Do We Know What We Need To Know? Objective and Subjective Knowledge Effects On Pro-Ecological Behaviors

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

Do We Know What We Need to Know?

Objective and Subjective Knowledge Effects


on Pro-Ecological Behaviors
Pam Scholder Ellen
GEORGIA STATE UNIVERSITY

The solution of societal problems such as coping with solid waste depends, One example is the emphasis by some cereal manufacturers
in part, on identifying desirable behaviors and developing programs to that their products come in recycled boxes implying that oth-
encourage citizens to act accordingly. This research empirically examines ers don’t. In fact, the majority of cereal boxes have been made
the relationship between objective and subjective knowledge, pro-ecological from recycled materials for years because of the associated
behaviors and related attitudes. Results indicate not only that objective lower costs. In the most extreme case, some marketers have
knowledge is relatively low among a group of environmentally concerned simply employed the use of slogans or symbols designed to
individuals but that it is not significantly related to perceived knowledge lead consumers to believe the product is more “friendly” than
suggesting that persons who believe they are knowledgeable may not, in its own previous forms or its competitors-a practice termed
fact, have the requisite knowledge to make sound ecological decisions. “green washing.” As a result of the latter practices, uninformed
Examination of the effects of each type of knowledge on behaviors indicates consumers may not be able to make wise decisions in their
that objective knowledge is only sign$cantly related to committed recy- effort to reduce waste.
cling behaviors, wherens perceived knowledgeis positively associated with “Since progress toward the solution of environmental prob-
committed recycling, source reduction, and political action behaviors. The lems is likely to depend more on proenvironmental behaviors
results suggest the need for cooperation among marketers, government, than proenvironmental attitudes, the reasons for the weak re-
and environmental-oriented organizations to develop standardized mean- lationship between environmental attitudes and behaviors and
ings for ecological terms nnd symbols for labeling and advertising and to the conditions under which it can be strengthened clearly de-
devise effective educational programs for consumers. serve examination” (Van Liere and Dunlap, 1981, p. 671). As
such, there are four basic goals for this research. The first is
to develop measures of objective and subjective knowledge to
Introduction make sound precycling and recycling-based shopping deci-
opular polls show that consumers are professing in- sions. Precycling (or source reduction) refers to behaviors in-

P creased concern over the effects of the disposal of waste


on the state of the environment (Miller and Keller,
1991). However, there is substantial doubt that the actual be-
tended to reduce “garbage” by purchasing products/packages
that will have less waste (e.g., buying the frozen dinner with-
out a throwaway plate and extra packaging). Recycling refers
havior of most consumers has changed dramatically (Schwartz to the diversion of products/packages from the waste stream
and Miller, 1991). Mindful of changes in consumer tastes and and into the production of other products/packages. The other
preferences, marketers have responded with changes in their goals are to examine the association between objective and
product offerings and/or promotion to reflect environmental perceived knowledge, to determine whether knowledge leads
concern. “New products that benefit or at least don’t destroy to the performance of certain pro-environmental behaviors,
the planet are growing 20 times faster than all other new prod- and finally, to assess the effects of certain potentially conflict-
ucts hitting the market” (Courtney, 1990, p. 40). ing attitudes on the performance of these behaviors.
Marketers’ responses to the environmental trend have taken The specific focus of this research will be on the effects of
two forms. First, some companies have found ways to actually disposal on landfills rather than on air pollution or use of
reduce the deleterious effects of their products and/or pack- nonrenewable resources for two reasons. First, it is estimated
aging. Others, though, have changed or repositioned their that by the year 1995, almost 80% of all existing landfills will
product to emphasize “environmentally friendly” attributes be closed. This issue is of tantamount importance to consum-
while minimizing other negative attributes. In many cases, the ers and marketers since about one-third of all landfill waste is
changes are more cosmetic than substantive and in others, from packaging. Second, the “best” choice among product of-
offerings do not actually provide the ecological benefits touted. ferings is seldom clear-cut, i.e., a product that is “good” in
terms of one environmental problem may contribute to the
Address correspondence to Pam Scholder Ellen, Department of Marketing,
other. For simplicity, this research focuses only on the solid
Georgia State University, University Plaza, Atlanta, GA 30303.3083. waste effects of individuals’ choices.

Journal of Business Research 30,43-52 (1994) 0148-2963/94/$7.00


0 1994 Elsevier Science Inc.
655 Avenue of the Americas, New York, NY 10010
44 J Busn Res P. 5. Ellen
1994:30:43-52

selves as knowledgeable when by objective measurement they


Behavioral Determinants are not. Taylor and Brown (1988) argue that there is a per-
Positive attitudes toward the solutions to the solid waste prob- vasive human tendency to evaluate oneself as better than oth-
lem of this country are prerequisites to behavior. Specifically, ers or above average. These persons may actually be poorly
the greater the level of expressed concern the higher the per- informed and their purchase decisions or other actions may
formance of pro-ecological behaviors should be. Beyond the actually contribute to the problem rather than reducing it.
impact of expressed concern, substantial previous research on Objective knowledge measures have been used in previous
attitudes toward energy and environmental conservation has research (Maloney and Ward, 1973; Maloney, Ward, and
focused on the effects of perceived consumer effectiveness (Al- Braucht, 1975, Synodinos, 1990). These measures, however,
len, 1982; Allen and Dillon, 1979; Good, 1979; Kinnear, Tay- assessed very broad levels of knowledge (i.e., not specific to
lor, and Ahmed, 1974; Seligman et al., 1979; Webster, 1975). making choices) and covered numerous ecological issues. In-
Perceived consumer effectiveness (PCE), the extent to which terestingly, Synodinos (1990) found no significant relation-
the consumer believes the efforts of an individual acting alone ship between objective knowledge and the attitudinal and
can make a difference, has been shown to impact individual- behavioral dimensions of the revised Maloney, Ward, and
oriented behaviors (i.e., recycling) but not participation in Braucht (1975) scale. It is unclear whether this lack of rela-
group activities (i.e., joining environmental groups) (Ellen, tionship is a function of the broad nature of the knowledge
Wiener, and Cobb-Walgren, 1991). questions or reflects the fact that stronger attitudes do not
necessarily encourage knowledge development. Assessing the
former required developing a measure of objective knowledge
Effects of Prior Knowledge for the specific issue which tapped information necessary for
Failure to act in concert with attitudes may be due to a lack making behavioral choices.
of information or knowledge to make informed decisions
(Olney and Bryce, 1991). There is substantial debate among Effects of Con. ic ting Attitudes
environmentalists over the “best” products and the lack of Finally, conflicting attitudes may supersede consumers’ ex-
“hard” meanings for the phrases used to communicate ec- pressed concern for the environment and therefore inhibit be-
ological benefits (i.e., recyclable, recycled, degradable, bio- havior. An “interesting aspect of energy and environmental
degradable and environmentally friendly) across contexts problems is that the social, collective costs, and benefits of the
provides manufacturers and marketers considerable latitude individual’s actions are distant and seem negligible in com-
in applying them to their products (Mayer, Scammon, and parison to the immediate, personal rewards [or lack thereof]
Zick, 1992; Olney and Bryce, 1991). Therefore, consumers the individual obtains by his behavior” (Hummel, Levitt, and
with lower levels of knowledge may find it difficult to make Loomis, 1978, p. 39). Conflicts may occur because the re-
“good” choices because of the potential for confusion. On quired actions for solving one problem are traded-off against
the other hand, persons with greater knowledge (both ob- costs for another. According to Goldman (1991, p. 25), “the
jective and subjective) would be expected to be more likely targets for green products are the same unregenerate,
to engage in meaningful pro-environmental behaviors be- wasteful, convenience-addicted consumers that corporations
cause of greater discriminatory ability as indicated by Kao have been selling to for decades,” and who now want easy,
and Monroe (1988) findings. They showed that persons convenient products that care for the environment. They may
with higher degrees of familiarity or knowledge were more compensate by guilt reduction shopping (i.e., insisting on and
likely to use intrinsic (i.e., physical product) cues to make buying more environmentally safe products) (Courtney,
quality assessments where as those with lesser knowledge 1990). Thus, a person may value the benefits afforded by ex-
relied on extrinsic (i.e., attributes not related to the physical isting behavior (i.e., convenience, time savings) more than
product, i.e., price) cues. The appropriateness of using ex- they value the long-term, societal effects. Specifically, given
trinsic cues depends on experience having shown that such the other significant demands in their lives, people may be
indicators were accurate. In this case, extrinsic cues, such unwilling to expend significant time and effort to actively
as labeling, are at best equivocal in communicating the en- change their behavior (i.e., reducing usage, reusing, and re-
vironmental impact of the product or package. cycling products, joining environmental organizations, writ-
Measures of perceived and objective knowledge might be ing public officials).
expected to be highly correlated, and thus redundant; how-
ever, they may in fact have differential effects on behavior.
One person may have the requisite knowledge to make in-
Method
formed decisions, However, they may not feel well-informed Focus interviews were conducted with 37 area citizens, uni-
because the “right” choice is not perfectly clear in all situa- versity employees and students. They were questioned about
tions Their uncertainty may be used as an excuse for not their response to increasing media attention and social pres-
acting accordingly and to foist their responsibilities on others. sure to make changes in their purchasing and disposal habits
On the other hand, another person may perceive them- and asked the reasons for their participation/nonparticipation
....V..,cU.,~ Effects on Pro-Ecological Behaviors J Busn Res 45
1994:30:43-52

Table 1. Factor Analysis Results


Convenience Committed Source Political
Recycling Recycling Reduction Action

Newspapers .92
Aluminum cans .92
Glass .80
Plastic containers .81
Paper .58
Cloth or string bag .47
Larger sizes .68
Recycled paper products .63
Refused shopping bag .73
Avoided polystyrene .75
Requested name be removed from mailing lists .47
Attended public hearings/meetings .72
Telephoned or wrote public officials .57
Contributed to environmental organization .54
Participated in environmental organization .70
Note: Loadmgs < .4 were suppressed to fmlitate interpretanon.

in recommended actions. Then they were asked the meaning legal definitions of such terms, (2) examining guidelines pro-
of a variety of symbols and phrases found on packages and in vided by accepted authorities (i.e., Environmental Protection
advertisements. Their answers were the basis for items for Agency) on the environmental impact of consumer choices,
several of the constructs and for alternative answers for the (3) an extensive review of guidelines provided by popular
objective knowledge scale. press sources (Corson et al., 1990; Makower, Elkington, and
Prior research in the areas of energy and environmental Hailes, 1991; The Earth Works Group, 1989, 1990), and (4)
conservation provided a rich source of items as a starting place an examination of phrases, logos, and symbols used on pack-
for the development of specific items for each of the constructs aging and in advertisements to transmit or infer information
(cf., Allen and Dillon, 1979; Gill, Crosby, and Taylor, 1986; about the environmental impact of the products or packaging.
Maloney, Ward, and Braucht, 1975). However, because of the The latter again was used to assess the existence and use of
diversity of issues addressed in this earlier research, applicable cues which may lead consumers to infer that certain products
items were adapted to the specific topic (i.e., concern with the may be “environmentally friendly.” The focus interviews pro-
effects of disposal on landfills) and supplemented with items vided believable incorrect answers. Finally, a solid waste en-
from specific comments from focus interviews. gineer from the Environmental Protection Agency served as a
Measures were developed for both subjective and objective final expert judge of the accuracy of all items. (The final ob-
knowledge of the information necessary to choose products jective knowledge scale is available from the author.)
with consideration for disposal and its impact. The subjective Behavior was measured as the degree to which respondents
measure used was a four-item indicator reflecting perceived performed 15 different behaviors. A range of behaviors were
knowledge of both precycling and recycling issues and the selected from those most commonly recommended by envi-
certainty about choices. ronmental groups and government agencies. The behaviors
The objective measure of knowledge was designed as a mul- fell into three basic types: (1) recycling behaviors, (2) precy-
tiple choice measure that incorporated items assessing both cling or source reduction behaviors, and (3) political actions.
declarative and procedural knowledge. According to Brucks The set of items are shown in Table 1 along with factor analysis
(1985), declarative knowledge is knowledge about concepts, results indicating four factors. The first five items were recy-
objects, and events. This would include specific verbal and cling activities that factor analysis indicated formed two fac-
non-verbal symbols. Procedural knowledge is rules for taking tors. These are labeled convenience and committed recycling.
action based on the declarative knowledge. In developing the Convenience recycling included saving and recycling news-
knowledge scale, items were included to tap both declarative papers and aluminum, two of the most readily recycled prod-
and procedural knowledge for precycling and recycling issues, ucts. The second three items (i.e., recycling glass, plastic and
using words/phrases as well as symbols. In addition, at least paper) generally require more time and effort and therefore
one item used a “meaningless” symbol used by a manufacturer are described as committed recycling activities. The next five
that consumers might use to infer ecological benefits. Thus, behaviors represent precycling activities, or source reduction
the scale assesses the potential for attaching meaning to any activities. The fourth factor, labeled political action, comprises
“ecology-related” information. behaviors more indicative of politically or group-based activ-
The specific items were developed through: (1) a review of ities rather than individual actions. Overall measures for each
46 J Busn Res P. S. Ellen
1994:30:43-52

Table 2. Comparison of Sample to Population on Selected and education than the population as a whole (see Table 2).
Demographtc Variables It is understood that generalizations to the population as a
Sample Population whole are significantly affected by self-selection, particularly
Age 46.13 32.58 among the second sample. However, the primary concern
home $4&Y50 $37,414 here is on assessing knowledge and its impact on pro-
Median years of educanon 16.0 13.16 environmental behaviors. Since participants are more likely
% White 78.8% 79% concerned with the problem and therefore may be more likely
Sour<c CACl Sourcehook of Drmographm and Bujmg Power for Every Zip Code I” rhe
to perform some of the activities, the general purpose of the
USA. 1988
project is not hindered.

type of behavior were created by summing the answers for the


set of items.
Results
A ptlot study with two envtronmentally oriented groups The average reported participation for each of the four types
served as a pretest for most items. Based on the results, certain of behavior were 6.30, 6.24, 9.25, and 1.04 on scales with
items were adapted. The behavioral measures showed the maximum values of 8, 12, 20, and 5, respectively. The set of
same factor structure indicated in Table 1. The objective items for measuring the additional constructs were created by
knowledge scale was identical for eight of the nine items. One summing across items. Descrtptive statistics and coefficient
item was slightly rewritten because of ambiguity. For the eight alphas are shown for each in Table 3. Of particular note is the
common items, the correlation between the pilot test and the relatively high level of reported concern and perceived con-
final research results was .94 (p < O.Ol), indicating reliability sumer effectiveness (means = 22.43 and 23.84 on 28-point
across groups. scales, respectively). While the reliabilities for several of these
Previous research has shown that there are often significant constructs are below the desirable level of coefficient alpha,
differences in pro-environmental attitudes and behaviors they are, in most cases, consistent with those found in pre-
based on demographic variables (Van Liere and Dunlap, vious research in this area (cf. Allen and Dillon, 1979; Ellen,
1980). For this reason, the final survey also included demo- Wiener, and Cobb-Walgren, 1991). Examination of the items
graphic questions such as age, sex, education, and political and protocols with respondents indicated that many items are
party affiliation. interpreted differently based on the context and the individ-
A mail questionnaire was designed to reduce the potential ual. Measures of concern, for example, indicate both the de-
high degree of social pressure to respond in an environmen- gree to which a person believes there is a problem and the
tally correct way. This should provide respondents with a personal relevance of the problem. Similarly, the measure of
greater feeling of anonymity and reduce interviewer effects. perceived knowledge taps perceived knowledge about recy-
To ensure participation and honest answers, care was taken cling and precycling as well as level of certainty. For this rea-
to develop counterbiasing instructions and statements to son, it is more appropriate to view these items as formative
make the respondents more comfortable giving socially “un- indicators, making measures of inter-item consistency, such
acceptable” answers. as coefficient alpha, unsuitable (Howell, 1987).
A two-stage sampling procedure was used. First, a system- A total score for the objective knowledge scale was created
atic random sample of metropolitan residents for a major by summing the number of correct answers to the nine ques-
southeastern city were selected from the area residential tel- tions. The average number of correct answers was 4.0 on a
ephone book. The qualification for participation was that the nine-point scale, indicating that, on average, respondents got
person be at least 18 years of age and shopped for groceries less than half the questions correct. Of the sample, 83.9% got
at least once a month. Potenttal respondents were precon- five or fewer items correct. In light of the fact that this was a
tacted by phone, and 321 agreed to complete the mail ques- mail questionnaire, which provided motivated respondents
tionnaire sent within the next week. Of those, 231 returned the opportunity, if desired, to look for answers, this result is
usable questtonnaires for a 72%-response rate. Because of the disappointing.
previous difficulty of getting nonwhite respondents in envi- The percentage of respondents giving the correct answer
ronmental surveys and because initial returns showed poor for each item ranged from 83% for the term recyclable to 20%
nonwhite response, a second cold mail survey was conducted. for recognizing the symbol indicating a product or package
Two census tracts were identified with large percentages of was made from recycled materials. The fact that people do not
nonwhite residents (i.e., 84.6% and 86.5% black residents) recognize the difference between the recycled symbol and the
who also had education levels of at least some high school recyclable symbol is particularly noteworthy since the two are
(77.8% and 61% with median education of at least 12 years). often not differentiated in educational materials. In fact, many
Eight hundred surveys were mailed to these households, with materials show the recyclable symbol and indicate it can mean
76 returned by the cutoff date for a 9.5%-response rate. either recycled or recyclable. A review of labels also shows
The characteristics of the combined sample indicate that that the symbols are often used incorrectly.
the sample is significantly older than and has higher income Particularly noteworthy, but not surprising, is that most
nnowleage Effects on Pro-Ecological Behaviors J Busn Res 47
1994:30:43-52

Table 3. Survey Measures


Perceived Knowledge (Mean = 13.51; SD = 3.78; Range = 4-28; Alpha = .43)
Trying to figure out the best packaging in terms of the effects on the environment is very confusing. (REVERSED)
1 am very confident that 1buy products in packages which are environmentally safe.
I know more about recycling than the average person.
Every time 1 turn around, the “experts” are changing their minds about what are the “best” products or packages to buy. (REVERSED)

Environmental Concern (Mean = 22.43; SD = 4.13; Range = 4-28; Alpha = .56)


Compared to other things in my life, environmental problems are not that important to me. (REVERSED)
Environmental problems are of great concern to me personally.
Environmental problems are not that serious because in the long run things will balance out. (REVERSED)
I can think of many things I’d rather do than work toward improving the environment. (REVERSED)

Perceived Consumer Effectiveness (Mean = 23.84; SD = 3.60; Range = 4-28; Alpha = .48)
The recycling efforts of one person do make a difference.
Each consumer’s behavior can have an effect on reducing landfill problems.
There is not much that any one individual can do about environmental problems. (REVERSED)
What 1 purchase as a consumer does have an effect on the nation’s landfill problems.

Sacrifices (Mean = 6.76; SD = 3.01; Range = 2-14; Alpha = .42)


Buying products to make my life easier is more important right now than buying to reduce the amount of garbage.
It is difficult to give up the convenience of disposable products.

Shopping Effort (Mean = 8.86; SD = 2.96; Range = 2-14; Alpha = .79)


Choosing products in packaging that is environmentally safe means spending a lot more time shopping.
Shopping for recycling or for products in recycled packages requires a lot of extra effort.

Recycling Effort (Mean = 7.00; SD = 3.32; Range = 2-14; Alpha = .73)


Recycling requires a lot of extra time.
Recycling is messy and requires a lot of space.

consumers (88.2%) ascribed a meaning to a symbol which has tion differences were for Republicans who indicated higher
no recognized meaning. The symbol used in the test was taken objective knowledge than either other group. The only differ-
from a battery package which included claims about the de- ences in marital status were for convenience recycling and
gree to which the product was mercury-free. Over 50% inter- political action in which married respondents reported higher
preted the symbol to indicate the package was made from incidences.
recycled pulp materials. This supports the contention that Correlations between the two measures of knowledge were
marketers using symbols, phrases, colors and other cues may then examined. Interestingly, there was no significant rela-
lead consumers to erroneously believe that their product/ tionship between perceived and objective knowledge indicat-
package offers certain benefits it does not. ing that a person’s confidence in their ability to make “right”
Differences along demographic lines in objective and sub- choices is not reflective of their objective knowledge (r = .08).
jective knowledge and four types of behavior are shown in Given this result, the effect of each type of knowledge on the
Tables 4, 5, and 6. For age, persons 55 and over indicated less behaviors was examined separately to determine how their
objective and perceived knowledge than other age groups, effects differed.
more recycling of newspaper and aluminum and less source Multivariate analysis of covariance was used to examine the
reduction behavior. Persons with an income under $50,000 effect of objective and subjective knowledge as well as concern
also showed less objective and subjective knowledge as well and PCE on the performance of the sets of behaviors. Median
as significantly less performance across all four types of be- splits were used to form high and low levels of each construct,
havior. Similarly, persons with a high school degree or less Perceptions of the sacrifices, shopping effort and recycling ef-
had less knowledge (both objective and perceived) as well as fort required to behave ecologically were examined as covar-
less committed recycling and less political behavior. Sex dif- iates. (There was no significant interaction between any of the
ferences were found only for source reduction and political covariates and factors.) Since none of the two-way or three-
behaviors with men doing significantly less precycling and way interactions were significant, Table 7 shows only the main
women participating less in political activities. Blacks scored effects and covariate effects while Table 8 provides the cell
lower on objective knowledge as well as the performance of means for each. For objective knowledge, the covariates were
both types of recycling behaviors while the only party affilia- significant overall, indicating their effects on all behavior sets
48 J Busn Res P. S. Ellen
1994:30:43-52

Table 4. Differences in Objective and Perceived Knowledge on Demographic Characteristics


Objective Knowledge Perceived Knowledge

Significant Significant
F Means Contrasts F Means Contrasts

Age 3.89" <35 = 6.76 1-3, 2-3 8.83” <35 = 3.61 l-3.2-3
35-54 = 7.58 35-54 = 3.43
>54 = 3.61 >54 = 3.03
Income 4.72” <50K = 3.87 l-2 4.20” <50K = 3.26 l-2
50K+ = 4.28 50K+ = 3.49
Education 12.10” <College = 3.37 1-2 5.08 <College = 3.13 l-2
College+ = 4.16 College+ = 3.43
Sex .oo Male = 4.04 - .77 Male = 3.43 -
Female = 4.04 Female = 3.33
Ethnic 19.38” Black = 3.17 l-2 .05 Black = 3.41 -
White = 4.23 White = 3.37
Party 4.12” Republican = 4.34 1-2, l-3 1.44 Republican = 3.33 -
Democrat = 3.78 Democrat = 3.51
Independent = 3.81 Independent = 3.29
Marital 1.50 Married = 4.08 - .76 Married = 3.34 -
Not Married = 3.84 Not Married = 3.34
“pi 0 05.

except political action. Specifically, persons perceiving greater While results for concern and PCE were similar to those
sacrifice reported less convenience recycling (t = -2.62, p = found for objective knowledge, the effects of perceived knowl-
0.01) while those viewing recycling effort as greater reported edge were greater than for objective knowledge. Subjective
less convenience (t = -2.48, p = 0.01) and committed re- knowledge positively affected not only committed recycling
cycling (t = -3.25, p < 0.01). Those who saw shopping effort but also reported source reduction and political activism. This
as being greater reported less source reduction behavior (t = suggests that while objective knowledge was only related to
-1.73, p = 0.09). Thus, as expected, persons who perceived one type of behavior, perceived knowledge affected several.
greater personal loss or effort reported performing less of Given the lack of significant relationship between objective
those behaviors. People who reported greater recycling and and perceived knowledge, it seems likely that persons per-
source reduction may perform the behaviors because they do forming these behaviors may believe they have the requisite
not perceive any substantial loss or, after beginning the ac- knowledge to make “good” choices. However, objective
tions, adjust their perceptions of the associated losses to justify knowledge scores do not necessarily indicate this. Although
the behavior. The lack of effect on political action is not totally there is no direct evidence that the actions of those with high
surprising since the specific sacrifices did not address the ef- perceived knowledge are ineffective, there certainly is some
fort/loss associated with such behaviors. basis for concern.
All three main effects were significant. The effect of objec-
tive knowledge, while significant (F = 2.07, p s 0.05). only
affected committed recycling such that those with higher
Implications and Recommendations
knowledge reported greater recycling. The level of concern The results of this research indicate that while the level of
affected both committed recycling, source reduction and po- reported environmental concern and perceived consumer ef-
litical activism with higher concern associated with higher per- fectiveness were relatively high for the respondents, their level
formance of the behaviors. Persons reported higher levels of of objective knowledge was not correspondingly high. In ad-
all four behaviors with higher levels of perceived effectiveness. dition, the level of objective knowledge is not significantly
Results were then considered for the effects of subjective correlated with their level of perceived knowledge. Examining
knowledge (also shown in Table 7). The covariates were again results across a set of different behaviors indicated that objec-
significant but only for convenience and committed recycling. tive knowledge was predictive only of committed recycling
As before, persons feeling greater sacrifice reported less con- while perceived knowledge was an important indicator of
venience recycling (t = -2.53, p = 0.01) and those perceiving committed recycling, source reduction, and political action to
greater recycling effort reported less convenience (t = -2.20, reduce the problems of solid waste.
p = 0.03) and committed recycling (t = -3.16, p c 0.01). While concern and effectiveness were predictive of the per-
Contrary to the previous model, there was no effect of shop- formance of almost all of the behaviors in both analyses, sev-
ping effort on source reduction behaviors. eral conflicting attitudes actually reduced the perceived value
Knowledge Effects on Pro-Ecological Behaviors J Bum Res 49
1994:30:43-52

Table 5. Differences in Convenience and Committed Recycling on Demographic Characteristics


Convenience Recycling Committed Recycling
Significant Significant
F Means Contrasts F Means Contrasts
Age 6.55” <35 = 5.68 l-2,1-3,2-3 1.48 <35 = 6.49 -
35-54 = 6.26 35-54 = 6.35
>54 = 6.94 >54 = 5.75
Income 2.85” <50K = 5.97 l-2 2.66b <50K = 5.92 l-2
50K+ = 6.44 50K+ = 6.50
Education .29 <College = 3.37 - 4.33” <College = 5.51 l-2
College+ = 4.16 College+ = 6.41
Sex .56 Male = 6.18 - .58 Male = 6.41 -
Female = 6.38 Female = 6.14
Ethnic 7.87” Black = 5.54 l-2 3.62b Black = 5.49 l-2
White = 6.53 White = 6.41
Party 1.56 Republican = 6.47 - .06 Republican = 6.09 -
Democrat = 5.98 Democrat = 6.22
Independent = 6.47 Independent = 6.22
Marital 2.92b Married = 6.44 l-2 .33 Married = 6.64 -
Not Married = 5.96 Not Married = 6.35
“p < 0.05
hp c 0 10.

of performing the behaviors because of the associated personal behaviors. The lack of effects on political action may be be-
loss or effort. As expected, the degree of sacrifice associated cause of the limited extent to which these types of behaviors
with convenience and time-saving products affected the de- were performed as well as the fact that the covariates more
gree of convenience recycling; similarly, the perceived effort closely addressed the losses and efforts associated with pre-
to recycle reduced the extent of committed recycling and the cycling and recycling.
perceived greater effort of shopping reduced source reduction The disparity between perceived knowledge and objective

Table 6. Differences in Source Reduction and Political Actions on Demographic Characteristics


Source Reduction Activities Political Actions
Significant Significant
F Means Contrasts F Means Contrasts
Age 4.27” <35 = 10.84 l-3,2-3 1.34 <35 = 1.09 -
35-54 = 11.02 35-54 = 1.13
>54 = 9.62 >54 = .86
Income 1.95 <50K = 10.34 - 12.75” <50K = .79 l-2
50K+ = 10.92 50K+ = 1.31
Education .lO <College = 10.44 - 8.05” <College = .63 l-2
College+ = 10.61 College+ = 1.14
Sex 4.39” Male = 10.13 l-2 4.29” Male = 1.22 l-2
Female = 10.99 Female = .91
Ethnic .73 Black = 10.18 - 1.24 Black = .88 -
White = 10.65 White = 1.10
Party 1.45 Republican = 10.22 .96 Republican = .98 -
Democrat = 11 .OO Democrat = 1.18
Independent = 10.75 Independent = .93
Marital .03 Married = 10.61 4.19” Married = 1.25 l-2
Not Married = 10.54 Not Married = .94
“p ‘= 0 05.
“p 5 0 10
50 J Bum Res P. 5. Ellen
1994:30:43-52

Table 7. Multivariate and Umvariate Analysis of (Co)Variance


Perceived Consumer
Concern Effectiveness Knowledge Covariates

Wilks Wilks’ Wilks Wilks


Dependent Variables lambda F-value lambda F-value lambda F-value lambda F-value

Multlvariate for objective knowledge .94 3.87” .95 2.98” .97 2.07” .a7 3.08”
Univariate
Convenience recycling 2.13 8.48” 1.43 4.35”
Committed recycling 4.94” 5.83” 7.79” 6.50”
Source reduction 5.16” 3.15” 2.08 3.66”
Political activism 12.76” 3.03h .64 .60
Multivariate for subjective knowledge .95 3.53” .96 2.83” .96 2.62” .90 2.25”
Univariate
Convenience recycling 2.34 7.01” .04 3.67”
Committed recycling 5.13” 7.31” 6.61” 4.13”
Source reduction 6.19 3.74” 3.74” 1.32
Political activism 10.27” 2.20” 3.17h .51

knowledge deserves increased attention. While this research such terms as degradable, compostable, recycled, and recycl-
did not allow for such assessment, it is important to determine able, the states have passed strict legislation and the task force
whether those persons reporting higher perceived knowledge issued its own guidelines. A review of states’ legislation shows
are in fact performing effective behaviors-that is, are they that many now mandate the definition of certain terms/
making “good” purchasing choices and recycling materials phrases (many of which vary by state) along with requiring
correctly. Anecdotal evidence suggests that the “rules” for pur- quotas for products and/or packaging. These specifications in-
chasing and recycling are often complicated so that materials clude recycled content claims which will have to be specific,
are handled inappropriately (i.e., nonrecyclable or mixed pa- disclosing the percentage of recycled materials and include
pers are combined with white paper; nonrecyclable plastic only postconsumer material. This legislation has forced many
materials are combined with recyclable resin products). Such manufacturers to drop “green” claims since legislation varies
actions result in increased costs for handling materials, lost so dramatically and in most cases, locale-specific labeling is
revenues from contamination of materials and, in some cases, impractical (Lawrence, 1991). In other cases, certain pack-
discontinuation of services. aging materials such as polystyrene will be strictly banned
In addition, if persons reporting high perceived knowledge within the next few years except where food and drug regu-
do not show higher levels of objective knowledge of relevant lations offer no acceptable alternative.
terms, symbols, and phrases, what information are they using? To provide consistency, the FTC recently released guide-
The best way to assess this would be through in-store inter- lines for specific or implied environmental claims used in
views with concerned consumers in which they indicate what labelling, advertising, promotional materials and other mar-
they look for in making purchases and deciding how to dis- keting vehicles (Federal Trade Commission, 1992). These
pose of the wastes of consumption. guidelines may preempt other states from setting their own
The low levels of actual knowledge demand action from guidelines and thus ease the difficulties faced by some mar-
several constituencies. For those governmental, marketing, keters in creating proper labeling for their products,
consumer and other groups interested in environmental pro- For marketers who are genuinely concerned with environ-
tection, understanding the motivations and ability to make mental effects, standard definitions would allow them to: (1)
“good” decisions can be used to tailor educational and per- more accurately promote the benefits associated with their
suasive messages. Specifically, this research indicates the need products, and (2) encourage and/or monitor other marketers
for not only improved education but the development of stan- to develop more accurate labels and promotion before gov-
dard definitions that can be conveyed to consumers to im- ernment intervention. The practice of “green washing”-the
prove their ability to make “good’ choices. use of slogans or symbols designed to lead the consumer to
The role of federal government agencies such as the Federal believe the product is more “friendly” than its own previous
Trade Commission, Environmental Protection Agency, and forms or its competitors’-likely results in uninformed con-
Food and Drug Administration in establishing standards has sumers being unable to make “wise” decisions in their effort
been usurped by other groups. A task force of 11 state attor- to reduce waste. Such practices encourage governmental in-
neys general issued Green Report II, warning marketers that tervention
the states won’t wait for other agencies to act (Lawrence, At the current time, there is often substantial disparity be-
1991). While calling for federal government standards for tween claims made by marketers and the “true” meaning of
Knowledge Effects on Pro-Ecological Behaviors J Busn Res 51
1994:30:43-52

environmental concern and perceived effectiveness, it seems


Table 8. Cell Means
clear that the sample through self-selection disproportionately
Low Objective High Objective
reflects those persons for whom environmental issues are im-
Knowledge Knowledge
portant. While these efforts to develop concern should con-
Low High Low High tinue, particularly with respect to other segments of the
Concern Concern Concern Concern
population, it is apparent that greater effort must be expended
Low perceived effectiveness in the effective education of consumers. Future research needs
Convenience recycling 5.33 6.40 5.00 6.22 include determining what specific information consumers are
Committed recycling 4.75 6.52 5.30 6.67
using to make pro-ecological decisions and to determine from
Source reduction 9.35 11.00 9.10 11.33
Political activism .90 1.00 .35 1.50 that the degree to which corrective information may be nec-
Sacrifice 8.90 8.20 9.80 7.72 essary. The literature on corrective advertisement should offer
Shopping effort 8.92 8.36 9.15 8.22 significant insights into the proper design of such programs.
Recycling effort 8.35 6.32 7.80 6.50 For consumer researchers, this area of research broadens
High perceived effectiveness
7.08 7.47 the focus of product/service evaluation and choice for con-
Convenience recycling 6.32 6.39
Committed recycling 5.68 6.40 7.08 8.16 sumption to include issues of disposal. The results should
Source reduction 10.03 11.04 11.39 11.96 provide greater insight into consumers’ response to environ-
Political activism .65 1.46 1.15 1.49 mental issues and more specifically to the development and
Sacrifice 8.13 7.10 8.19 6.71 testing of social marketing programs.
Shopping effort 9.42 8.17 9.69 8.64
Recycling effort 6.94 6.27 7.85 6.33
The author gratefully acknowledges the assistance of Deborah Webb and
Low Subjective High Subjective Laura Thomson in conducting this research.
Knowledge Knowledge
Low High Low High
Concern Concern Concern Concern References
Allen, Chris T., Self-Perception Based Strategies for Stimulating En-
Low perceived effectiveness ergy Conservation. J. Consumer Res. 8 (March 1982): 381-390.
Convenience recycling 4.88 6.24 5.59 6.39
Allen, Chris T., and Dillon, Wilham R., On Receptivity to Information
Committed recycling 4.68 5.41 5.15 7.35
Furnished by the Public Policymaker: The Case of Energy. Edu-
Source reduction 9.00 10.53 9.56 11.54
cator’s Conference Proceedings. Neil Beckwith et al., eds., Amencan
Political activism .62 1.12 .85 1.27
Marketing Association, Chicago, IL. 1979. pp. 550-556.
Sacrifice 9.91 9.47 8.41 7.04
Shopping effort 9.32 9.41 8.65 7.58 Brucks, Merrie, A Typology of Consumer Knowledge Content. Ad-
Recycling effort 8.88 6.29 7.50 6.46 vances in Consumer Research. Richard J. Lutz, ed., Association for
High perceived effectiveness Consumer Research, Provo, UT. 13 (1985): 58-63.
Convenience recycling 6.69 6.76 6.64 7.06 Corson, Ben, Downey. John, Marlin, Alice Tepper, Schorsch, Jona-
Committed recycling 5.89 6.33 6.88 8.02 than, Swaab, Emily, and Will, Rosalyn, Shoppingfor a Better World.
Source reduction 10.13 10.47 11.32 12.47 Council on Economic Priorities, New York. 1990.
Political activtsm .84 1.09 .92 1.78
Courtney, Alexandra, It’s Not Easy Being Green. Marketing Insights
Sacrifice 8.75 7.87 7.40 6.00
(Fall 1990): 39-48.
Shopping effort 10.63 9.24 8.16 7.53
Recycling effort 8.53 6.42 5.84 6.04 Ellen, Pam Scholder, Wiener, Joshua and Cobb-Walgren, Cathy, The
Role of Perceived Consumer Effectiveness in Motivating Environ-
mentally Conscious Behaviors. J. Public Policy and Marketing 10
(2)(1991): 102-117.
Federal Trade Commission, Guides for the Use of Environmental Mar-
those claims in local communities. This provides a opportu- keting Claims. Washington, D.C. July 1992.
nity for local environmental groups (i.e., Sierra Club) to focus Gill, James D., Crosby, Lawrence A., and Taylor, James R., Ecological
their efforts on educating consumers about the meaning of Concern, Attitudes, and Social Norms in Voting Behavior. Public
such claims in terms of the availability of local disposal or Opinion Q. 50 (1986): 537-554.
recycling facilities. In addition, local government can work for Goldman, Debra, Second Thoughts on the “Green Revolution.” Ad-
increased access to and knowledge of collection facilities. In week’s Marketing Week (February 18, 1991): 24-25.
some communities, certain products/packages may be col- Good, Walter S., Canadian Attitudes toward Energy Conservation:
lected through curbside recycling while others may have to Implications for Public Policymakers, in Public Policy Issues in Mar-
be taken to a variety of locations. The time and effort required keting, vol. 1, Cynthia Frey, Thomas Kinnear and Bonnie Reece,
eds., University of Michigan Press, Ann Arbor, MI. 1979, pp. 146-
for these “good Samaritan” activities decrease consumers’ in- 159.
terest in performing them.
Howell, Roy D., Covariance Structure Modeling and Measurement
Much of the effort by governmental and other agencies or Issues: A Note on Interrelations Among a Channel Entity’s Power
organizations has been allocated to increasing awareness of Sources. J. Marketing Res. 24 (February 1987): 119-126.
and concern for the environment. While the sample of persons Hummel, Carl F., Levitt, Lynn, and Loomis, Ross J., Perceptions of
included in this research indicates a generally high level of the Energy Crisis: Who Is Blamed and How Do Citizens React to
52 J Bum Res P. S. Ellen
1994:30:43-52

Environmental-Lifestyle Trade-Offs? Environment and Behavior 10 Rao, Akshay R., and Monroe, Kent B., The Moderating Effect of Prior
(1)(1978): 37-89. Knowledge on Cue Utilization in Product Eva1uations.J. Consumer
Kinnear, Thomas, Taylor, James R., and Ahmed, Sadrudin, Ecologi- Res. 15 (Z)(September 1988): 253-264.
cally Concerned Consumers: Who Are They?J. Marketing 38 (April Schwartz, Joe, and Miller, Thomas, The Earth’s Best Friends. Am.
1974): 20-24. Demographics (February 1991): 26-35.
Lawrence, Jennifer, State Guides Define Green Terms. Advertising Age Seligman, Clive, Kriss, M., Darley, John M., Fazio, R. H., Becker, L.
(May 27, 1991): 3. J., and Pryor, J. B., Predicting Summer Energy Consumption from
Makower, Joel, Elkington, John, and Hailes, Julia, The Green Con- Homeowner’s Attitudes. J. Appl. Sot. Psychof. 9 (1979): 70-90.
sumer Supermarket Guide, Penguin Books, New York. 1991. Synodinos, Nicolaos E., Environmental Attitudes and Knowledge: A
Comparison of Marketing and Business Students with Other
Maloney, Michael P., and Ward, Michael P., Ecology: Let’s Hear It
Groups. J. Bus. Res. 20 (1990): 161-170.
from the People: An Objective Scale for the Measurement of Ec-
ological Attitudes and Knowledge. Am. Psychol. 28 (July 1973): Taylor, Shelley E., and Brown, Jonathon D., Illusion and Well-Being:
583-586. A Social Psychological Perspective on Mental Health. Psychol. Bull.
103 (2) (1988): 193-210.
Maloney, Michael P., Ward, Michael P., and Braucht, G. N., A Revised
Scale for the Measurement of Ecological Attitudes and Knowledge. The Earth Works Group, 50 Simple Things You Can Do to Save the
Am. Psychol. 30 (1975): 787-790. Earth, Earthworks Press, Berkeley, CA. 1989.
Mayer, Robert, Scammon, Debra and Zick, Cathleen, Turning the The Earth Works Group, The Recycfer’s Handbook, Earthworks Press,
Competition Green: The Regulation of Environmental Claims. Pro- Berkeley, CA. 1990.
ceedings of the 1992 Marketing and Public Policy Conference. Paul N. Van Liere, Kent D. and Dunlap, Riley E., The Social Bases of Envi-
Bloom and Richard G. Starr, Jr., eds., Washington, DC. 1992. pp. ronmental Concern: A Review of Hypotheses, Explanations and
152-165. Empirical Evidence. Public Opinion Q. 4 (1980): 181-197.
Miller, Thomas A. W., and Keller, Edward B., What the Public Van Liere, Kent D., Environmental Concern: Does It Make a Differ-
Thinks. EPAJoumal (March-Apnl 1991): 40-43. ence How It’s Measured? Environment and Behavior 13 (6)(1981):
Olney, T. J., and Bryce, Wendy, Consumer Responses to Environ- 651-676.
mentally Based Product Claims. Advances in Consumer Research. Webster, Frederick E., Jr., Determining the Characteristics of the So-
Rebecca H. Holman and Michael R. Solomon, eds., Association for cially Conscious Consumer. J. Consumer Res. 2 (December 1975):
Consumer Research, Provo, UT. 18 (1991): 693-696. 188-196.

You might also like