to packaged foods. This study, however,

focused on consumers’ nutritional information use in a restaurant context; the study also

identified significant factors that influence consumers’ nutritional information processes,

such as motivation to process, nutritional knowledge and menu context. Additionally, it

was found that there is a more complicated structure underlying consumer food decision

making. The three dependent variables in this study – nutritional perception, overall food

111
attitude, and purchase intention – were significantly correlated. In addition, nutritional

perception did not have a direct relationship with the actual food choice, while other two

dependent variables were significant predictors of consumer food choice (See Table 4.14).

This result implies that overall food attitude and purchase intention mediated the

relationship between nutritional perception and actual food choice, and that consumers’

nutritional perception should be considered as one of the relevant attributes in a food

choice decision. Accordingly, this study provides the initial outline of the consumer food

decision process when nutritional information is provided; this issue should be dealt with

in future research.

5.3 Limitations and Future Research

Although this study provides significant contributions to both research and

industry, it has several limitations that may have influenced the results. First, the sample

of this study consisted of mainly undergraduate students. This homogenous group

facilitated control of other influential demographic factors in the experiment, but the

homogeneity also limits the scope of this study to one typical segment of consumers.

Student perception and behavior in a restaurant may differ from the behavior of others,

such as people who are older or less educated. Thus, the findings cannot be generalized

to all consumers in other market segments. It would therefore be helpful to conduct the

same experiment targeting other market segmentation.

Additionally, the number of respondents in each condition was only 26 or 27,

which is a relatively small sample size even though the total sample size was sufficient,

more than 300. Thus, a larger sample size would allow for stronger results.

112
 Target items were always accompanied by alternatives on the menu so as to

mimic a real restaurant situation as closely as possible: there is no restaurant menu where

only one item is offered on the menu. However, this requires that menu context must

always be controlled for in testing other effects such as the main effects of nutritional

information disclosure, motivation to process, and nutritional knowledge. Identification

of the effects of other variables would have been facilitated by the presence of a control

group, a group which was not provided with a menu context. Thus, future study should

employ a modified experimental design including three different types of menu context:

absence, healthy, and unhealthy contexts.

Another limitation of the study is due to the selected menu items. This study

adopted menu items only available in a casual or fine dining restaurant, and only chicken

items were offered to eliminate food preferences. Thus, there is a possibility that

consumers use different evaluation criteria or decision making processes when nutritional

information is provided in fast food restaurants. Furthermore, it was found that meat was

another preferred food source (34%), so consumers’ food preferences were not

represented in food decision making. Even if vegetarians only represented a small

portion of the sample (3%), they remain another segment of consumers. Thus, there

might be differences in evaluation or decision making processes depending upon

consumers’ food preference.

In addition to food preference, taste of food is another critical factor that

influences food choice. In the study conducted by Dorms (2006), it was found that

consumers are more likely to choose tasty foods when they eat out because they prefer to

enjoy themselves regardless of money or health. Lee and Cranage (2007) also found that

113
the relative importance of taste in menu selection was greater than the importance of

nutrition. Therefore, in future research, how food preferences and taste influence

consumers’ use and process of the provided nutritional information should be

investigated.

Next, the questions used to measure objective knowledge were selected from

existing questionnaires, which were related to the provided six nutrient contents.

Although the questions were tested through two pilot tests, they did not provide a

standardized instrument for measuring the nutritional knowledge level. There were 7

questions with rates of the correct answer between 50% and 80%, and more than 80% of

respondents answered correctly for other 2 questions (See Table E.1 in Appendix E).

Despite of the moderate level of nutritional knowledge questions, the mean score of

objective nutritional knowledge was 8.17, which indicates the average rate of correct

answers is just 54%. Thus, in this study, the objective nutritional knowledge measures in

particular may be of concern. A scale to measure nutritional knowledge efficiently and

accurately should be developed.

A final issue that can be considered in a future study is the relationship among

influential factors on consumer food decision. In this study, how taste and diet concerns

influenced food choice was investigated. However, consumer food choice at a restaurant

could be affected by other important factors such as food quality, taste, menu variety,

price, convenience, and ambience (Lee & Cranage, 2007; Neumark-Sztainer, Story, Perry,

& Casey, 1999; Yoon, et al., 2009). Similar to the results obtained though the logistic

regression in this study, the taste of food, which was considered as the most important

factor on food choice, was closely associated with health. According to Raghunathan, et

114
al. (2006), consumers would choose a tasty food rather than a healthy food if they have to

make a choice between them. Additionally, other individual characteristics, besides of

motivation to process information and nutritional knowledge, should influence the use of

nutritional information and consumer food choice in restaurants. As shown in the logistic

regression result, for instance, consumers’ diet concern was a significant predictor of food

choice. Thus, how the disclosure of nutritional information is associated with other

influential factors and what other factors can influence the use of nutritional information

should be investigated in future study.

115
 References

Almanza, B. A., Nelson, D., & Chai, S. (1997). Obstacles to nutrition labeling in
restaurants. Journal of the American Dietetic Association, 97(2), 157-161.

Andrews, J. C., Burton, S., & Netemeyer, R. G. (2000). Are Some Comparative Nutrition
Claims Misleading? The Role of Nutrition Knowledge, Ad Claim Type and
Disclosure Conditions. [Article]. Journal of Advertising, 29(3), 29-42.

Andrews, J. C., Netemeyer, R. G., & Burton, S. (1998). Consumer generalization of

nutrient content claims in advertising. The Journal of Marketing, 62(4), 62-75.

Axelson, M. L., & Penfield, M. P. (1983). Factors associated with food expenditures of
elderly persons living alone. Family and Consumer Sciences Research Journal,
12(2), 228-236.

Bagozzi, R. P., Gurhan-Canli, Z., & Priester, J. R. (2002). The social psychology of
consumer behaviour: Open University Press Buckingham, England.

Balasubramanian, S. K., & Cole, C. (2002). Consumers' search and use of nutrition
information: the challenge and promise of the nutrition labeling and education act.
The Journal of Marketing, 66(3), 112-127.

Bettman, J. R., Johnson, E. J., & Payne, J. W. (1991). Consumer decision making. In T. S.
Robertson & H. H. Kassarjian (Eds.), Handbook of Consumer Behavior (pp. 50-
84): Prentice-Hall.

Binkley, J. K., Eales, J., & Jekanowski, M. (2000). The relation between dietary change
and rising US obesity. International Journal of Obesity, 24(8), 1032-1039.

Bowman, S. A., & Vinyard, B. T. (2004). Fast food consumption of US adults: impact on
energy and nutrient intakes and overweight status. Journal of the American
College of Nutrition, 23(2), 163-168.

Burton, S., & Andrews, J. C. (1996). Age, product nutrition, and label format effects on
consumer perceptions and product evaluations. Journal of Consumer Affairs,
30(1), 68-89.

116
Burton, S., & Creyer, E. H. (2004). What consumers don't know can hurt them: consumer
evaluations and disease risk perceptions of restaurant menu items. Journal of
Consumer Affairs, 38(1), 121-146.

Burton, S., Creyer, E. H., Kees, J., & Huggins, K. (2006). Attacking the obesity epidemic:
the potential health benefits of providing nutrition information in restaurants.
American Journal of Public Health, 96(9), 1669-1675.

Burton, S., Garretson, J. A., & Velliquette, A. M. (1999). Implications of accurate usage
of nutrition facts panel information for food product evaluations and purchase
intentions. Journal of the Academy of Marketing Science, 27(4), 470-480.

Byrd-Bredbenner, C. (2000). The ability of college women aged 17 to 25 to perform

tasks using nutrition facts labels. International Electronic Journal of Health
Education, 3(2), 97-106.

CDC (2007a). Economic Consequences in Overweight and Obesity, Retrieved May 28,
2007, Center for Disease and Control Prevention,
from http://www.cdc.gov/obesity/causes/economics.html

CDC (2007b). Health Consequences in Overweight and Obesity, Retrieved May 28, 2007,
Centers for Disease Control and Prevention,
from http://www.cdc.gov/nccdphp/dnpa/obesity/consequences.htm

CDC (2009). About BMI for Adults, Retrieved July 27, 2009, Centers for Disease Control
and Prevention,
from http://www.cdc.gov/healthyweight/assessing/bmi/adult_bmi/index.html

Chandon, P., & Wansink, B. (2007a). The Biasing Health Halos of Fast-Food Restaurant
Health Claims: Lower Calorie Estimates and Higher Side-Dish Consumption
Intentions. Journal of Consumer Research, 34(3), 301-314.

Chandon, P., & Wansink, B. (2007b). Is obesity caused by calorie underestimation? A

psychophysical model of meal size estimation. Journal of Marketing Research,
44(1), 84-99.

Chou, S. Y., Grossman, M., & Saffer, H. (2004). An economic analysis of adult obesity:
results from the Behavioral Risk Factor Surveillance System. Journal of Health
Economics, 23(3), 565-587.

Conlin, R. (July 27, 2007). Making it work, Seattle City Council

from www.cityofseattle.net/council/Conlin/miw/0707miw.htm

117
Cranage, D. A., Conklin, M. T., & Lambert, C. U. (2004). Effect of Nutrition Information
in Perceptions of Food Quality, Consumption Behavior and Purchase Intentions.
Journal of Foodservice Business Research, 7, 43-61.

CSPI (2008). Nutrition Labeling at Fast-food and Other Chain Restaurants, Center for
Science in the Public Interest. ,
from http://www.cspinet.org/nutritionpolicy/Nutrition_Labeling_Fast_Food.pdf

Cutler, D. M., Glaeser, E. L., & Shapiro, J. M. (2003). Why have Americans become
more obese? Journal of Economic Perspectives, 17(3), 93-118.

Dean, A., & Voss, D. (1999). Design and Analysis of Experiments: Springer.

Dhar, R., Nowlis, S. M., & Sherman, S. J. (1999). Comparison effects on preference
construction. Journal of Consumer Research, 26(3), 293-306.

Dhar, R., & Simonson, I. (1992). The effect of the focus of comparison on consumer
preferences. Journal of Marketing Research, 29(4), 430-440.

. Dietary Guidelines for Americans (2005). from

<http://www.fda.gov/ohrms/dockets/DOCKETS/06q0091/06q-0091-pav0001-19-
Ref-18-Dietary-Guidelines-for-Americans-.pdf>.

Drichoutis, A. C., Lazaridis, P., & Nayga, R. M. (2005). Nutrition knowledge and
consumer use of nutritional food labels. European Review of Agricultural
Economics, 32(1), 93-118.

Droms, C. (2006). When I Go Out to Eat I Want to Enjoy Myself: An Investigation into
Consumers' Use of Nutrition Information. Advances in Consumer Research, 33,
282.

Ebbin, R. (2000). Americans Dining-out habits. Restaurants USA, 20(9), 38-42.

Edwards, J. S., & Meiselman, H. L. (2005). The influence of positive and negative cues
on restaurant food choice and food acceptance. International Journal of
Contemporary Hospitality Management, 17(4), 332-344.

FDA (1994). Guide to Nutrition Labeling and Education Act (NLEA) Requirements,
Retrieved April 30, 2009 U.S. Food and Drug Administration,
from http://www.fda.gov/ICECI/Inspections/InspectionGuides/ucm074948.htm

FDA (1999). Definition of "Healthy Food", Medical Dictionary,

from http://www.cspinet.org/menulabeling/resources.html

118
Finkelstein, E., French, S., Variyam, J. N., & Haines, P. S. (2004). Pros and cons of
proposed interventions to promote healthy eating. American Journal of Preventive
Medicine, 27(3S), 163-171.

Ford, G. T., Hastak, M., Mitra, A., & Ringold, D. J. (1996). Can consumers interpret
nutrition information in the presence of a health claim? A laboratory investigation.
Journal of Public Policy & Marketing, 15(1), 16-27.

Frazier, P. A., Tix, A. P., & Barron, K. E. (2004). Testing Moderator and Mediator
Effects in Counseling Psychology. Journal of Counseling Psychology, 51(1), 115-
134.

French, S. A., Story, M., Neumark-Sztainer, D., Fulkerson, J. A., & Hannan, P. (2001).
Fast food restaurant use among adolescents: associations with nutrient intake,
food choices and behavioral and psychosocial variables. International Journal of
Obesity, 25(12), 1823-1833.

Garretson, J. A., & Burton, S. (2000). Effects of nutrition facts panel values, nutrition
claims, and health claims on consumer attitudes, perceptions of disease-related
risks, and trust. Journal of Public Policy & Marketing, 19(2), 213-227.

Glanz, K., Sallis, J. F., Saelens, B. E., & Frank, L. D. (2005). Healthy nutrition
environments: concepts and measures. American Journal of Health Promotion,
19(5), 330-333.

Guthrie, J. F., Fox, J. J., Cleveland, L. E., & Welsh, S. (1995). Who uses nutrition
labeling, and what effects does label use have on diet quality? Journal of
Nutrition Education, 27(4), 163-163.

Haugtvedt, C. P., & Petty, R. E. (1992). Personality and persuasion: Need for cognition
moderates the persistence and resistance of attitude changes. Journal of
Personality and Social Psychology, 63(2), 308-319.

Hill, J. O., & Peters, J. C. (1998). Environmental contributions to the obesity epidemic.
Science, 280(5368), 1371-1374.

Hill, J. O., Wyatt, H. R., Reed, G. W., & Peters, J. C. (2003). Obesity and the
environment: where do we go from here? , 299(5608), 853-855.

Hochradel, R. F. (2007). The Effect of Nutrition Information on Menu Selection When

Eating Food Away From Home. Old Dominion University.

HSUS (2006). "Farm Animal Statistics: Meat Consumption" The Human Society of the
United States. Retrieved November 30, 2006,
from http://www.hsus.org/farm/resources/pubs/stats_meat_consumption.html

119
Hu, F. B., Li, T. Y., Colditz, G. A., Willett, W. C., & Manson, J. E. (2003). Television
watching and other sedentary behaviors in relation to risk of obesity and type 2
diabetes mellitus in women. The Journal of American Medical Association,
289(14), 1785-1791.

Huber, J., Payne, J. W., & Puto, C. (1982). Adding asymmetrically dominated
alternatives: Violations of regularity and the similarity hypothesis. Journal of
Consumer Research, 9(1), 90-98.

Huber, J., & Puto, C. (1983). Market boundaries and product choice: Illustrating
attraction and substitution effects. Journal of Consumer Research, 10(1), 31-44.

Hwang, J., & Lorenzen, C. L. (2008). Effective nutrition labeling of restaurant menu and
pricing of healthy menu. Journal of Foodservice, 19(5), 270-276.

Keller, S. B., Landry, M., Olson, J., Velliquette, A. M., Burton, S., & Andrews, J. C.
(1997). The effects of nutrition package claims, nutrition facts panels, and
motivation to process nutrition information on consumer product evaluations.
Journal of Public Policy & Marketing, 16(2), 256-269.

Kemp, E., Burton, S., Creyer, E. H., & Suter, T. (2007). When Do Nutrient Content and
Nutrient Content Claims Matter? Assessing Consumer Tradeoffs Between
Carbohydrates and Fat. Journal of Consumer Affairs, 41(1), 47-73.

Kim, S. (2008, October 29). Calories Do Count. New York Times,

Kozup, J. C., Burton, S., & Creyer, E. H. (2006). The provision of trans fat information
and its interaction with consumer knowledge. Journal of Consumer Affairs, 40(1),
163-176.

Kozup, J. C., Creyer, E. H., & Burton, S. (2003). Making healthful food choices: The
influence of health claims and nutrition information on consumers' evaluations of
packaged food products and restaurant menu items. Journal of Marketing, 67(2),
19-34.

Lee, S. J., & Cranage, D. A. (2007). The Relative Importance of Menu Attributes at Point
of Menu Selection Through Conjoint Analysis: Focused on Adolescents. Journal
of Foodservice Business Research, 10(2), 3-18.

Levy, A. S., & Fein, S. B. (1998). Consumers' ability to perform tasks using nutrition
labels. Journal of Nutrition Education, 30(4), 210-217.

Lynch, J. G., Chakravarti, D., & Mitra, A. (1991). Contrast effects in consumer
judgments: changes in mental representations or in the anchoring of rating scales?
Journal of Consumer Research, 18(3), 284-297.

120
Marietta, A. B., Welshimer, K. J., & Anderson, S. L. (1999). Knowledge, Attitudes, and
Behaviors of College Students regarding the 1990 Nutrition Labeling Education
Act Food labels. Journal of American Dietetic Association, 99(4), 445-449.

Moorman, C. (1990). The effects of stimulus and consumer characteristics on the

utilization of nutrition information. Journal of Consumer Research, 17(3), 362-
374.

Moorman, C. (1993). A Model of Consumers' Preventive Health Behaviors: The Role of

Health Motivation and Health Ability. Journal of Cosumer Research 20(2), 208-
228.

Moorman, C., Diehl, K., Brinberg, D., & Kidwell, B. (2004). Subjective knowledge,
search locations, and consumer choice. Journal of Consumer Research, 31(3),
673-680.

Moorman, C., & Matulich, E. (1993). A model of consumers' preventive health behaviors:
The role of health motivation and health ability. Journal of Consumer Research,
20(2), 208-228.

Nayga, R. M., Lipinski, D., & Savur, N. (1998). Consumers' use of nutritional labels
while food shopping and at home. Journal of Consumer Affairs, 32(1), 106-120.

NCDA&CS. "Kids World Nutrition" North Carolina Department Agriculture and

Consumer Service,
from http://www.agr.state.nc.us/cyber/kidswrld/nutrition/NutritionQuiz.html

Neter, J., Kutner, M. H., Nachtsheim, C. J., & Wasserman, W. (1996). Applied Linear
Regression Models (2 ed.): McGraw-Hill/Irwin.

Neumark-Sztainer, D., Story, M., Perry, C., & Casey, M. A. (1999). Factors Influencing
Food Choices of Adolescents Findings from Focus-Group Discussions with
Adolescents. Journal of the American Dietetic Association, 99(8), 929-937.

NRA (2008). Philadelphia gives menu labeling a twist, Retrieved November 6, 2008,
Nation's Restaurant News,
from http://www.nrn.com/breakingNews.aspx?id=359940&menu_id=1368

Palmer, J., & Leontos, C. (1995). Nutrition training for chefs: Taste as an essential
determinant of choice. Journal of American Dietetic Association, 95(12), 1418-
1421.

Petty, R. E., & Brock, T. C. (1981). Thought disruption and persuasion: Assessing the
validity of attitude change experiments. Cognitive responses in persuasion.
Hillsdale, NJ: Erlbaum, 55-79.

121
Petty, R. E., & Cacioppo, J. T. (1981). Attitude and Persuation: classic and contemprary
approaches: Dubuque, IA: William C. Brown.

Petty, R. E., & Cacioppo, J. T. (1984). The effects of involvement on response to

argument quantity and quality: Central and peripheral routes to persuasion.
Journal of Personality and Social Psychology, 46(1), 69-81.

Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion.
Advances in Experimental Social Psychology, 19, 123-205.

Petty, R. E., Cacioppo, J. T., & Schumann, D. (1983). Central and peripheral routes to
advertising effectiveness: The moderating role of involvement. Journal of
Consumer Research, 10(2), 135-146.

Petty, R. E., Haugtvedt, C. P., & Smith, S. M. (1995). Elaboration as a determinant of

attitude strength: Creating attitudes that are persistent, resistant, and predictive of
behavior. In R. E. Petty & J. A. Krosnick (Eds.), Attitude strength: Antecedents
and consequences (Vol. 4, pp. 93-130): Lawrence Erlbaum Associates.

Petty, R. E., & Wegener, D. T. (1999). The elaboration likelihood model: Current status
and controversies. In S. Chaiken & Y. Trope (Eds.), Dual-process theories in
social psychology (pp. 41-72): Guilford Press.

Philipson, T. J., & Posner, R. A. (2003). The long-run growth in obesity as a function of
technological change. Perspectives in Biology and Medicine, 46(3), 87-107.

Raghunathan, R., Naylor, R., & Hoyer, W. (2006). The unhealthy= tasty intuition and its
effects on taste inferences, enjoyment, and choice of food products. Journal of
Marketing, 70(4), 170-184.

Saegert, J., & Young, E. A. (1983). Nutrition knowledge and health food consumption.
Nutrition and Behavior, 1, 103-113.

Simonson, I. (1989). Choice based on reasons: The case of attraction and compromise
effects. Journal of Consumer Research, 16(2), 158-174.

Simonson, I., & Tversky, A. (1992). Choice in context: Tradeoff contrast and
extremeness aversion. Journal of Marketing Research, 29(3), 281-295.

Stubenitsky, K., Aaron, J. I., Catt, S. L., & Mela, D. J. (2007). The influence of recipe
modification and nutritional information on restaurant food acceptance and
macronutrient intakes. Public Health Nutrition, 3(02), 201-209.

122
USDA (2008). "Food CPI, Prices and Expenditures: Food Expenditures Table", U.S.
Department of Agriculture. Retrieved September 3, 2008,
from http://www.ers.usda.gov/Briefing/CPIFoodAndExpenditures/data/

Wansink, B., & Chandon, P. (2006). Can HLow-Fat?Nutrition Labels Lead to Obesity?
Journal of Marketing Research, 43(4), 605-617.

Wardle, J., Parmenter, K., & Waller, J. (2000). Nutrition knowledge and food intake.
Appetite, 34(3), 269-275.

Wikipedia (2009). Nutritional Information,

from http://en.wikipedia.org/wiki/Nutritional_information

Wolf, A. M., & Colditz, G. A. (1998). Current estimates of the economic cost of obesity
in the United States. Obesity Research, 6(2), 97-106.

Wootan, M. G. (2003). Anyone's Guess: Center for Science in the Public Interest.

Yoon, H., Sydny-Busso, S., & Parsa, H. G. (2009). Bayesian approach to assess
consumers’ brand selection process and identification of brand attributes in a
service context. International Journcal of Hospitality Management, 28(1), 33-41.

Yung, L., Contento, I., & Gussow, J. D. (1984). Use of health foods by the elderly.
Journal of nutrition education (USA), 16(3), 127-131.

Zoumas-Morse, C., Rock, C. L., Sobo, E. J., & Neuhouser, M. L. (2001). Children'
patterns of macronutrient intake and associations with restaurant and home eating.
Journal of the American Dietetic Association, 101(8), 923-925.

123
Appendix A: Questionnaire of Pre-test I

124
Pilot Test I

We are seeking your reactions to a situation that you might encounter

concerning meal selection in a restaurant. Please respond to the questions in a
natural manner.
There are no right or wrong answers; We want your honest opinions.

Please carefully read all information presented on the menu. Later we will ask
you to provide your reaction to the information.

Participation is voluntary. If you decide not to participate, there will be no

penalty or loss of benefits. You can stop participating at any time, without any
penalty. You will still receive extra credits if you stop.

If you have any additional questions concerning this research or your

participation in it, please feel free to contact me or my advisor at any time.

Hae Jin Yoon Dr. Thomas George

Graduate Student Associate Professor
Department of Consumer Sciences Department of Consumer Sciences
The Ohio State University The Ohio State University
Phone: (614) 537-8779 Phone: (614) 292-6219
Email: [email protected] Email: [email protected]
0

125
Imagine you are about to order an entrée from the following menu at a casual dining
restaurant.

• Chicken Caesar Salad

Grilled chicken served on top of romaine lettuce in a creamy and
garlicky Caesar dressing topped with parmesan cheese and
croutons
• Classic Sirloin
Sirloin is justifiably America's favorite steak. A generous 10–oz.
cut, expertly seasoned and fire–grilled just the way you like it
• Lemon-Herb Roasted Chicken
Roasted half–chicken seasoned with a Light Lemon Garlic-Herb
Sauce.
• Chicken Fajitas
Juicy chicken served on a skillet that sizzles like the Southwest
sun. Served with southwest rice, guacamole, sour cream, pico de
gallo, caramelized onions, green peppers and Jack-cheddar
cheese.
• Chicken Marsala
Sauteed chicken breasts in a savory sauce of mushrooms, garlic
and marsala wine.
• Grilled Salmon
An herb seasoned filet of Coho Salmon that is fire-grilled served
over rice pilaf comes with broccoli, tartar sauce, and a lemon
wedge, for a light and delicious seafood feast.
• Grilled Pork Chop
A generous 10–oz. cut, seasoned and fire–grilled pork chop.
• California Turkey Club
Slices of roasted turkey piled high on marble wheat bread and
topped with Jack cheese, lettuce, sliced tomatoes, smoked bacon
and creamy avocado-ranch
* All menu items will be served in a dinner portion size.

126
1. Think about your preferences. Please select three items you like the most from the
menu above.

2. Of the items you listed in Question 1, which would you most prefer?

The following questions are general questions about nutritional knowledge. Please
answer based on your knowledge. If you don’t know the answer, please do not guess.
Circle the “Don’t know” option.

1. Saturated fats are usually found in ________.

a. Vegetable and vegetable oils b. Animal products like meat and dairy
c. Grain products such as bread and d. None of the above
cereal
e. Don’t know

2. Which kind of fat is more likely to raise the blood cholesterol level?
a. Saturated fats b. Polyunsaturated fats
c. Both of them d. None of the above
e. Don’t know

3. Risk of high blood pressure is most likely to be reduced by eating a diet with U .
a. Less sugar b. More fiber
c. More iron d. Less salt
e. Don’t know

4. Which food group provides protein, B vitamins, iron, and zinc?

a. Meat, poultry and fish b. Milk and daily products
c. Fruits d. Grain products such as bread and cereal
e. Don’t know

5. Cholesterol is found in _______.

a. Vegetables and vegetables oils b. All foods containing fat or oil
c. Animal products like meat and dairy d. None of above
e. Don’t know

127
6. If you eat 2000 calories a day, your daily sodium intake should be less than how many
milligrams?
a. 500mg b. 2400mg
c. 4300mg d. 6000mg
e. Don’t know

7. Which food group is our body's best source of energy?

a. Meat Group b. Fats, oils and sweets
c. Breads and cereals d. Milk and cheese
e. Don’t know

8. Which of the following is not considered a nutrient?

a. Vitamins b. Minerals
c. Fiber d. Fats
e. Don’t know

9. Which of the following is added to a food label because people sometimes don’t eat
enough of this?
a. Fat b. Calcium
c. Sodium d. Cholesterol
e. Don’t know

10. The bread, cereal, rice and pasta group is a good source of .
a. Carbohydrate b. Vitamin C
c. Calcium d. Vitamin D
e. Don’t know

11. Which of the following is required on a food label?

a. Total carbohydrate b. Sugars
c. Iron d. All of the above
e. Don’t know

12. Citrus fruits are an excellent source of .

a. Calcium b. Vitamin C
c. Vitamin B d. Calories
e. Don’t know

13. Foods such as the meat, poultry, fish, dry beans, eggs and nuts are an important
source of .
a. Iron b. Fiber
c. Beta carotene d. Calcium
e. Don’t know

128
14. In general, which food contains the most fat?
a. Graham crackers b. Brownies
c. Pudding d. Angel food cake
e. Don’t know

15. The milk, cheese and yogurt group are important for .
a. Strong bones b. Teeth
c. Muscles d. All of the above
e. Don’t know

16. How many servings of vegetables do we need each day?

a. 1-2 b. 2-3
c. 3-5 d. 6-11
e. Don’t know

How important is each of the following when you choose a menu item in a restaurant?

Not important at all Very Important

Calories from fat 1--------------2--------------3--------------4--------------5
Carbohydrates 1--------------2--------------3--------------4--------------5
Cholesterol 1--------------2--------------3--------------4--------------5
Dietary fiber 1--------------2--------------3--------------4--------------5
Protein 1--------------2--------------3--------------4--------------5
Saturated fat 1--------------2--------------3--------------4--------------5
Sodium 1--------------2--------------3--------------4--------------5
Sugar 1--------------2--------------3--------------4--------------5
Total calories 1--------------2--------------3--------------4--------------5
Total fat 1--------------2--------------3--------------4--------------5
Trans fat 1--------------2--------------3--------------4--------------5

129
Consider the following menu item and answer the questions;

• Chicken Caesar Salad

Grilled chicken served on top of romaine lettuce in a creamy and garlicky Caesar
dressing topped with parmesan cheese and croutons
930 Calories 71g Total Fat 13g Saturated Fat
28g Carbohydrates 43g Protein 6g Dietary Fiber 1840mg Sodium
1. Based on the information, how important would this menu item be as part of a
healthy diet?
Not Important at all 1---------2---------3---------4---------5 Very Important
2. How would you rate the level of nutrition suggested by the information of this menu
item?
Not Nutritious at all 1---------2---------3---------4---------5 Very Nutritious
3. How would you evaluate the healthiness of this menu item?
Very Unhealthy 1---------2---------3---------4---------5 Very Healthy
4. To what extent do you consider the nutritional information of this menu item to be
accurate?
Very Inaccurate 1---------2---------3---------4---------5 Very Accurate
5. To what extent do you consider the nutritional information of this menu item to be
credible?
Very Unreliable 1---------2---------3---------4---------5 Very Reliable

Consider the following menu item and answer the questions;

• Classic Sirloin Steak

Sirloin is justifiably America's favorite steak. A generous 10–oz. cut, expertly
seasoned and fire–grilled just the way you like it
690 Calories 44g Total Fat 18g Saturated Fat
4g Carbohydrates 58g Protein 0g Dietary Fiber 430mg Sodium
1. Based on the information, how important would this menu item be as part of a
healthy diet?
Not Important at all 1---------2---------3---------4---------5 Very Important
2. How would you rate the level of nutrition suggested by the information of this menu
item?
Not Nutritious at all 1---------2---------3---------4---------5 Very Nutritious
3. How would you evaluate the healthiness of this menu item?
Very Unhealthy 1---------2---------3---------4---------5 Very Healthy
4. To what extent do you consider the nutritional information of this menu item to be
accurate?
Very Inaccurate 1---------2---------3---------4---------5 Very Accurate
5. To what extent do you consider the nutritional information of this menu item to be
credible?
Very Unreliable 1---------2---------3---------4---------5 Very Reliable

130
Consider the following menu item and answer the questions;

• Lemon-herb Roasted Chicken

Roasted half–chicken seasoned with a Light Lemon Garlic-Herb Sauce.
410 Calories 10g Total Fat 2g Saturated Fat
50g Carbohydrates 34g Protein 9g Dietary Fiber 300mg Sodium
1. Based on the information, how important would this menu item be as part of a
healthy diet?
Not Important at all 1---------2---------3---------4---------5 Very Important
2. How would you rate the level of nutrition suggested by the information of this menu
item?
Not Nutritious at all 1---------2---------3---------4---------5 Very Nutritious
3. How would you evaluate the healthiness of this menu item?
Very Unhealthy 1---------2---------3---------4---------5 Very Healthy
4. To what extent do you consider the nutritional information of this menu item to be
accurate?
Very Inaccurate 1---------2---------3---------4---------5 Very Accurate
5. To what extent do you consider the nutritional information of this menu item to be
credible?
Very Unreliable 1---------2---------3---------4---------5 Very Reliable

Consider the following menu item and answer the questions;

• Chicken Fajitas
Juicy chicken served on a skillet that sizzles like the Southwest sun. Served with
southwest rice, guacamole, sour cream, pico de gallo, caramelized onions, green
peppers and Jack-cheddar cheese.
851 Calories 29g Total Fat 6g Saturated Fat
90g Carbohydrates 56g Protein 6g Dietary Fiber 2002mg Sodium
1. Based on the information, how important would this menu item be as part of a
healthy diet?
Not Important at all 1---------2---------3---------4---------5 Very Important
2. How would you rate the level of nutrition suggested by the information of this menu
item?
Not Nutritious at all 1---------2---------3---------4---------5 Very Nutritious
3. How would you evaluate the healthiness of this menu item?
Very Unhealthy 1---------2---------3---------4---------5 Very Healthy
4. To what extent do you consider the nutritional information of this menu item to be
accurate?
Very Inaccurate 1---------2---------3---------4---------5 Very Accurate
5. To what extent do you consider the nutritional information of this menu item to be
credible?
Very Unreliable 1---------2---------3---------4---------5 Very Reliable

131
Consider the following menu item and answer the questions;

• Grilled Pork Chop

A generous 10–oz. cut, seasoned and fire–grilled pork chop.
310 Calories 12g Total Fat 6g Saturated Fat
0g Carbohydrates 0g Protein 0g Dietary Fiber 75mg Sodium
1. Based on the information, how important would this menu item be as part of a
healthy diet?
Not Important at all 1---------2---------3---------4---------5 Very Important
2. How would you rate the level of nutrition suggested by the information of this menu
item?
Not Nutritious at all 1---------2---------3---------4---------5 Very Nutritious
3. How would you evaluate the healthiness of this menu item?
Very Unhealthy 1---------2---------3---------4---------5 Very Healthy
4. To what extent do you consider the nutritional information of this menu item to be
accurate?
Very Inaccurate 1---------2---------3---------4---------5 Very Accurate
5. To what extent do you consider the nutritional information of this menu item to be
credible?
Very Unreliable 1---------2---------3---------4---------5 Very Reliable

Consider the following menu item and answer the questions;

• Grilled Salmon
An herb seasoned filet of Coho Salmon that is fire-grilled served over rice pilaf
comes with Broccoli, tartar sauce, and a lemon wedge, for a light and delicious
seafood feast.
393 Calories 19g Total Fat 4.5g Saturated Fat
14g Carbohydrates 44g Protein 7g Dietary Fiber 520mg Sodium
1. Based on the information, how important would this menu item be as part of a
healthy diet?
Not Important at all 1---------2---------3---------4---------5 Very Important
2. How would you rate the level of nutrition suggested by the information of this menu
item?
Not Nutritious at all 1---------2---------3---------4---------5 Very Nutritious
3. How would you evaluate the healthiness of this menu item?
Very Unhealthy 1---------2---------3---------4---------5 Very Healthy
4. To what extent do you consider the nutritional information of this menu item to be
accurate?
Very Inaccurate 1---------2---------3---------4---------5 Very Accurate
5. To what extent do you consider the nutritional information of this menu item to be
credible?
Very Unreliable 1---------2---------3---------4---------5 Very Reliable
132
Consider the following menu item and answer the questions;

• California Turkey Club

Slices of roasted turkey piled high on marble wheat bread and topped with Jack
cheese, lettuce, sliced tomatoes, smoked bacon and creamy avocado-ranch
510 Calories 20.5g Total Fat 3.5g Saturated Fat
49g Carbohydrates 30.7g Protein 3.6g Dietary Fiber 1734mg Sodium
1. Based on the information, how important would this menu item be as part of a
healthy diet?
Not Important at all 1---------2---------3---------4---------5 Very Important
2. How would you rate the level of nutrition suggested by the information of this menu
item?
Not Nutritious at all 1---------2---------3---------4---------5 Very Nutritious
3. How would you evaluate the healthiness of this menu item?
Very Unhealthy 1---------2---------3---------4---------5 Very Healthy
4. To what extent do you consider the nutritional information of this menu item to be
accurate?
Very Inaccurate 1---------2---------3---------4---------5 Very Accurate
5. To what extent do you consider the nutritional information of this menu item to be
credible?
Very Unreliable 1---------2---------3---------4---------5 Very Reliable

Consider the following menu item and answer the questions;

• Classic Marsala
Sautéed chicken breasts in a savory sauce of mushrooms, garlic and marsala wine
973 Calories 57g Total Fat 0g Saturated Fat
65g Carbohydrates 37g Protein 9g Dietary Fiber 1399mg Sodium
1. Based on the information, how important would this menu item be as part of a
healthy diet?
Not Important at all 1---------2---------3---------4---------5 Very Important
2. How would you rate the level of nutrition suggested by the information of this menu
item?
Not Nutritious at all 1---------2---------3---------4---------5 Very Nutritious
3. How would you evaluate the healthiness of this menu item?
Very Unhealthy 1---------2---------3---------4---------5 Very Healthy
4. To what extent do you consider the nutritional information of this menu item to be
accurate?
Very Inaccurate 1---------2---------3---------4---------5 Very Accurate
5. To what extent do you consider the nutritional information of this menu item to be
credible?
Very Unreliable 1---------2---------3---------4---------5 Very Reliable

133
You have received and evaluated the nutritional information of all 8 menu items.
Do you want to change your choice of menu item that you selected at the beginning of
the survey?
a. No b. Yes

If yes, which alternative menu item do you want to choose?

Please rank the following from 1 to 11 by the order of importance when you choose a
menu item in a restaurant. (1 = Most Important; 11 = Least Important)
Calories from fat ( )
Carbohydrates ( )
Cholesterol ( )
Dietary Fiber ( )
Protein ( )
Saturated fat ( )
Sodium ( )
Sugar ( )
Total Calories ( )
Total fat ( )
Trans fat ( )

The following questions are general questions about you.

1. Please indicate your gender.

a. Male b. Female

2. Please indicate your age.

a. 18-20 years old b. 21-23 years old c. 24-26 years old
d. 27-30 years old e. 31-35 years old f. More than 35 years old

3. Please indicate which of the following best describe your status.

a. Freshman b. Sophomore c. Junior
d. Senior e. Graduate f. Faculty or Staff

4. Please indicate your ethnicity/nationality.

a. Caucasian b. African American c. Hispanic
d. Asian e. Native American f. Other ( )

5. Please indicate your monthly income.

a. Under $1,000 b. $1,001 - $2,000 c. $2,001 - $3,000
d. $3,001 - $5,000 e. $5,001 - $7,000 f. Over $7,000

6. Please indicate your monthly expenditure on food away from home.

a. Under $100 b. $100 - $200 c. $201 - $300
d. $301 - $400 e. $401 - $500 f. Over $500

134
Appendix B: Questionnaire of Pre-test II

135
Pilot Test II.
Type3: Nutritional Information of 6 nutrition contents

We are seeking your reactions to a situation that you might encounter concerning
meal selection in a restaurant. Please respond to the questions in a natural manner.
There are no right or wrong answers; We want your honest opinions.

Please carefully read all information presented on the menu. Later we will ask
you to provide your reaction to the information.

Participation is voluntary. If you decide not to participate, there will be no

penalty or loss of benefits. You can stop participating at any time, without any
penalty. You will still receive extra credits if you stop.

If you have any additional questions concerning this research or your

participation in it, please feel free to contact me or my advisor at any time.

Hae Jin Yoon Dr. Thomas George

Graduate Student Associate Professor
Department of Consumer Sciences Department of Consumer Sciences
The Ohio State University The Ohio State University
Phone: (614) 537-8779 Phone: (614) 292-6219
Email: [email protected] Email: [email protected]

Imagine you are about to order an entrée from the following menu at a
casual dining restaurant.

136
Section I: Menu Evaluation

The following questions are general questions about Chicken Caesar Salad on the menu.
Please answer what is being asked based on your opinion.
• Chicken Caesar Salad
Grilled chicken served on top of romaine lettuce in a creamy and
garlicky Caesar dressing topped with parmesan cheese and
croutons.
900 Calories 43g Total fat (66% DV) 13g Saturated Fat (65% DV)
201mg Cholesterol (67% DV) 1560mg Sodium (65% DV) 13g Protein (26% DV)
Percent Daily Values (DV) are based on a 2,000 calorie diet.
1. Based on the information shown, to what extent do you expect the taste of Chicken
Caesar Salad?
Very Poor 1----------2----------3----------4----------5 Very Good
2. Based on the information provided, how important would Chicken Caesar Salad be
as part of a healthy diet?
Not Important at all 1----------2----------3----------4----------5 Very Important
3. How would you rate the level of nutrition suggested by the information of Chicken
Caesar Salad?
Not Nutritious at all 1----------2----------3----------4----------5 Very Nutritious
4. How would you evaluate the healthiness of Chicken Caesar Salad?
Very Unhealthy 1----------2----------3----------4----------5 Very Healthy
5. Overall, how much are you favorable toward Chicken Caesar Salad?
Very Unfavorable 1----------2----------3----------4----------5 Very Favorable
6. Overall, how much do you prefer Chicken Caesar Salad?
Very Dislike 1----------2----------3----------4----------5 Very Like
7. What is your overall feeling toward Chicken Caesar Salad?
Very Negative 1----------2----------3----------4----------5 Very Positive
8. How probable is it that you would consider the purchase of Chicken Caesar Salad?
Not Probable at all 1----------2----------3----------4----------5 Very Probable
9. How willing would you be to purchase Chicken Caesar Salad again?
Very Unlikely 1----------2----------3----------4----------5 Very Likely
10. How likely would you be to purchase Chicken Caesar Salad based on the given
information?
Very Unlikely 1----------2----------3----------4----------5 Very Likely

137
The following questions are general questions about Lemon Herb Roasted Chicken on the
menu. Please answer what is being asked based on your opinion.

• Lemon Herb Roasted Chicken

Roasted half–chicken seasoned with a Light Lemon Garlic-Herb
Sauce.

460 Calories 8g Total fat (12% DV) 2.5g Saturated Fat (12.5% DV)
40mg Cholesterol (13% DV) 264mg Sodium (11% DV) 34g Protein (68% DV)
Percent Daily Values (DV) are based on a 2,000 calorie diet.
1. Based on the information shown, to what extent do you expect the taste of Lemon
Herb Roasted Chicken?
Very Poor 1----------2----------3----------4----------5 Very Good
2. Based on the information provided, how important would Lemon Herb Roasted
Chicken be as part of a healthy diet?
Not Important at all 1----------2----------3----------4----------5 Very Important
3. How would you rate the level of nutrition suggested by the information of Lemon
Herb Roasted Chicken?
Not Nutritious at all 1----------2----------3----------4----------5 Very Nutritious
4. How would you evaluate the healthiness of Lemon Herb Roasted Chicken?

Very Unhealthy 1----------2----------3----------4----------5 Very Healthy

5. Overall, how much are you favorable toward Lemon Herb Roasted Chicken?

Very Unfavorable 1----------2----------3----------4----------5 Very Favorable

6. Overall, how much do you prefer Lemon Herb Roasted Chicken?

Very Dislike 1----------2----------3----------4----------5 Very Like

7. What is your overall feeling toward Lemon Herb Roasted Chicken?

Very Negative 1----------2----------3----------4----------5 Very Positive

8. How probable is it that you would consider the purchase of Lemon Herb Roasted
Chicken?
Not Probable at all 1----------2----------3----------4----------5 Very Probable
9. How willing would you be to purchase Lemon Herb Roasted Chicken again?

Very Unlikely 1----------2----------3----------4----------5 Very Likely

10. How likely would you be to purchase Lemon Herb Roasted Chicken based on the
given information?
Very Unlikely 1----------2----------3----------4----------5 Very Likely

138
The following questions are general questions about Chicken Fajitas on the menu.
Please answer what is being asked based on your opinion.

• Chicken Fajitas
Juicy chicken served with southwest rice, guacamole, sour
cream, pico de gallo, caramelized onions, green peppers and
Jack-cheddar cheese.

830 Calories 38g Total fat (58% DV) 12g Saturated Fat (60% DV)
186mg Cholesterol (62% DV) 1464mg Sodium (61% DV) 15g Protein (30% DV)
Percent Daily Values (DV) are based on a 2,000 calorie diet.
1. Based on the information shown, to what extent do you expect the taste of Chicken
Fajitas?
Very Poor 1----------2----------3----------4----------5 Very Good
2. Based on the information provided, how important would Chicken Fajitas be as part
of a healthy diet?
Not Important at all 1----------2----------3----------4----------5 Very Important
3. How would you rate the level of nutrition suggested by the information of Chicken
Fajitas?
Not Nutritious at all 1----------2----------3----------4----------5 Very Nutritious
4. How would you evaluate the healthiness of Chicken Fajitas?

Very Unhealthy 1----------2----------3----------4----------5 Very Healthy

5. Overall, how much are you favorable toward Chicken Fajitas?

Very Unfavorable 1----------2----------3----------4----------5 Very Favorable

6. Overall, how much do you prefer Chicken Fajitas?

Very Dislike 1----------2----------3----------4----------5 Very Like

7. What is your overall feeling toward Chicken Fajitas?

Very Negative 1----------2----------3----------4----------5 Very Positive

8. How probable is it that you would consider the purchase of Chicken Fajitas?

Not Probable at all 1----------2----------3----------4----------5 Very Probable

9. How willing would you be to purchase Chicken Fajitas again?

Very Unlikely 1----------2----------3----------4----------5 Very Likely

10. How likely would you be to purchase Chicken Fajitas based on the given
information?
Very Unlikely 1----------2----------3----------4----------5 Very Likely

139
The following questions are general questions about Grilled Chicken on the menu.
Please answer what is being asked based on your opinion.

• Grilled Chicken
Grilled chicken breasts in an apricot citrus sauce. Served with
broccoli, asparagus and diced tomatoes.

520 Calories 12g Total fat (18% DV) 4g Saturated Fat (20% DV)
54mg Cholesterol (18% DV) 520mg Sodium (21% DV) 31g Protein (62% DV)
Percent Daily Values (DV) are based on a 2,000 calorie diet
1. Based on the information shown, to what extent do you expect the taste of Grilled
Chicken?
Very Poor 1----------2----------3----------4----------5 Very Good
2. Based on the information provided, how important would Grilled Chicken be as part
of a healthy diet?
Not Important at all 1----------2----------3----------4----------5 Very Important
3. How would you rate the level of nutrition suggested by the information of Grilled
Chicken?
Not Nutritious at all 1----------2----------3----------4----------5 Very Nutritious
4. How would you evaluate the healthiness of Grilled Chicken?

Very Unhealthy 1----------2----------3----------4----------5 Very Healthy

5. Overall, how much are you favorable toward Grilled Chicken?

Very Unfavorable 1----------2----------3----------4----------5 Very Favorable

6. Overall, how much do you prefer Grilled Chicken?

Very Dislike 1----------2----------3----------4----------5 Very Like

7. What is your overall feeling toward Grilled Chicken?

Very Negative 1----------2----------3----------4----------5 Very Positive

8. How probable is it that you would consider the purchase of Grilled Chicken?

Not Probable at all 1----------2----------3----------4----------5 Very Probable

9. How willing would you be to purchase Grilled Chicken again?

Very Unlikely 1----------2----------3----------4----------5 Very Likely

10. How likely would you be to purchase Grilled Chicken based on the given
information?
Very Unlikely 1----------2----------3----------4----------5 Very Likely

140
The following questions are general questions about Chicken Marsala on the menu.
Please answer what is being asked based on your opinion.

• Chicken Marsala
Sauteed chicken breasts in a savory sauce of mushrooms, garlic
and marsala wine.

920 Calories 46g Total fat (71% DV) 15g Saturated Fat (75% DV)
216mg Cholesterol (72% DV) 1680mg Sodium (70% DV) 10g Protein (20% DV)
Percent Daily Values (DV) are based on a 2,000 calorie diet
1. Based on the information shown, to what extent do you expect the taste of Chicken
Marsala?
Very Poor 1----------2----------3----------4----------5 Very Good
2. Based on the information provided, how important would Chicken Marsala be as
part of a healthy diet?
Not Important at all 1----------2----------3----------4----------5 Very Important
3. How would you rate the level of nutrition suggested by the information of Chicken
Marsala?
Not Nutritious at all 1----------2----------3----------4----------5 Very Nutritious
4. How would you evaluate the healthiness of Chicken Marsala?

Very Unhealthy 1----------2----------3----------4----------5 Very Healthy

5. Overall, how much are you favorable toward Chicken Marsala?
Very Unfavorable 1----------2----------3----------4----------5 Very Favorable
6. Overall, how much do you prefer Chicken Marsala?
Very Dislike 1----------2----------3----------4----------5 Very Like
7. What is your overall feeling toward Chicken Marsala?

Very Negative 1----------2----------3----------4----------5 Very Positive

8. How probable is it that you would consider the purchase of Chicken Marsala?
Not Probable at all 1----------2----------3----------4----------5 Very Probable
9. How willing would you be to purchase Chicken Marsala again?
Very Unlikely 1----------2----------3----------4----------5 Very Likely
10. How likely would you be to purchase Chicken Marsala based on the given
information?
Very Unlikely 1----------2----------3----------4----------5 Very Likely

141
The following questions are general questions about Italian Chicken Sandwich on the
menu. Please answer what is being asked based on your opinion.

• Italian Chicken Sandwich

Grilled, marinated chicken breast topped with sliced portobello
mushrooms, a slice of tomato and chunky marinara sauce.

440 Calories 6.5g Total fat (10% DV) 2g Saturated Fat (10% DV)
33mg Cholesterol (11% DV) 216mg Sodium (9% DV) 36g Protein (72% DV)
Percent Daily Values (DV) are based on a 2,000 calorie diet
1. Based on the information shown, to what extent do you expect the taste of Italian
Chicken Sandwich?
Very Poor 1----------2----------3----------4----------5 Very Good
2. Based on the information provided, how important would Italian Chicken Sandwich
be as part of a healthy diet?
Not Important at all 1----------2----------3----------4----------5 Very Important
3. How would you rate the level of nutrition suggested by the information of Italian
Chicken Sandwich?
Not Nutritious at all 1----------2----------3----------4----------5 Very Nutritious
4. How would you evaluate the healthiness of Italian Chicken Sandwich?

Very Unhealthy 1----------2----------3----------4----------5 Very Healthy

5. Overall, how much are you favorable toward Italian Chicken Sandwich?

Very Unfavorable 1----------2----------3----------4----------5 Very Favorable

6. Overall, how much do you prefer Italian Chicken Sandwich?

Very Dislike 1----------2----------3----------4----------5 Very Like

7. What is your overall feeling toward Italian Chicken Sandwich?

Very Negative 1----------2----------3----------4----------5 Very Positive

8. How probable is it that you would consider the purchase of Italian Chicken
Sandwich?
Not Probable at all 1----------2----------3----------4----------5 Very Probable
9. How willing would you be to purchase Italian Chicken Sandwich again?

Very Unlikely 1----------2----------3----------4----------5 Very Likely

10. How likely would you be to purchase Italian Chicken Sandwich based on the given
information?
Very Unlikely 1----------2----------3----------4----------5 Very Likely

142
Section II: Menu Choice

1. Based on the previous menu items, which one are you most likely to choose?

a. Chicken Caesar Salad

b. Lemon Herb Roasted
c. Chicken Fajitas
d. Grilled Chicken
e. Chicken Marsala
f. Italian Chicken Sandwich

2. Which of side items on the below menu would you like to select accompanying with
your entrée that you chose? (Please select one)

a. House Salad
(210 calories, 12g Total Fat, 6g Saturated Fat, 10g protein, 310mg Sodium without
dressing)

b. Mashed Potato
(270 calories, 14g Total Fat, 2g Saturated Fat, 5g protein, 950mg Sodium)

c. Baked Potato
(250 calories, 18g Total Fat, 11g Saturated Fat, 9g protein, 910mg Sodium)

d. Basket of Fries
(430 calories, 26g Total Fat, 4.5g Saturated Fat, 4g protein, 240mg Sodium)

e. Fresh Steamed Vegetables

(70 calories, 5g Total Fat, 1g Saturated Fat, 3g protein, 195mg Sodium)

f. French Onion Soup

(388 calories, 21g Total Fat, 0g Saturated Fat, 19g protein, 2317mg Sodium)

g. None

143
Section III: Nutritional Knowledge

The following questions are general questions about nutritional knowledge. Please
answer based on your knowledge. If you don’t know the answer, please do not guess.
Circle the “Don’t know” option.

1. Which kind of fat is higher in calories?

a. Saturated fats b. Polyunsaturated fats
c. They are both the same d. None of the above
e. Don’t know

2. Risk of high blood pressure is most likely to be reduced by eating a diet with .
a. Less sugar b. More fiber
c. More iron d. Less salt
e. Don’t know

3. Which food group provides protein, B vitamins, iron, and zinc?

a. Meat, poultry and fish b. Milk and daily products
c. Fruits d. Grain products such as bread, cereal,
and rice
e. Don’t know

4. Cholesterol is found in _______.

a. Vegetables and vegetables oils b. All foods containing fat or oil
c. Animal products like meat and dairy d. None of above
e. Don’t know

5. If you eat 2000 calories a day, your daily sodium intake should be less than how many
milligrams?
a. 500mg b. 2400mg
c. 4300mg d. 6000mg
e. Don’t know

6. Which food group is our body's best source of energy?

a. Meat Group b. Fats, oils and sweets
c. Breads and cereals d. Milk and cheese
e. Don’t know

7. Which of the following is not considered a nutrient?

a. Vitamins b. Minerals
c. Fiber d. Fats
e. Don’t know

144
8. Which of the following is added to a food label because people sometimes don’t eat
enough of this?
a. Fat b. Calcium
c. Sodium d. Cholesterol
e. Don’t know

9. The bread, cereal, rice and pasta group is a good source of .

a. Carbohydrate b. Vitamin C
c. Calcium d. Vitamin D
e. Don’t know

10. Which of the following is required on a food label?

a. Total carbohydrate b. Sugars
c. Iron d. All of the above
e. Don’t know

11. Citrus fruits are an excellent source of .

a. Calcium b. Vitamin C
c. Vitamin B d. Calories
e. Don’t know

12. Foods as the meat, poultry, fish, dry beans, eggs and nuts are an important source
of .
a. Iron b. Fiber
c. Beta carotene d. Calcium
e. Don’t know

13. In general, which food contains the most fat?

a. Graham crackers b. Brownies
c. Pudding d. Angel food cake
e. Don’t know

14. The milk, cheese and yogurt group are important for .
a. Strong bones b. Teeth
c. Muscles d. All of the above
e. Don’t know

15. How many servings of vegetables do we need each day?

a. 1-2 b. 2-3
c. 3-5 d. 6-11
e. Don’t know

145
Section IV: Individual Characteristics

The following questions are asking about your general personal characteristics.
Please answer based on your personality.

1. In general, I am interested in looking for nutritional information of menu items in a

restaurant.
Strongly Disagree 1----------2----------3----------4----------5 Strongly Agree
2. I would like to receive additional nutritional information about menu items in a
restaurant.
Strongly Disagree 1----------2----------3----------4----------5 Strongly Agree
3. In general, I intend to pay attention to nutritional information while choosing a menu
item in a restaurant.
Strongly Disagree 1----------2----------3----------4----------5 Strongly Agree
4. It is important to me that nutritional information is available on the menu.
Strongly Disagree 1----------2----------3----------4----------5 Strongly Agree
5. I usually ignore nutritional information while choosing a menu item in a restaurant
Strongly Disagree 1----------2----------3----------4----------5 Strongly Agree
6. I am quite knowledgeable about nutritional information compared to most people.
Strongly Disagree 1----------2----------3----------4----------5 Strongly Agree
7. I am confident in using nutritional information compared to most people.
Strongly Disagree 1----------2----------3----------4----------5 Strongly Agree
8. I feel confident about my ability to comprehend nutritional information on the menu.
Strongly Disagree 1----------2----------3----------4----------5 Strongly Agree
9. I will look for nutritional information of menu items in a restaurant in future.
Strongly Disagree 1----------2----------3----------4----------5 Strongly Agree
10. I will ask additional nutritional information about menu items to a server in a
restaurant in future.
Strongly Disagree 1----------2----------3----------4----------5 Strongly agree
11. I will pay more attention to nutritional information while choosing a menu item at a
restaurant in future.
Strongly Disagree 1----------2----------3----------4----------5 Strongly Agree

The following questions are asking about your health Status.

1. Overall, current my health status is .

Very Poor 1---------2---------3---------4---------5 Excellent
2. How serious have your health problems been?
Not at all serious 1---------2----------3--------4---------5 Very Serious

146
Section V: Demographic Questions

The following questions are general questions about you.

1. Please select the major information source that you use to obtain the nutritional
information of menu items offered by restaurant from the following.
(Please select only one).
a. TV b. Restaurant Website c. Other Website (ex. Yahoo, Google)
d. Newspaper e. Menu f. Ask to a server
g. Others

2. How many times do you eat at a restaurant during a week?

a. 0 b. 1-2 c. 3-4 d.5-6 e. more than 7

3. How often do you eat at a fast food restaurant during a week?

a. 0 b. 1-2 c. 3-4 d.5-6 e. more than 7

4. Please indicate your gender.

a. Male b. Female

5. Please indicate your age.

a. 18-20 years old b. 21-23 years old c. 24-26 years old
d. 27-30 years old e. 31-35 years old f. More than 35 years old

6. Please indicate which of the following best describe your status.

a. Freshman b. Sophomore c. Junior
d. Senior e. Graduate f. Faculty or Staff

7. Please indicate your ethnicity/nationality.

a. Caucasian b. African American c. Hispanic
d. Asian e. Native American f. Other ( )

8. Please indicate your monthly income.

a. Under $1,000 b. $1,001 - $2,000 c. $2,001 - $3,000
d. $3,001 - $5,000 e. $5,001 - $7,000 f. Over $7,000

9. Please indicate your monthly expenditure on food away from home.

a. Under $100 b. $100 - $200 c. $201 - $300
d. $301 - $400 e. $401 - $500 f. Over $500

Thank you very much for your participation!

147
Appendix C: Examples of Menu

148
• Menu including a Healthy Target Item in Healthy Context without
Nutritional Information

• Menu including a Healthy Target Item in Unhealthy Context without

Nutritional Information

149
• Menu including a Unhealthy Target Item in Healthy Context without
Nutritional Information

• Menu including a Unhealthy Target Item in Unhealthy Context without

Nutritional Information

150
• Menu including a Healthy Target Item in Healthy Context with Calories

• Menu including a Healthy Target Item in Unhealthy Context with Calories

151
• Menu including a Unhealthy Target Item in Healthy Context with Calories

• Menu including a Unhealthy Target Item in Unhealthy Context with Calories

152
• Menu including a Healthy Target Item in the Healthy Context with
Nutritional Information of 6 nutrition contents

• Menu including a Healthy Target Item in the Unhealthy Context with

Nutritional Information of 6 nutrition contents

153
• Menu including a Unhealthy Target Item in the Healthy Context with
Nutritional Information of 6 nutrition contents

• Menu including a Unhealthy Target Item in the Unhealthy Context with

Nutritional Information of 6 nutrition contents

154
Appendix D: Results of Additional Multiple Regression Analysis
(Dependent Variable: Source Credibility)

155
 Source Credibility
Independent Model I Model II Model III Model IV
Variables B (S.E.) B (S.E.) B (S.E.) B (S.E.)
Constant 3.63 (0.10)*** 3.63 (0.13)*** 3.60 (0.10)*** 3.60 (0.10)***
Target Item -0.03 (0.09) -0.03 (0.09) -0.01 (0.09) -0.01 (0.09)
Context -0.09 (0.09) -0.09 (0.09) -0.08 (0.09) -0.08 (0.09)
Total Calories -0.03 (0.11) -0.03 (0.11) -0.004 (0.11) 0.002 (0.11)
Nutritional Info
-0.01 (0.11) -0.01 (0.11) 0.02 (0.11) 0.02 (0.11)
of 6 contents
Objective NK
0.05 (0.02)** 0.06 (0.03)*
(CONK)
Subjective NK
0.09 (0.05)* 0.02 (0.09)
(CSNK)
Cal*CONK -0.01 (0.05)
NI*CONK -0.02 (0.05)

Cal*CSNK 0.09 (0.13)

NI*CSNK 0.13 (0.12)
Adj. R2 0.005 0.07 <0.001 <0.001
Model fit
1.29 0.94 0.86 0.77
(F-Value)
R2 Change <0.001 0.004
F Change 0.07 0.60
Note: *Significant at p<0.1; **Significant at p<0.05; ***Significant at p<0.01.
Standard errors are in the parentheses.

Table D.1: Results of Regression Analysis Result: Effect of Nutritional Knowledge on

Source Credibility

156
 Overall Food Attitude Purchase Intention
Independent Variables
B (S.E.) B (S.E.)
Constant 4.26 (0.13)*** 4.27 (0.15)***
Target Item 0.11 (0.11) 0.04 (0.13)
Context -0.46 (0.11)*** -0.58 (0.13)***
Total Calories -0.34 (0.14)** -0.41 (0.16)**
Nutritional Info
-0.43 (0.14)*** -0.49 (0.16)***
of 6 contents
Motivation (CMOT) -0.04 (0.06) -0.05 (0.07)
Objective NK (CONK) 0.003 (0.02) <0.001 (0.03)
Subjective NK
0.04 (0.07) 0.08 (0.09)
(CSNK)
CMOT*CONK 0.004 (0.02) -0.002 (0.03)
CMOT*CSNK -0.12 (0.05)** -0.13 (0.06)**
Adj. R2 0.08 0.08
Model fit (F-Value) 3.94** 4.13***
Note: *Significant at p<0.1; **Significant at p<0.05; ***Significant at p<0.01.
Standard errors are in the parentheses.

Table D.2: Results of Regression Analysis: Interaction Effect b/w Nutritional Knowledge
and Motivation to Process

157
Appendix E: Objective Nutritional Knowledge Questions

158
 “Don’t
Objective Nutritional Knowledge Question Correct
Know”
1. Which kind of fat is higher in calories?
a. Saturated fats b. Polyunsaturated fats 18 78
c. They are both the same d. None of the above (5.6%) (24.5%)
e. Don’t know
2. Risk of high blood pressure is most likely to be reduced
by eating a diet with .
188 62
a. Less sugar b. More fiber
(58.9%) (19.4%)
c. More iron d. Less salt
e. Don’t know
3.Which food group provides protein, B vitamins, iron, and
zinc?
234 41
a. Meat, poultry and fish b. Milk and daily products
(73.4%) (12.9%)
c. Fruits d. Grain products such as
e. Don’t Know bread, cereal, and rice
4. Cholesterol is found in _______.
a. Vegetables and b. All foods containing fat or
vegetables oils oil 103 52
c. Animal products like d. None of above (32.3%) (16.3%)
meat and dairy
e. Don’t know
5. If you eat 2000 calories a day, your daily sodium intake
should be less than how many milligrams?
48 225
a. 500mg b. 2400mg
(15.0%) (70.5%)
c. 4300mg d. 6000mg
e. Don’t know
6. Which food group is our body's best source of energy?
a. Meat Group b. Fats, oils and sweets 183 23
c. Breads and cereals d. Milk and cheese (57.4%) (7.2%)
e. Don’t know
7. Which of the following is not considered a nutrient?
a. Vitamins b. Minerals 82 46
c. Fiber d. Fats (25.7%) (14.4%)
e. Don’t know
Note: The correct answer is in bold.

Continued

Table E.1: Frequency of Correct Answers of Objective Nutritional Knowledge Questions

159
Table E.1 continued

8. Which of the following is added to a food label because

people sometimes don’t eat enough of this?
251 45
a. Fat b. Calcium
(78.7%) (14.1%)
c. Sodium d. Cholesterol
e. Don’t know
9. The bread, cereal, rice and pasta group is a good source
of .
304 6
a. Carbohydrate b. Vitamin C
(95.3%) (1.9%)
c. Calcium d. Vitamin D
e. Don’t know
10. Which of the following is required on a food label?
a. Total carbohydrate b. Sugars 215 39
c. Iron d. All of the above (67.4%) (12.2%)
e. Don’t know
11. Citrus fruits are an excellent source of .
a. Calcium b. Vitamin C 300 14
c. Vitamin B d. Calories (94.0%) (4.4%)
e. Don’t know
12. Foods as the meat, poultry, fish, dry beans, eggs and nuts
are an important source of .
151 87
a. Iron b. Fiber
(47.3%) (27.3%)
c. Beta carotene d. Calcium
e. Don’t know
13. In general, which food contains the most fat?
a. Graham crackers b. Brownies 170 76
c. Pudding d. Angel food cake (53.3%) (23.8%)
e. Don’t know
14. The milk, cheese and yogurt group are important
for .
220 8
a. Strong bones b. Teeth
(69.0%) (2.5%)
c. Muscles d. All of the above
e. Don’t know
15. How many servings of vegetables do we need each day?
a. 1-2 b. 2-3 141 41
c. 3-5 d. 6-11 (44.2%) (12.9%)
e. Don’t know

160