Articulo Investigacion

Martin Eisend
Have We Progressed Marketing

Knowledge? A Meta-Meta-Analysis of
Effect Sizes in Marketing Research
This study measures the value and progress of knowledge produced in marketing research by using meta-analytic
effect sizes as a measure of scientific knowledge. The author combines the results of 176 meta-analyses that
include data from more than 7,500 primary studies published between 1918 and 2012. The 1,841 meta-analytic
effect sizes show that a considerable body of marketing knowledge has been developed, as expressed by a metameta-analytic correlation of .24. This medium-sized effect is as strong or stronger than effects that have been found
in compilations of meta-analyses in other, more basic fields of inquiry (e.g., psychology), which shows that
marketing is a successful academic discipline. The effect sizes vary across subject areas, with pricing showing the
strongest effects, followed by consumer behavior; methods and new product development show the weakest
effects. This finding reveals different degrees of knowledge production and varying benchmarks to assess the
contribution of future research outcomes in these subject areas. Marketing knowledge follows a discontinuous
model of progress: knowledge has increased over time, but at a decreasing rate; the marketing field, which is
currently characterized by fragmentation and specialization, has reached a stage of maturity. The findings provide
implications for further research regarding how to measure, evaluate, and progress knowledge in marketing.
Keywords: marketing knowledge, knowledge accumulation, effect size, meta-analysis
Online Supplement: http://dx.doi.org/10.1509/jm.14.0288
ince being recognized as a separate field of inquiry

almost 100 years ago, the marketing discipline has
generated an impressive body of knowledge. This
knowledge is based on myriad theories, models, and concepts as well as thousands of empirical studies. The field
has matured to a point at which researchers are taking stock
of the knowledge that marketing research has produced and
investigating how knowledge has progressed so far. Several
articles and books on the history, development, domains,
and principles of marketing knowledge attest to this idea
(e.g., Hanssens 2009; Sheth and Sisodia 1999; Stewart and
Zinkhan 2006; Wilkie and Moore 2003). However, the conclusions on the progress of marketing knowledge differ
somewhat. Some believe that marketing knowledge has
advanced remarkably and made significant progress (e.g.,
Stewart and Zinkhan 2006), while others object that marketing knowledge has become less important and influential
over time because the discipline has neglected conceptual
developments (Yadav 2010) or has traded relevance for
analytical complexity and technical sophistication (Lehmann,
McAlister, and Staelin 2011).
How much do we know about marketing and how has

this knowledge developed? Unlike previous studies that discuss the value and progress of marketing knowledge, this
study measures the value and the progress of knowledge in
marketing research by using meta-analytic effect sizes as
measures of scientific knowledge. Meta-analytic effect
sizes describe the explanatory potential of the concepts and
theories that marketing has developed to explain behavior
in markets.
This study contributes to the extant literature that reflects
on the status of marketing research in several ways. First, by
measuring the value of the body of knowledge in marketing
with meta-analytic effect sizes, researchers can evaluate the
body of knowledge against common benchmarks for the
size of an effect, compare it with the meta-analytic findings
of other disciplines, and decide whether marketing research
has been successful. Second, by providing an empirical and
quantitative account of how much variance of behavior in
markets we can explainthat is, how much we know about
market behaviorthe meta-analytic effect size shows the
predictive power of our knowledge for the whole field, and
particularly for different subject areas. The meta-analytic
figures make knowledge production in different subject
areas accountable and comparable and set benchmarks for
assessing the value and contribution of future research projects. Third, by analyzing how knowledge has developed
over time and how progress varies across subject areas in
marketing, the study identifies the model of knowledge
Martin Eisend is Professor of Marketing, European University Viadrina

(e-mail: [email protected]). The author acknowledges the helpful
comments provided by Donald R. Lehmann and thanks the JM review
team for insightful suggestions, which improved this article considerably.
Roland T. Rust served as area editor for this article.
2015, American Marketing Association

ISSN: 0022-2429 (print), 1547-7185 (electronic)
23
Journal of Marketing
Vol. 79 (May 2015), 2340
development that applies to marketing research and

assesses the maturity of the discipline. These findings complement previous reviews on the thought development of
marketing research (Wilkie and Moore 2003). In particular,
the current era of marketing research is characterized by
specialization and fragmentation, and this study finds that
knowledge can increase or decrease in this era depending
on how a subject area deals with specialization. If a specialized area uses exchange across disciplinary borders to provide efficient response to new challenges, the areas knowledge base can increase. Fourth, the findings provide
implications for further research in marketing and for the
academic community on how to assess, evaluate, and
progress knowledge; how to avoid a deceleration in knowledge development; and how topic decisions can benefit academic careers.
The reminder of the article is organized as follows. The
following section discusses effect sizes as a measure of
knowledge in marketing research, presents several models
from sociology of science that describe how knowledge
progresses (i.e., develops over time), and develops a categorization of subject areas in marketing research. The metaanalytic data and meta-meta-analysis method that were used
for the analysis are presented, followed by the findings. The
article concludes with a discussion of the interpretations
and implications of the findings.
Effect Sizes as a Measure of

Scientific Knowledge
A major function of science is to produce explanations of

how the world works by building scientific knowledge
(Hempel 1965; Van Fraassen 1980). Scientific explanations
connect two or more concepts, usually in a causal manner
(Schmidt 1992; Stewart and Zinkhan 2006). Thus, the value
of scientific knowledge can be characterized by the
explanatory power of statements (i.e., hypotheses and
theories) that link concepts together.
A way to empirically assess explanatory power is by
examining the effect size, which provides empirical evidence on how strongly two variables are related or depend
on each other. Effect sizes indicate the merit of scientific
explanations and the usefulness and relevance of a scientific hypothesis and theory (Aguinis et al. 2011): the more
the variance in the dependent variable is explained, the
more useful and relevant the underlying theory is thought to
be, and the more valuable the knowledge generated by the
theory. In addition to theoretical relevance, effect sizes also
have practical relevance, which is important for an applied
field of inquiry such as marketing. Combs (2010, p. 11)
explains the theoretical and practical relevance of effect
sizes: A theory might find support, but its explanatory
powerthat is, the effect size observedis so weak that
further efforts to develop the theory might not be warranted.... Small effects also raise questions about managerial relevance.... If managers begin to act on theories that
are supported by small effects, they are not likely to notice
positive results even when they occur.
24 / Journal of Marketing, May 2015
The value of scientific knowledge depends not only on

the magnitude of an effect but also on the generalization of
that effect. Meta-analytic effect sizes provide a high degree
of generalization. Meta-analysis is a means of combining
effects sizes in multiple studies. By relying on large samples and repeated tests, meta-analysis averages out sampling error deviations from correct values and can correct
mean values for distortions due to measurement error and
other artifacts (Schmidt 1992). Therefore, a meta-analytic
effect size serves as a quantifiable and generalizable measure of the merit of scientific explanations and the value of
scientific knowledge.
The Progress of Scientific

Knowledge
How does scientific knowledge develop over time?

Although philosophers of science have different views on
the trajectory of scientific progress (Anderson 1983), there
is broad consensus that science is cumulative. Pieces of
knowledge build on one another because all contributions
to research yield some cognitive benefit that is assessed
before publication by referees (i.e., through peer review)
and after publication by colleagues (i.e., by means of citations). The accumulation of knowledge suggests a model of
continuous growth that has often been associated with the
natural sciences: each contribution continues the knowledge
of prior contributions (Crane 1972). Several proponents in
the sociology of science have provided arguments for a discontinuous growth. The major proponent of this view,
Thomas Kuhn (1962), has suggested that science progresses
through revolutions. In phases of normal science,
researchers follow an accepted paradigm, which leads to
continuous accumulation of knowledge. If the paradigm is
discredited, a crisis begins and knowledge production slows
down or even stops until a new paradigm is accepted and
the field resumes its growth. Several researchers in the sociology of science have attested to the discontinuous model.
Laudan (1977) has proposed that science is an evolving
process and that mere collection of empirical evidence does
not really advance science; rather, science is advanced by
the occasional resolution of anomalies provided by various
theories. Lakatos (1977) has suggested that researchers
work within research programs (i.e., a sequence of theories
within a domain of inquiry) and that a research program can
progress with supportive evidence or by modification of
theories, but it can also degenerate if evidence is not supportive and modification of theories does not sufficiently
explain new phenomena.
This study suggests that the explanatory power of statements that link concepts with one another characterizes the
value of scientific knowledge and that the progress of scientific knowledge can be assessed by variations in explanatory power over time. In line with the sociology of science
literature, different models of scientific progress of a
research field (here, marketing) and its subject areas (e.g.,
advertising, consumer behavior) apply: continuous growth,
discontinuous growth, and stasis. The models can be identified and assessed by the variation of effect sizes, which fol-
lows a linear trend, takes the form of a logistic or quadratic

curve, or remains unchanged.
Scientific Progress Is Continuous
Scientific progress is continuous if the research field starts

with small ideas, weakly developed research methods, and
findings of low reach. Additional empirical research
increases the scope of explanations. As a result, the first
findings are low in explanatory power, but this power
increases with each additional finding. This model is often
associated with the natural sciences (Cole 1992; Crane
1972) but has also been found for specific research topics in
marketing such as research on market orientation (Goldman
and Grinstein 2010). The continuous increase, as measured
by effect sizes, is not necessarily linear but rather follows a
logistic curve of exponential growth (Price 1986). Such
exponential growth has been suggested for publication
activities in marketing (Wilkie and Moore 2003), which has
attracted more researchers from different disciplines and
varying geographic areas, leading to higher productivity
and possibly also to an increase in explanatory power.
Scientific Progress Is Discontinuous
As noted previously, the sociology of science proposes a

discontinuous growth model. The simplest deviation from a
continuous growth model can be described by a quadratic
function of the variation of effect sizes that illustrates how
growth can level off. Scientific knowledge first increases by
adding explanatory power to an initial finding; eventually, it
finds a saturation level, after which further research findings provide decreasing contributions, and the strength of
the cumulated effect is weakened by contingency factors or
exhaustion. As described by the idea of research programs
(Lakatos 1977), main and strong contributions are made
when a research field is initiated, and subsequent contributions might provide less additional explanatory power. A
strong main effect found in a seminal article can become
weaker in future articles if additional research identifies and
adds contingency factors. Related to this notion is the concept of exhaustion (Goldman and Grinstein 2010): the
idea that certain problems are exhausted or solved or lose
research attention in the field for other reasons (e.g., the
environment is not interested in the issue). Such patterns
have occurred in research on the diffusion of innovation
paradigm in rural sociology, in which changing policy environments have led to a decline of rural sociological research
on diffusions (Valente and Rogers 1995). The quadratic
function that suggests that research contributions can first
build up before they decline has been shown for citations to
scientific publications in the econometrics literature. These
citations follow a diffusion curve similar to product life
cycles (Fok and Franses 2007).
Scientific Progress Is Static
If no accepted paradigm guides research, a chaotic research

environment evolves in which any problem stands a similar
chance of being selected for research (Crane 1972; Price
1986). Researchers do not build on one anothers work, and
knowledge generation is not continuous, accumulating, or
discontinuous but remains static (i.e., effect sizes remain

unchanged). Although there might be fluctuations in the
growth of specific topics within an area, the overall pattern
is static. Notably, the static model can be temporary and
does not rule out the possibility that a research field will
follow another growth modelfor example, if a strong
paradigm develops or the area becomes attractive for other
reasons.
The changes in research contributions in a field over
time can be related to other processes and characteristics of
research. Two important characteristics are the maturity of a
research topic and the intensity of research on a topic.
Research in a field or subject area (e.g., advertising) typically covers multiple topics (e.g., comparative advertising,
metaphors). Both maturity and intensity refer to the notion
that the explanatory power of findings depends on not only
a general development of the field or the subject area over
time but also how long and how intensively researchers
have dealt with a specific research topic. Both maturity and
intensity can differ over time. For example, an increase in
the number of journal outlets that develop over time enables
researchers to publish more research findings on a particular topic in a shorter time frame than in the past. This
increases the intensity of research on certain topics, which
accelerates the process of knowledge development over
time. As for maturity, each research topic has a life cycle
and attracts different types of researchers and different
types of research at given points in its life cycle. For example, a new research topic might attract younger researchers
who are not yet locked into another, more established
research topic; moreover, it might allow for more innovative research methods if a research tradition has not yet
been established. Thus, maturity and intensity can provide
alternative explanations for the effect of knowledge development as measured by the variation of effect sizes in a
research field or in a subject area.
Subject Areas in Marketing
Marketing is a broad research field with many subject areas.

Researchers have developed several classification schemes
for these subject areas. For example, the Institute for Operations Research and the Management Sciences (2014) has
classified marketing into 18 subjects (e.g., buyer behavior,
industrial marketing, promotion). Journal of Marketing has
classified marketing into five broad subject areas (marketing environment, marketing functions, special marketing
applications, marketing research, and other topics)
(Leonard 2012). For the purpose of this research, eight subject areas from the scheme developed by Cano, Carrillat,
and Jaramillo (2004) were used: advertising, channels, consumer behavior, research methods, new product development, pricing, sales, and strategy.1 Table 1 lists the subject
areas and describes the research focus of each subject
areathat is, the main relationship between concepts that
research in a subject area investigates.
1The subject area diffusion of innovation was not considered,

because only one meta-analysis referred to this field.
Have We Progressed Marketing Knowledge? / 25
Subject Area
TABLE 1
Subject Areas in Marketing Research
Research Focus
Typical Publication Outlets for Primary

Research
International Journal of Advertising

Journal of Advertising
Journal of Advertising Research
Channels
Effect of channel characteristics and relationships between Industrial Marketing Management
Journal of Business to Business Marketing
channel members on channel member responses and
Journal of Retailing
behavior and firm performance
Consumer
Effect of marketing stimuli and consumer characteristics on Journal of Consumer Psychology
behavior
consumer responses
Journal of Consumer Research
Journal of Marketing Research
Method
Effect of data collection and measurements on evaluation International Journal of Research in Marketing
criteria related to analyses and methods
Marketing Science
New product
Effect of product development strategies, processes, con- Journal of Product Innovation Management
development texts, and product characteristics on new product success
Pricing
Effect of price variations and strategies, promotion and
deal characteristics and contexts on consumer responses Journal of Retailing
related to prices
Sales
Effect of salesperson characteristics, sales contexts, and
Journal of Personal Selling and Sales
tasks on sales, salesperson performance, and consumer
Management
responses
Strategy
Effect of firm characteristics, strategies, tactics, and
International Journal of Research in Marketing
contexts on customer responses and firm performance
Journal of Marketing
Marketing Science
Advertising
Effect of advertising characteristics/stimuli on consumer

responses
Notes: Typical publication outlets were selected on the basis of their occurrences in the meta-analyses in each subject area.
This classification scheme is debatable (as is any other

subject classification), but it is instrumental in studying
research in different subject areas in marketing for several
reasons. First, the concepts and the relationships between
concepts in each subject area have a common research
focus (see Table 1) that enables a broad but comprehensive
interpretation of a combined effect size. For example, the
findings in the subject area of new product development
refer to the explanation of new product success using products, product development processes, strategies, and contexts. A combined effect size provides an interpretation of
the extent of explanation of new product success provided
by these concepts. Second, the research in each subject area
is sufficiently distinct from research in other subject areas,
as indicated by specialized journals in most of the areas.
These journals allude to a confined academic community,
which shares interests in the same topics and follows the
same research approaches in a subject area. Third, the classification scheme is neither too detailed nor too broad. It
allows scholars to analyze a substantial number of metaanalytic effect sizes per subject area. This is important for
the kind of analysis (regression) that is performed here to
test the progress of scientific knowledge in each subject
area.
Method
To investigate the amount and development of knowledge

in marketing and its subject areas, a meta-meta-analysis is
performedthat is, a meta-analysis of meta-analyses on

marketing topics. Meta-meta-analyses have been performed
previously in other research fields, such as psychology (e.g.,
Barrick, Mount, and Judge 2001; Lipsey and Wilson 1993;
Richard, Bond, and Stokes-Zoota 2003), but not in business
research. A meta-meta-analysis is also called a second-order
meta-analysis, which synthesizes the findings of traditional,
first-order meta-analyses. The methods of second-order
meta-analyses are a straightforward generalization of firstorder meta-analysis methods to the analysis of mean effect
size estimates (i.e., meta-analytic effect sizes) across multiple meta-analyses (Schmidt and Oh 2013). A second-order
meta-analysis combines these mean effect sizes across
meta-analyses of interest without accessing the primary
studies. The method enables researchers to model the
between-meta-analyses variance that can be used to assess
homogeneity (i.e., whether the variation between metaanalyses findings are homogeneous). If the second-order
sampling error variance fully accounts for the total variance, the meta-analytic effect sizes are homogeneous. If
they are heterogeneous, appropriate moderator variables
can explain the variation between meta-analytic effect sizes
and reduce heterogeneity.
Identification of Meta-Analyses
To locate published meta-analyses in marketing research,

the following steps were performed: (1) electronic databases (e.g., ABI/INFORM, EBSCO) were searched by
using keywords (metaanaly*, meta-analy*, quantita-
tive review, synthesis, and generalization, combined

with marketing or consumer), (2) articles on the Internet (using Google Scholar) were browsed to retrieve unpublished meta-analyses, (3) journals that turned out to be
major outlets of meta-analyses were systematically
searched,2 (4) the proceedings of major marketing conferences were searched,3 and (5) lists of meta-analyses that
had been compiled by others were consulted (e.g., Cano,
Carrillat, and Jaramillo 2004; Peterson 2001).
To establish meaningful inclusion criteria, definitions of
both what a meta-analysis is and what qualifies as a marketing topic are needed. This research defines a meta-analysis
as follows (see also Richard, Bond, and Stokes-Zoota
2003): the study must report a numerical measure of a relationship between two variables (e.g., correlation, standardized mean difference) and must systematically summarize
evidence of this effect collected within two or more primary
scholarly studies by two or more researchers or research
teams. This definition resulted in exclusions of several studies that are sometimes referred to as meta-analytic. First,
review studies that did not report numerical measures for a
relationship between two variables but rather presented
descriptive results on a single variable such as vote-counting,
percentages, mean values, and validity coefficients were
excluded. Second, studies that compiled effect sizes from
various data sets from a single practitioner source were
excluded because these studies lack comparability with
scholarly meta-analyses and do not provide information to
code all independent variables (e.g., Lodish et al. 1995).
Third, studies that combined raw data from previous studies
instead of effect size measures were excluded. Fourth, some
studies that were published as abstracts only and for which
further information could not be retrieved from the authors
for various reasons (e.g., the study is still a work in
progress) were excluded.
If a meta-analysis was published more than once, more
than one meta-analysis was included only when one of the
meta-analyses provides additional results. Otherwise, the
meta-analysis that was published first was used; if the information between both meta-analyses differs, the one that
provides more information was used (e.g., a journal publication that followed a conference abstract publication). If
other authors reanalyzed the same meta-analytic data (e.g.,
data from Peterson, Albaum, and Beltramini [1985] were
reanalyzed by Rust, Lehman, and Farley [1990] and by
2The journals are European Journal of Marketing, International

Journal of Advertising, International Journal of Research in Marketing, Journal of Advertising, Journal of Business Research,
Journal of Consumer Psychology, Journal of Consumer Research,
Journal of Marketing, Journal of Marketing Research, Journal of
Product Innovation Management, Journal of Public Policy &
Marketing, Journal of Retailing, Journal of Service Research,
Journal of the Academy of Marketing Science, Marketing Letters,
Marketing Science, and Psychology and Marketing.
3The conferences are the meetings organized by the Association
for Consumer Research, American Marketing Association, European Marketing Academy, and INFORMS Marketing Science
Society.
Kayande and Bhargava [1994]), the primary meta-analysis

was included.
As for defining topics in marketing, all meta-analyses
that address marketing issues were includedthat is, articles that broadly refer to the activity, set of institutions,
and processes for creating, communicating, delivering, and
exchanging offerings that have value for customers, clients,
partners, and society at large (American Marketing Association 2013). For this purpose, all meta-analyses that were
published in traditional marketing outlets as well as unpublished meta-analyses on marketing topics were considered.
Journals were categorized as marketing journals following
the comprehensive journal classification provided by Harzings (2012) list, which is in line with other marketing journal lists in the literature (e.g., Lehmann 2005). In addition,
meta-analyses that had not appeared in the usual publication outlets were considered if they addressed topics that
qualified as marketing topics. They qualified as such if the
majority of articles (more than 50%) included in the metaanalyses came from marketing journals. If fewer than 50%
of the articles included in a meta-analysis were from marketing journals, three expert coders made a final decision on
whether the topic of the meta-analysis qualified as a marketing topic. If fewer than 10% of the articles were from
marketing journals, the meta-analysis was considered as
focused on a nonmarketing topic and was not included.
After this procedure, 192 manuscripts that were available
by the end of 2012 were retrieved for possible inclusion in
this study.
Coding of Meta-Analytic Effect Sizes
At least one meta-analytic effect size was selected from

each meta-analysis. To achieve the broadest generalization,
the most highly aggregated effect was selected. In some
cases, the meta-analysis provides only one meta-analytic
effect size. For example, Elliss (2006) meta-analysis provides one aggregated effect size for the relationship
between market orientation and business performance.
Other meta-analyses have combined findings from primary
studies into several meta-analytic effect sizes rather than
into a single meta-analytic effect size: for example, a metaanalysis on the effects of comparative advertising has provided several meta-analytic effect sizes for the relationship
between comparative advertising and attitudes, intentions,
behavior, and so on (Grewal et al. 1997). These meta-analytic
effect sizes were coded separately.
Because most of the meta-analyses in marketing used
the correlation coefficient as the effect size, it was chosen
as the meta-analytic effect size for this study as well;
absolute values were coded because this study pertains to
the size of the effect rather than its direction. If the metaanalysis applied an effect size metric other than correlation
coefficients (e.g., standardized mean differences), effect
sizes were transformed to correlation coefficients according
to common recomputation methods provided in the literature (e.g., Borenstein et al. 2009). The raw mean effect size
was used because this measure was provided in most cases.
If only the sample sizeweighted mean or varianceweighted mean was provided, this measure was used
because no difference between raw mean and weighted

mean values in meta-analyses that provided both estimates
were found (t(1, 172) = .24, p = .81). Some meta-analyses
provided attenuated effect sizes only. The meta-analyses
that provided both attenuated and unattenuated effect sizes
revealed that attenuation-corrected estimates are larger than
raw means (t(1, 91) = 5.88, p < .001) and weighted means
(t(1, 363) = 13.71, p < .001) because the correction factor
increases the unattenuated effect sizes. The average ratio of
unattenuated effect sizes to attenuated effect sizes was .866.
This value was used to correct the estimates from metaanalyses that provide attenuation-corrected estimates only
(i.e., these estimates were multiplied by .866).
Coding of Independent and Control Variables
The main independent variable is time, which explains the

size of the effect as a function of the development of marketing research over time. This development was analyzed
for the research field overall (i.e., for all effect sizes) as well
as for eight subject areas separately, provided that the metaVariable (Symbol)
Effect size (ES)
TABLE 2
Variables Used in the Analysis
Description/Coding
Meta-analytic correlation
(absolute values)
Development of
The average publication year of
research over
the studies that are included in
time (Time)
the meta-analysisa
Maturity of research Time frame between the oldest
topic (Maturity)
and most recent study that was
included in the meta-analysis
Intensity of
Effect sizes per year, calculated
research efforts
by the number of effect sizes
(Intensity)
divided by the time frame of studies included in the meta-analysis
Subject area
Subject areas (dummy variables):
advertising, channels, consumer
behavior, method, new product
development, pricing, sales, and
strategy (taken from Cano, Carrillat, and Jaramillo 2004)
Journal in which the meta-analysis
Journal type
is published (1 = top journal,b
(Journal)
0 = other)
Manuscript status
(Manuscript)
Method type
aThe
analytic effect sizes for these subject areas are heterogeneous. In addition to these variables, several other variables
were coded to control for possible confounding effects.
These variables were either suggested in the literature (e.g.,
journal type, manuscript status, method type; see, e.g.,
Richard, Bond, and Stokes-Zoota 2003) or included to rule
out alternative explanations for the effects of time on effect
sizes (maturity and intensity). Table 2 provides an overview
of the coding of the variables used in the analysis. The table
describes these variables, explains why they were considered (i.e., their function), and provides relevant data characteristics. The choice of variables used for the regression
analysis is necessarily restricted to variables that can be
coded from the meta-analyses and should be restricted to
those that theory or logic indicates could affect the metaanalytic effect size. Meta-analyses provide information on
primary studies on an aggregated level only, and few characteristics on this level can be coded consistently from all
meta-analyses. This explains the restriction to three major
Function
Dependent variable
Explains the size of the effect as a

function of the general development
and progress of knowledge over time
function of the maturity of the
research topic
function of the intensity of research
on the research topic
Data Characteristics
Min = .00, max = .91,

M = .24, SD = .17
Min = 1957, max = 2008,
M = 1991.61, SD = 9.93
Min = 4, max = 78,
M = 24.59, SD = 12.34
Min = .18, max = 218.17,

M = 13.87, SD = 26.79
Explains the size of the effect by dif- Advertising (294 effect sizes),
ferent subject areas
channels (274), consumer
behavior (457), method (56),
new product development
(238), pricing (57), sales
(386), strategy (231)
Controls for the effect of publication Top journal (431 effect sizes),
outlet. Meta-analyses with strong
other (1,410 effect sizes)
results are published in high-impact
journals (Tierney, Clarke, and Stewart
2000)
Controls for publication bias
Yes (921 effect sizes), no
Meta-analysis includes unpub(inclusion of unpublished studies
(920 effect sizes)
lished studies (1 = yes, 0 = no)c
deflates the magnitude of the effect
size [Sharpe 1997])
Controls for the possibility of conMainly nonexperimental
Method approach of the majority
founding effects in nonexperimental (1,449 effect sizes), mainly
of the studies included in the
experimental (392 effect
meta-analysis (1 = nonexperimen- studies that can lead to weaker
tal, 0 = experimental)
effects (Field and Hole 2003)
sizes)
list of studies of 20 meta-analyses could not be retrieved. For these meta-analyses, the mean value was computed using information provided on the time difference between the publication year of the oldest and most recently published study in the meta-analysis. On average,
the mean is located after 66% of the time has passed since the publication of the oldest study; the median is located after 70% of the time has
passed.
bTop journals are Journal of Consumer Research, Journal of Marketing, Journal of Marketing Research, and Marketing Science.
cAcross all meta-analyses that provide a list of studies, on average 6.5% of the studies in a meta-analysis were unpublished.
and presumably influential control variables in addition to

the main independent variables.
Two coders independently assigned all meta-analytic
effect sizes to eight subject areas. An effect size can be
assigned to more than one area. The intercoder agreement
was 95%, and disagreements were resolved after discussion.
The remaining variables are based on coding that did not
leave room for interpretation; these variables were doublecoded by a second coder for only 25% of the meta-analyses,
with a 100% agreement rate in coding. Table 3 provides
some descriptive information on the meta-analyses in each
subject area. Following the definition of subject areas as
provided in Table 1, Table 3 shows how often certain categories of independent and dependent variables occur per
subject area. Table 3 also provides an idea of the time
frames captured by each subject area and, thus, information
on the development and maturity of research in each subject
area. For example, strategy is a relatively new subject area
according to both the publication years of the meta-analyses
and the primary studies, whereas method is a more established subject area. The longest research tradition, according to primary studies, is sales.
In the final analyses, only 176 meta-analyses (providing
1,841 meta-analytic effect sizes) were included. Sixteen
meta-analyses were excluded because the effect size in the
meta-analysis (e.g., elasticities, willingness-to-pay measures) could not be readily transformed into a correlation
coefficient on the basis of the information provided in the
meta-analysis; for example, transforming the elasticity
requires the mean values for both variables involved, which
are not usually provided in a meta-analysis. Included and
excluded meta-analyses were compared regarding the main
variables in Table 2. The meta-analyses do not differ in
terms of the average publication year of the studies
included in a meta-analysis (t(1, 196) = 1.58, p = .12) or in
terms of the time frame of the studies (t(1, 196) = 1.00, p =
.32), manuscript status (c2(1) = .12, p = .73), or method
type (c2(1) = 1.73, p = .19). However, the excluded studies
are more likely to be published in top journals with a
greater number of effect sizes per time frame, and they differ across subject areas. This is because the meta-analyses
that apply elasticity-based effect sizes come from particular
subject areas (in particular, pricing and advertising) and are
often published in Journal of Marketing Research. This difference is considered when interpreting the findings.
The 176 meta-analyses included more than 7,500 primary studies (43.15 studies per meta-analysis) published
between 1918 and 2012 and more than 54,000 effect sizes
(307.34 effect sizes per meta-analysis). The 131 metaanalyses that provided information on the sample sizes of
the primary studies indicate an average of 63,641 subjects
per meta-analysis and a total of 8,337,096 subjects. Assuming the same sample size for the remaining meta-analyses,
the 176 meta-analyses are based on an overall sample of
more than 11 million subjects.
scientific progress, different functional forms were estimated: a linear function (for continuous and linear
progress), a logistic or growth function (for continuous and
nonlinear progress), and a quadratic function (for discontinuous progress). The model of static progress is captured
by a nonsignificant effect. The analytical procedure is as
follows. First, the time variable was regressed on all other
independent variables and the residuals were saved. The
residuals then represent the variance in the given independent variable after all other independent variables are controlled. The residuals were used as the independent variable
to predict the effect size under any of the suggested models.
In addition, the functional form was explored by applying locally weighted scatterplot smoothing, a nonparametric
regression method. The procedure fits simple models to
localized subsets of the data to build a function that
describes the deterministic part of variation in the data point
by point. Although the procedure does not provide a test to
identify the best-fitting functional form, the smoothed curve
allows for an interpretation of the best-fitting function.
After the appropriate functions were identified, a
regression model was applied in which all effect sizes were
regressed on all dummy variables for subject area as predictors in addition to time, maturity, intensity, journal, manuscript status, and method. In a second step, the functional
term of the time variable was added to check for nonlinear
effects. In addition, the same regression models were estimated for each subject area separately. Before running the
regression models, all relevant conditions for the regression
models were checked (normal distribution of residuals,
homoskedasticity, and no multicollinearity). The time
variable was mean-centered before the functional term was
added to reduce collinearity problems. The model specification for all effect sizes is as follows:
The effect sizes were modeled as a linear function of the

independent variables. To explore the different models of
Figure 1 presents the frequency distribution of the observed

1,841 meta-analytic effect sizes (absolute values). The
Analytical Procedure
ESi = 0 +
subject area + time +

i =1
10 f
( time )i
ESi + 11maturityi + 12 intensity i + 13 journali
ESi + 14 manuscript statusi + 15 method typei + ei ,
where ESi is the meta-analytic effect size, b0 is a constant,

and bk are parameter estimates of the determinants detailed
in Table 2.
The model specification for each subject area reduces to
ESi = b0 + b1timei + b2f(time)i + b3maturityi + b4intensityi
+ b5journali + b6manuscript statusi + b7method typei +ei.
Table 4 provides the correlation matrix of the variables. For

correlation matrices for each subject area, see the Web
Appendix.
Results
How Much Do We Know About Marketing? The

Meta-Meta-Analytic Effect Size
457
Consumer
behavior
18
Dependent Variables
(Number of Effect Sizes)
Message (88), receiver (119), situation (33), Behavior (42), cognitions (50), credibility
source (54)
(21), emotions (14), memory (19), persuasion (104), processing (26), other (18)
Environment (26), investments (23), partner Commitment (34), conflict (10), functionalcharacteristics (23), partnership cooperation ity and norms (32), performance (30),
relationship quality (47) satisfaction (30),
(26), partnership inequality/dependence
(49), relationship quality (76), structure (47), trust (42), uncertainty and opportunism
other (4)
(45), other (4)
Attitude/evaluation (10), behavior (114),
Brand characteristics (32), consumer
demographics/biographical information (38), emotions (8), intention (115),
judgment/decision (28), memory (4),
perceptions/beliefs/behavior (215),
perceptions (60), relationship quality (64),
consumer psychographics (33), marketing
communication (51), other marketing stimuli satisfaction (44), other (10)
(75), other (13)
Measurement (29), other research
Measurement quality (38), response rates
characteristics (6), sampling (6), survey
(16), other (2)
characteristics (15)
Marketplace/environment (30), organization Innovation (10), market orientation (7),
(37), process (122), product characteristics new product development speed/efficiency
(26), strategy (23)
(38), new product performance/success
(153), organizational functions (23), other
consumer responses (7)
Price variations and strategies (30), product Demand (1), perceived savings (35), price
(5), promotion and deals (16), store (5),
perceptions (16), price recall (2)
other (1)
Salesperson biographical information (26),
Adaptive selling (19), customer responses
aptitude (68), experience (20), motivation
(25), sales performance (163), sales(62), skill level (108), role perceptions (40); person responses (155)
organization/environment (27), task characteristics (11)
Customer response (64), employee
Consumer relations (49), environment/
response (15), firm operations (33), firm
marketplace (11), firm processes (44), firm
performance (48), firm value (16), market
structure (13), investments/resources (22),
orientation (23), strategic outcomes (27),
marketing instruments (47), relations with
other (5)
other businesses (10), strategies/strategic
orientation (35)
Independent Variables
(Number of Effect Sizes)
19182012
(M = 1990)
19612010
(M = 1998)
20022012
(M = 2007)
19502010
(M = 1989)
19572011
(M = 1998)
19312007
(M = 1983)
19502012
(M = 1996)
19852012
(M = 2006)
19872012
(M = 2001)
19902012
(M = 2007)
19852011
(M = 1995)
19812012
(M = 2006)
19672009
(M = 1993)
19392012
(M = 1992)
19852011
(M = 2004)
19762012
(M = 2005)
Publication
Years of Primary
Studiesb
Publication
Years of MetaAnalysesb
add up to 176/1,841 because some meta-analyses provided meta-analytic effect sizes that were assigned to more than one subject area.
and most recent meta-analysis/primary study and the mean publication year of meta-analyses/primary studies per subject area. For additional information regarding the computation of a mean publication year, see Table 2, footnote a.
231
Strategy
20
16
14
65
12
41
Number
of MetaAnalysesa
aThe number of meta-analyses/effect sizes does not

bFigures given are the publication year of the oldest
362
57
238
Sales
Pricing
New
product
development
56
274
Channels
Method
294
Number
of Effect
Sizesa
Advertising
Subject
Area
TABLE 3
Descriptive Information of Meta-Analyses per Subject Area
1. Effect size
2. Advertising
3. Channels
4. Consumer
behavior
5. Method
6. New product
development
7. Pricing
8. Sales
9. Strategy
10. Time
11. Maturity
12. Intensity
13. Journal
14. Manuscript
15. Method type
Variable
.050
.216
.161
.083
.007
.168
.143
.234
.733
.077
.168
.129
.021
.026
.063
.206
.005
.031
.088
.011
.084
.073
1
.182
.251
1
.051
.005
.160
.075
.207
.158
.039
.018
.160
.037
.122
.217
.074
.161
1
.240
.059
.256
.067
.150
.081
.124
.012
.081
.082
.008
.214
.032
.080
.067
.234
.010
.135
.066
.051
.077
1
.068
.069
.191
.146
.197
.202
.141
.079
.217
.200
1
.088
.068
.108
.105
.053
.084
.066
.053
1
.167
.176
.377
.359
.049
.033
.231
TABLE 4
Correlation Matrix
.184
.155
.006
.402
.244
.197
1
.417
.039
.139
.119
.004
10
1
.135
.202
.178
.047
11
.002
.093
.012
12
.220
.175
13
1
.146
14
15
1.0
success). The mean effect size in this meta-meta-analysis

translates into a BESD of 23.6. If half of this figure is added
to and subtracted from 50%, the increase in success rate
ranges from 38% to 62%. That is, by applying more knowledge in marketing, market behavior can change and/or marketers success can increase by 24%.
The meta-meta-analytic effect size differs across subject
areas (Table 5), with the lowest effect size found for method
and new product development; pricing has the strongest
effect size, followed by consumer behavior. The BESD success rate varies accordingly and reaches only 20% in new
product development but 36% in pricing. Table 5 further
indicates the proportion of second-order sampling error to
total variance that is used to assess the heterogeneity of
effect sizes. The second-order sampling error variance
accounts for a maximum of 25% of the total variance. Thus,
heterogeneity can be assumed and a moderator analysis is
appropriate.
observed mean effect size (i.e., the meta-meta-analytic

effect size) is .236. This effect size indicates less than 6%
explained variance and leaves more than 94% of the variance unaccounted for. Although such a result might seem
low in practical terms, another way of viewing it is the success rate equivalent as provided by the binomial effect size
display (BESD; e.g., Rosenthal and Rubin 1982). The
BESD is a convenient way to estimate the practical difference one makes by applying the knowledge a research field
produces. The BESD illustrates the practical importance of
an effect as the difference in outcome rates between two
groupsusually an experimental treatment and a control
group, but the idea can be applied to any type of independent variable that is or can be (artificially) dichotomized. As
such, research has recommended the BESD for assessing
the practical importance of findings in management literature (e.g., Eden 2002). For example, the BESD can describe
the relationship between market orientation and success as
the difference in success if half the companies have a strong
market orientation and the other half have a weak market
orientation. If there is no difference, both groups are
assumed to have a success rate of 50% (i.e., the base rate of
Table 6 provides the results for the curve estimation. The

time variable does not show any effects for pricing and
sales. For all other models, the quadratic term explains
more variance than the exponential growth function. The
locally weighted scatterplot smoothing estimation (see the
Web Appendix) largely supports the functional form suggested by these models.
Table 7 provides the findings of the regression model
for all effect sizes. After examining the residuals, robust
standard errors are applied because the BreuschPagan test
for heteroskedasticity is significant (c2 = 34.03, p < .01).
The quadratic term of the time variable shows a significant
effect. Figure 2, Panel A, depicts the relationship between
time and effect size. The findings provide an answer to the
main research question: knowledge in marketing has progressed over the years, but the increase follows a discontinuous patternthat is, the progress slows down.
Table 8 provides the findings of the regression models
for each subject area. Robust standard errors are applied for
channels, sales, and strategy, because the BreuschPagan
test for heteroskedasticity is significant (all p < .05). Significant effects of the quadratic term are found for advertis-
FIGURE 1
Frequency Distribution of Meta-Analytic Effect
Sizes (Absolute Values)
150
Frequency
100
50
0
.2
.4
.6
Effect Size
.8
How Has Marketing Knowledge Developed? The

Influence of Time on the Meta-Analytic Effect Size
TABLE 5
Meta-Meta-Analytic Effect Sizes (Correlations)
All effect sizes

Advertising
Channels
Consumer behavior
Method
New product development
Pricing
Sales
Strategy
Number of
Meta-Analytic
Effect Sizes
1,841
294
274
457
56
238
57
362
231
Mean r
.236
.215
.238
.284
.154
.203
.361
.228
.248
Median r
.200
.173
.202
.260
.075
.199
.332
.205
.220
Explained
Variance (%)
5.560
4.640
5.657
8.069
2.372
4.111
13.032
5.219
6.128
BESD
(%)
23.579
21.540
23.785
28.405
15.403
20.277
36.099
22.845
24.754
Percentage
Sampling Error
Variance/
Total Variance
.113
.278
25.054
.173
.321
1.339
.009
.010
1.589
TABLE 6
Curve Estimation: Explained Variance of Three
Functional Forms of Time
All effect sizes
Advertising
Channels
Consumer behavior
Method
Pricing
Sales
Strategy
Linear
.003**
.024***
.009
.013**
.202***
.025**
.001
.004
.014*
Logistic
Quadratic Growth
.012***
.025**
.032**
.017**
.259***
.026**
.008
.010
.015
.005***
.011*
.005
.002
.097**
.020**
.004
.006
.009
*p < .10.
**p < .05.
***p < .01.
Notes: The dependent variable is the residual obtained after
regressing the time variable on the remaining independent
variables (see Table 2). The curve estimation regresses the
dependent variable on different functions of the time
variable.
ing, channels, and consumer behavior (see Figure 2, Panels

BD). The relationship for advertising follows a discontinuous life cycle model. The relationship for channels shows
that the effect sizes decrease but that the decrease levels off.
The relationship for consumer behavior indicates a nonlinear trend, in which research first levels off and later
increases. These findings show that knowledge accumulation develops differently across subject areas.
Additional Findings of the Regression Analysis
The regression models provide further notable findings. In

the overall model, the subject area variables support the pattern of meta-meta-analytic correlations depicted in Table 4.
Method provides significantly weaker effect sizes than the
other variables. As expected, maturity slows down the
research contribution, whereas intensity does not have an
effect. Maturity shows the same deceleration effects in
advertising, consumer behavior, and sales. Intensity shows
varying effects across subject areas; it increases effect sizes
in pricing and sales but decreases them for channels and
strategy.
As for the three control variables, the effect of journal
has a negative sign that is significant for channelsthat is,
the effect sizes in top journals tend to be weaker. The effect
sizes in top journals are stronger for strategy and new product development. They are the youngest subject areas in
terms of publication dates of primary studies, and the
increasingly strong competition for top journal space in
recent years can explain why only strong effect-size studies
have made it into the top journals in these subject areas.
The field of channels has witnessed decreasing effect sizes
over time, and the shift toward publishing in top journals in
recent years can explain why the effect sizes in top journals
are weaker compared with other journals that have published stronger effects in the past. As for manuscript status,
the findings show that in the overall model, for channels,
new product development, and sales, the inclusion of
TABLE 7
Variance in All Effect Sizes: Regression Estimates
(Predictor) Variable
Constant
Advertising
Channels
Consumer behavior
Method
Pricing
Sales
Strategy
Time
Time2
Maturity
Intensity
Journal
Manuscript status
Method type
R2
F
.262***
.064**
.079***
.117***
.054*
.017
.143***
.095***
.053**
.299**
.001**
.003***
.001
.011
.030**
.006
.113
16.16***
SE
.026
.026
.023
.042
.028
.023
.031
.022
.021
.129
.001
.001
.001
.011
.009
.017
*p < .10.
**p < .05.
***p < .01.
Notes: Robust standard errors were applied after a significant finding of the BreuschPagan test.
unpublished studies reduces the effect size, whereas for

advertising, the effect size increases. The negative sign was
expected. Variations in the acceptability of weak findings
for journal publications in different subject areas can
explain the finding that some subject areas are unaffected
by manuscript status and that advertising even leads to a
stronger effect. Finally, with regard to method type, no significant differences in effect size surfaced. The advantage
of control in experiments seems to be compensated by stimuli that are more realistic or by strong contrast effects in
field studies.
Robustness Checks
Additional analyses were performed to investigate the

robustness of the results. First, a different measure for the
time variable was used. In the initial analysis, the average
publication year of the studies in a meta-analysis was used
to describe the time the research occurred. The studies
examined in a meta-analysis are not equally distributed
over time but are more likely to be included the closer they
are to the publication date of the meta-analysis. To consider
an uneven distribution, the median instead of the average of
the publication years of the studies in the meta-analysis was
used. The average publication year over all 176 meta-analyses
is 1991.61, while the median is 1992.46. The findings of the
regression models remain stable (same signs and significant
test results).
Second, first-order meta-analyses usually weight effect
sizes according to the underlying sample sizes, giving more
weight to effects from large samples and less weight to
effects from smaller samples. To consider the differences in
sample size in the primary study in the current second-order
meta-analyses, the models were rerun as weighted regres-
FIGURE 2
Relationship Between Time and Effect Size for All Effect Sizes and Effect Sizes in Advertising, Channels,
and Consumer Behavior
B: Advertising
.1
.1
.15
.15
.2
.2
.25
.25
.3
.3
.35
A: Marketing (All Effect Sizes)
1960
1970
1980
1990
2000
$YHUDJH3XEOLFDWLRQ<HDURI6WXGLHVLQWKH0HWD$QDO\VLV
95% CI
2010
1970
Fitted values
1980
1990
2000
95% CI
C: Channels
2010
Fitted values
D: Consumer Behavior
.2
.2
.3
.3
.4
.4
.5
.5
.6
.6
1980
1985
1990
1995
95% CI
2000
Fitted values
sions with the aggregated sample size for each meta-analytic

effect size as a weighting factor. Although in the overall
model, the significance of some of the subject areas disappeared (advertising, channels, and sales), all other variables
showed the same signs and (marginally) significant test
results that resemble the findings of the unweighted regression model.
Finally, because the dependent variable distribution
deviates from normality (see Figure 1) and is negatively
skewed, which can affect the error terms in the model, the
dependent variable was transformed by taking the square
root to achieve normal distribution. Importing the transformed variable into the regression models does not change
the findings.
Additional Analysis: Effect Sizes as Indicators of

Scientific Contribution?
This study assumes that an effect size measures the

explanatory power of scientific statements and indicates the
degree of scientific contribution or the value of scientific
1970
1980
1990
2000
95% CI
2010
Fitted values
knowledge. However, effect sizes can become strong for

different reasons (e.g., sophisticated control and manipulation in experiments). If the effect size is indeed a measure
of scientific contribution and value of scientific knowledge,
academic community recognition should be related to the
strength of the effect size. That is, the stronger the effect
sizes, the greater the scientific contribution and the value of
scientific knowledge and, thus, the more academic community recognition. This assumption is supported by prior
research that shows that meta-analyses with strong results
are published in high-impact journals (i.e., journals that
receive a high number of citations) (Tierney, Clarke, and
Stewart 2000). These data exhibit support for the relationship between effect size and citations, too. The effect sizes
of 102 meta-analyses in this data set (published in journals
listed in the Web of Science database) positively correlated
with the meta-analyses average citations per year (r = .24,
p < .01). That is, the stronger the effect size, the more citations a meta-analysis receives, which indicates the assess-
.262***
.760*
.001*
.004***
.009
.031
.048*
.060
.059
2.57**
.035
.422
.001
.001
.007
.038
.029
.038
.344***
6.167***
.002***
.004
.017**
.105***
.116***
.
.093
7.59***
.043
1.999
.001
.003
.008
.032
.041
SE
SE
Channels
Advertising
.401***
1.277**
.001**
.001***
.001
.019
.020
.026
.147
11.07***
b
.027
.591
.001
.001
.001
.024
.021
.021
SE
Consumer
Behavior
.272***
3.009
.001
.001
.001
.011
.125
.
.246
2.67**
b
.089
2.333
.001
.004
.001
.080
.075
SE
Method
.229***
.571
.001
.001
.002
.117***
.060**
.
.142
5.08***
b
.035
.970
.001
.002
.003
.032
.027
SE
New Product
Development
.388
5.620
.001
.001
.003*
.885
.
.
.111
1.27
SE
.309
9.332
.002
.008
.005
.707
Pricing
.278***
.387
.001
.002**
.012*
.028
.074***
.062
.104
5.96***
SE
.060
.359
.001
.001
.006
.020
.017
.054
Sales
.625***
3.633
.001
.011***
.002***
.081*
.037
.
.257
29.42***
.068
2.683
.001
.001
.001
.043
.050
SE
Strategy
*p < .10.
**p < .05.
***p < .01.
Notes: Robust standard errors were applied after a significant finding of the BreuschPagan test for the following models: channels, sales, and strategy. For channels, new product development, and strategy, the method variable was dropped because it is a constant. Method was dropped for the subject area method because of high collinearity. For pricing, method and
manuscript status were dropped due to high collinearity.
Constant
Time
Time2
Maturity
Intensity
Journal
Manuscript
Method
R2
F
(Predictor)
Variable
TABLE 8
Variance in Effect Sizes per Subject Area: Regression Estimates
ment of higher scientific contribution and value of scientific

knowledge by the academic community.
Discussion
The findings of this study illustrate how much we know in

marketing and how marketing knowledge has progressed.
They provide implications for further research in marketing
and the academic community on how to evaluate and
progress this knowledge.
How Much We Know About Marketing
Previous literature has discussed what we know in marketing and how our knowledge has progressed, resulting in different conclusions. The current study is the first to measure
the value and progress of knowledge in marketing. The estimate in this study shows that the field has developed a considerable body of marketing knowledge that is able to
explain real-world phenomena, as expressed by a mean
meta-analytic correlation of .24. The literature suggests different ways to evaluate the magnitude of such an effect size
(e.g., Durlak and Lipsey 1991). According to Cohens
(1988) classification, a correlation of .24 can be considered
a medium-sized effect. Second, when this effect, found in
an applied discipline (marketing), is compared with effects
found in compilations of meta-analyses in other fields of
inquiry that constitute more basic disciplines with a longerlasting research tradition than marketing, same-sized or
smaller effects are found. Lipsey and Wilsons (1993)
review of meta-analyses on the effectiveness of psychological, educational, and behavioral treatments indicates an
effect size that corresponds to a mean meta-analytic correlation of .24. Richard, Bond, and Stokes-Zoota (2003) find
that social psychology effects in 322 meta-analyses yield a
mean meta-analytic correlation of .21. These figures show
that marketing is a relatively successful research discipline.
Although the mean effect indicates low explanatory
power from a practical point of view (it explains, on average, less than 6% of the variance in the dependent variable
and leaves more than 94% of the variance unaccounted for),
the practical significance of this effect in BESD terms is far
from trivial. The study shows a 24% increase in the dependent variable if we apply the knowledge that is produced by
marketing research. It should be noted that marketing
research focuses on variables that are under marketers control (e.g., price variations, strategic decisions) but neglects
variables that are less likely to be influenced (e.g., habituation, inertia). A great deal of variance in behavior in markets can be attributed to inertia, and previous purchases are
a very strong predictor of future purchases (Corstjens and
Lal 2000; Vogel, Evanschitzky, and Ramaseshan 2008). For
example, Beatty and Smith (1987) show that 40%60% of
consumers buy at the same retailer because of habit. Other
explanations that are out of marketers control are market
characteristics and number of alternatives available. Thus,
when these often-excluded effects are considered, marketing research might actually explain more variance than the
meta-analytic effect sizes suggest.
In conclusion, marketing research provides considerable

predictive power, and the variation that can be influenced
compared with the variation that cannot be influenced is
rather large. Marketing research can thus provide a strong
contribution to practitioner knowledge, but this practical
knowledge must be assessed for different subject areas
rather than the field overall.
How and Why Marketing Knowledge Varies by
Subject Area
The amount of marketing knowledge varies across subject

areas; all have provided explanations for real-world phenomena, but to varying degrees. Whereas some subject
areas have effect sizes that are close to the overall mean
(channels, sales, and strategy), the strongest effects and success rates are found in pricing, followed by consumer
behavior; in contrast, new product development evidences
considerably weaker effects and success. The weakest
effects occur in methods. The variations in successful
knowledge production across subject areas can be explained
by substantial and research-based differences between subject areas.
The substantial differences in effect sizes might be due
to different degrees of complexity of the research object.
For example, the determinants that have been used to
explain new product success are manifold, and each of them
provides low explanatory power (Evanschitzky et al. 2012).
Because of this complexity, research in this field still struggles to explain the high failure rate of new product introductions. Similar complexity has been found in advertising
research (Vakratsas and Ambler 1999), in which scholars
have indicated that advertising in general might have rather
small effects that depend on the many ways advertising
execution takes place and the way the effect is assessed
(e.g., Stewart 1992).
Research-based differences relate to conceptualizations
and manipulations, validity and reliability of measurements,
and the ability to isolate effects. Manipulations in pricing
and consumer behavior can contribute to the strong effects
found in these areas compared with other areas: manipulating prices is more concrete and less error prone than, for
example, manipulating advertising stimuli such as testimonial characteristics (e.g., credibility, attractiveness). Both
areas also show high reliability of dependent variables: consumer behavior research works more often with choice
than, for example, sales or advertising. Choice is a more
concrete and less error-prone dependent variable than commitment or attitudes. Informant bias is another reason for
bias in measurements: in subject areas in which key informants provide data (e.g., new product development, channels, strategy), the effects tend to be weaker than in subject
areas with data provided by individual consumers (e.g.,
consumer behavior, pricing). As for the isolation of effects,
pricing has strong immediate effects (e.g., Nijs et al. 2001)
that are captured in the effect sizes, which are largely based
on immediate measures, whereas a field such as advertising
has long-term carryover effects (Sethuraman, Tellis, and
Briesch 2011), and a field such as new product development has long-lasting performance effects that the effect
sizes might not fully capture. The maturity of different subject areas further supports the research-based differences in
effect sizes: a subject area such as new product development is more recent than others and has not yet reached the
explanatory power of subject areas with more established
methods and measures, such as pricing. Not all researchbased differences lead to variations in effect sizes, however.
The subject areas differ in their use of subjective and objective measures, but the findings do not reveal a systematic
pattern that shows that either measure leads to stronger
effects.
In conclusion, the amount of marketing knowledge
varies by subject areas. The findings by subject area indicate benchmarks for the amount of knowledge produced in
a particular field and the amount of knowledge that future
studies should provide or exceed to offer a meaningful contribution in this area. The complexity and maturity of a
field, the concreteness of concepts, the reliability and
validity of measures, and the ability to isolate effects all
explain the variation in effect sizes.
How Marketing Knowledge Has Progressed
Marketing knowledge has increased over time. As in any

other field of science, knowledge in marketing is cumulative, and over time, researchers succeed in providing better
explanations about real-world phenomena. Marketing as a
discipline (and as an industry) has experienced uneven
changes in different periods, and the evolution of marketing
is often nonlinear or even discontinuous. Although the data
in this study are too broad to investigate specific changes in
different periods, the findings show that the overall development is discontinuous, the increase in knowledge occurs
at a decreasing rate, and marketing knowledge has reached
a stage of maturity. This development meets both substantial and methodological explanations. The more mature a
research field, the less groundbreaking its new findings,
which therefore leads to less increase in knowledge compared with older findings. As for methodological explanations, early research in a discipline might have faced fewer
obstacles to publication and therefore provided weaker
results, but the advancement of methods and demands for
rigorous research might have strengthened the effects
extracted from research, at least up to a certain saturation
level.
The changes over time relate to different eras of marketing thought, as distinguished by Wilkie and Moore (2003).
Most empirical findings were generated during the third
and fourth eras. In era III (19501980), growth in scholarly
research in marketing boomed, and the managerial, behavioral, and quantitative perspectives evolved to become
major knowledge areas. Era IV (1980 to present) is characterized by maturity, strong fragmentation of the mainstream, and diversification into specialized research areas.
The development over time shows that effect sizes have
become weaker in the fourth era. Wilkie and Moore point
out that fragmentation might have disadvantages such as
breakdown of communication across fields and inability to
build on a rich knowledge base and to provide a cumulative
response to important new challenges. These disadvantages
offer another explanation as to why growth in effect sizes

has weakened: increasingly specialized scholars might lack
the comprehensive marketing knowledge needed to craft
research that provides strong results.
The progress of knowledge in different subject areas
varies. Whereas a static progress model finds support in the
subject areas of methods, new product development, pricing, sales, and strategy, discontinuous progress applies to
advertising, channels, and consumer behavior, which indicates a deceleration of knowledge in these areas. Advertising and channels have faced major external challenges due
to the advent of the Internet. As Wilkie and Moore (2003)
suggest, the increasing fragmentation during the most
recent era of marketing thought might make it difficult for a
specialized field to provide answers that require a broad
knowledge base and exchange of information across fields.
As for consumer behavior, the findings show that the deceleration has stopped, and the contributions in this area are
now increasing, making it the only subject area with
expanding knowledge production. Although consumer
behavior is a specialized field that has emerged from fragmentation in marketing, it is a multidisciplinary area that
fosters communication across fields, which might help the
field provide better answers to new challenges. The paradigmatic shift toward naturalistic research and qualitative
methods during the 1980s and 1990s might have led to a
breakdown in communication between consumer behavior
and other fields in marketing and would explain the specific
U-shaped pattern of the development over time. Consumer
behavior now experiences an increasing influence of psychological research that emphasizes strong experimental
controls and exclusion of confounding effects, which might
explain the recent increase in effect sizes. If one takes into
account the finding that research topic maturity slows down
contributions, the recent development in consumer behavior
might also be due to the high variety and innovativeness of
topics inspired by the exchange across disciplinary borders.
Variety and innovativeness ensure high explanatory power
and, at the same time, provide more efficient responses to
new challenges compared with other subject areas.
In conclusion, marketing knowledge has increased but
does so at a decreasing rate, and it has reached a stage of
maturity. The current era of marketing research is characterized by specialization and fragmentation, and increasingly
specialized scholars might lack the comprehensive marketing knowledge needed for research that provides strong
results. The subject area of consumer behavior uses the
exchange across disciplinary borders to provide efficient
responses to new challenges, thereby increasing its knowledge base, whereas advertising and channels struggle with
the challenges of fragmentation and specialization.
Implications for Research in Marketing and the
Marketing Community
The finding regarding how much we know about marketing

has implications for the entire marketing community (i.e.,
marketing scholars and marketing practitioners) because the
quantitative estimate provides a unique opportunity to
evaluate and quantify research outcomes. Our field of
inquiry often faces controversies about its low productivity

and lack of accountability (Sheth and Sisodia 2002). Marketing scholars and practitioners benefit from quantifying
their knowledge or success. Scholars can increase marketings reputation, and practitioners can increase their influence within the firm (Verhoef and Leeflang 2009). The
effect size, explained variance, and BESD findings in this
study provide quantitative evidence for productivity and
accountability in marketing because they show the degree
to which the current marketing knowledge explains realworld phenomena and the extent to which it can increase
firms performance.
The results regarding how and why marketing knowledge varies by subject area have implications for further
marketing research. The finding that some areas have lower
explanatory power than others is a reminder for the community to focus on these subject areas because there is still
much to be explained. Consequently, fields such as new
product development and methods are candidates for
research initiatives and funding programs that support
research in these areas. Explanations for variations of effect
sizes in subject areas are provided by the complexity and
maturity of a field, the concreteness of concepts, the reliability and validity of measures, and the ability to isolate
effects. These explanations provide the means to increase
the effect sizes in subject areas with low explanatory power,
for example, by focusing on more concrete concepts.
The findings related to the progress of knowledge over
time show that marketing has reached a stage of maturity.
Knowledge is still increasing, but at a decreasing rate. This
pattern indicates an advanced stage in the knowledge life
cycle of the discipline. Marketers know what to do in different stages of a life cycle: if we are interested in regaining
the trajectory of success, the community needs to think
about a relaunch of the field. We need to search for paradigmatic changes, new topics, and ways to provide better
answers to new challenges. The subject area of consumer
behavior provides a role model for successful knowledge
development despite challenges of increasing specialization. This specialization can benefit knowledge development over time if the field deals with the disadvantages of
specialization in appropriate ways. Engaging in communication and interaction between different areas and disciplines hinders the breakdown in communication across
fields and helps the area build on comprehensive marketing
knowledge, which is needed in crafting research that provides strong results and, thus, better answers to new challenges. Furthermore, the maturity variable shows that new
research topics provide the most knowledge in terms of
effect sizes at initiation, which decreases the more the
research topic becomes established. Thus, progressing
knowledge could also imply stronger specialization and
development of more new research topics at a faster rate.
Although a strong exchange between areas and fields aids in
responding to new challenges, specialized journals or journal sections, conferences, or financial support by associations can strategically support a successful specialization.
The findings provide implications for individual scholars career strategies, especially for less experienced schol-
ars who can choose from multiple research topics to pursue

or can even decide between different areas to join. Scholars careers strongly depend on the contribution they make
in a field. This study suggests that the explanatory power of
an effect size indicates the degree of contribution of
research. Indeed, the effect size is a good indicator for contribution and recognition by the community and is therefore
a driver of scholarly careers because it increases the likelihood of being published in a leading journal and is correlated with the average citations an article receives (Eisend
and Tarrahi 2014). The effect sizes provided in the current
study provide several considerations for marketing scholars strategy for a competitive academic market. First,
mature topics and fields with less concrete measures and
data from sources that tend to be biased lead to weaker
effect sizes and will therefore contribute less to career
development. Second, if areas can be chosen, pricing and
consumer behavior promise strong effect sizes; indeed, the
effect sizes in consumer behavior are still increasing. Third,
for scholars who already work in a particular subject area,
the meta-analytic effect size in a subject area provides a
benchmark of the explanatory power that is considered
normal in this specific area and can be considered a minimum requirement for new studies to provide a substantial
contribution to this area.
Limitations
This study is not without limitations. One concern relates to

the effect size as an indicator for knowledge and knowledge
progress. Although larger effect sizes are indicative of a
fields explanatory ability (in terms of variance accounted
for), they are not necessarily synonymous with progress.
For example, a disciplines focus on simple relationships
(e.g., price changes and sales) may yield high effect sizes
over time, whereas research questions with more conceptual and methodological challenges may have received less
attention. That is, effect sizes contain valuable information,
but they might emphasize simple concepts and relationships
rather than more complex conceptual frameworks and
views. Indeed, the meta-meta-analysis falls short in explaining the variation of effect sizes due to specific theories.
Although this is important to researchers, the contribution
of specific theories often cannot even be assessed in firstorder meta-analyses because they gather and aggregate
empirical results independent of the theory that has been
applied in primary research. Thus, the current study is
unable to explain the variation in effect sizes due to particular theories. Although adding theory differences as another
predictor to the regression model would probably increase
the models explained variance, its exclusion does not affect
the explanation of variance due to subject areas and time.
Another limitation refers to the classification scheme of
marketing subject areas. As described previously,
researchers have developed various schemes, and no consensus has been reached to date. Different classifications
might lead to distinct results. For the current study, a classification scheme was chosen for a combination of both substantial (e.g., consistent research focus in each area that differs from the focus in other areas) and pragmatic (e.g.,
sufficient number of meta-analytic effect sizes per area)

reasons that other classification schemes could not provide.
In particular, an analysis of the time trend would be difficult
in a more detailed classification scheme, while a more general categorization would lead to aggregation of concepts
that cannot be integrated meaningfully.
Related to the classification scheme is the question of
appropriate measures. The effect size usually refers to
absolute measures and describes, for example, the effect of
price changes on consumer response. Companies actions
are often influenced by competitors, and thus, companies
are interested in relative measures (e.g., the effect of price
relative to competitors on responses). Absolute versus relative measures lead to different results (e.g., Albers,
Mantrala, and Sridhar 2010) and have different meanings
and implications for companies (Steenburgh 2007). Subject
areas intended to describe market behavior (e.g., consumer
behavior) benefit from both absolute and relative measures,
whereas in other areas, relative measures (e.g., relative market share in strategy) are often more meaningful. If the
number of first-order meta-analyses that use relative measures increases, a second-order meta-analysis can complement the current meta-meta-analysis and provide a better
picture of relative versus absolute measures in marketing.
Aguinis, Herman, Dan R. Dalton, Frank A. Bosco, Charles A.

Pierce, and Catherine M. Dalton (2011), Meta-Analytic
Choices and Judgment Calls: Implications for Theory Building
and Testing, Obtained Effect Sizes, and Scholarly Impact,
Journal of Management, 37 (1), 538.
Albers, Snke, Murali K. Mantrala, and Shrihari Sridhar (2010),
Personal Selling Elasticities: A Meta-Analysis, Journal of
Marketing Research, 47 (October), 84053.
American Marketing Association (2013), Definition of Marketing, (accessed February 12, 2015), [available at https://www.
ama.org/AboutAMA/Pages/Definition-of-Marketing.aspx].
Anderson, Paul F. (1983), Marketing, Scientific Progress, and
Scientific Method, Journal of Marketing, 47 (Fall), 1831.
Barrick, Murray R., Michael K. Mount, and Timothy A. Judge
(2001), Personality and Performance at the Beginning of the
New Millennium: What Do We Know and Where Do We Go
Next? International Journal of Selection and Assessment, 9
(1/2), 930.
Beatty, Sharon E. and Scott M. Smith (1987), External Search
Effort: An Investigation Across Several Product Categories,
Journal of Consumer Research, 14 (1), 8395.
Borenstein, Michael, Larry V. Hedges, Julian P.T. Higgins, and
Hannah R. Rothstein (2009), Computing Effect Sizes for MetaAnalysis. Chichester, UK: John Wiley & Sons.
Cano, Cynthia Rodriguez, Francois A. Carrillat, and Fernando
Jaramillo (2004), A Meta-Analysis of the Relationship
Between Market Orientation and Business Performance: Evidence from Five Continents, International Journal of
Research in Marketing, 21 (2), 179200.
Cohen, Jacob (1988), Statistical Power Analysis for the Behavioral Sciences, 2nd ed. New York: Academic Press.
Cole, Stephen (1992), Making Science: Between Nature and Society. Cambridge, MA: Harvard University Press.
Combs, James G. (2010), Big Samples and Small Effects: Lets
Not Trade Relevance and Rigor for Power, Academy of Management Journal, 53 (1), 913.
Corstjens, Marcel and Rajiv Lal (2000), Building Store Loyalty
Through Store Brands, Journal of Marketing Research, 37
(August), 28191.
Crane, Diana (1972), Invisible Colleges: Diffusion of Knowledge
in Scientific Communities. Chicago: University of Chicago
Press.
Durlak, Joseph A. and Mark W. Lipsey (1991), A Practitioners
Guide to Meta-Analysis, American Journal of Community
Psychology, 19 (3), 291332.
Eden, Dov (2002), From the Editors: Replication, Meta-Analysis,
Scientific Progress, and AMJs Publication Policy, Academy
of Management Journal, 45 (5), 84146.
Eisend, Martin and Farid Tarrahi (2014), Meta-Analysis Selection Bias in Marketing Research, International Journal of
Research in Marketing, 31 (3), 31726.
Ellis, Paul D. (2006), Market Orientation and Performance: A
Meta-Analysis and Cross-National Comparisons, Journal of
Management Studies, 43 (5), 10891107.
Evanschitzky, Heiner, Martin Eisend, Roger J. Calantone, and
Yuanyuan Jian (2012), Success Factors of Product Innovation:
An Updated Meta-Analysis, Journal of Product Innovation
Management, 29 (S1), 2137.
Field, Andy and Graham Hole (2003), How to Design and Report
Experiments. Los Angeles: Sage Publications.
Fok, Dennis and Philip Hans Franses (2007), Modelling the Diffusion of Scientific Publications, Journal of Econometrics,
139 (2), 37690.
Goldman, Arieh and Amir Grinstein (2010), Stages in the Development of Market Orientation Publication Activity: A Longitudinal Assessment, European Journal of Marketing, 44 (9/10),
13841409.
Grewal, Dhruv, Sukumar Kavanoor, Edward F. Fern, Carolyn
Costley, and James Barnes (1997), Comparative Versus Noncomparative Advertising: A Meta-Analysis, Journal of Marketing, 61 (October), 115.
Hanssens, Dominique M., ed. (2009), Empirical Generalizations
About Marketing Impact: What We Have Learned from Academic Research. Cambridge, MA: Marketing Science Institute.
Harzing, Anne-Wil (2012), Journal Quality List, Forty-Third Edition, (January 1), (accessed January 28, 2012), [available at
www.harzing.com].
Hempel, Carl G. (1965), Aspects of Scientific Explanation. New
York: The Free Press.
Institute for Operations Research and the Management Sciences
(2014), Subject Classification Scheme for the OR/MS Index,
(accessed February 12, 2015), [available at https://www.informs.
org/content/download/16367/184683/file/orsubjec05-06t.pdf].
Kayande, Ujwal and Mukesh Bhargava (1994), An Examination
of Temporal Patterns in Meta-Analysis, Marketing Letters, 3
(2), 14151.
Kuhn, Thomas S. (1962), The Structure of Scientific Revolutions.
Chicago: University of Chicago Press.
Lakatos, Imre (1977), Lakatos: The Methodology of Scientific
Research Programmes: Philosophical Papers Volume 1. Cambridge, UK: Cambridge University Press.
Laudan, Larry (1977), Progress and Its Problems: Towards a
Theory of Scientific Growth. Berkeley: University of California
Press.
Lehmann, Donald R. (2005), Journal Evolution and the Development of Marketing, Journal of Public Policy & Marketing, 24
(Spring), 13742.
REFERENCES
, Leigh McAlister, and Richard Staelin (2011), Sophistication in Research in Marketing, Journal of Marketing, 75
(July), 15565.
Leonard, Myron (2012), Marketing Literature Review, Journal
of Marketing, 64 (March), 14255.
Lipsey, Mark W. and David B. Wilson (1993), The Efficacy of
Psychological, Educational, and Behavioral Treatment: Confirmation from Meta-Analysis, American Psychologist, 48 (12),
11811209.
Lodish, Leonard M., Magid Abraham, Stuart Kalmenson, Jeanne
Livelstoerger, Beth Lubetkin, Bruce Richardson, et al. (1995),
How TV Advertising Works: A Meta-Analysis of 389 RealWorld Split-Cable TV Advertising Experiments, Journal of
Marketing Research, 32 (May), 12539.
Nijs, Vincent R., Marnik G. Dekimpe, Jan-Benedict E.M.
Steenkamp, and Dominique M. Hanssens (2001), The CategoryDemand Effects of Price Promotions, Marketing Science, 20
(1), 122.
Peterson, Robert A. (2001), On the Use of College Students in
Social Science Research: Insights from a Second-Order MetaAnalysis, Journal of Consumer Research, 28 (3), 45061.
, Gerald Albaum, and Richard F. Beltramini (1985), A
Meta-Analysis of Effects Sizes in Consumer Behavior Experiments, Journal of Consumer Research, 12 (1), 97103.
Price, Derek de Solla (1986), Little Science, Big Science ... and
Beyond, 2nd ed. New York: Columbia University Press.
Richard, F.D., Charles F. Bond Jr., and Juli J. Stokes-Zoota (2003),
One Hundred Years of Social Psychology Quantitatively
Described, Review of General Psychology, 7 (4), 33163.
Rosenthal, Robert and Donald B. Rubin (1982), A Simple, General Purpose Display of Magnitude of Experimental Effect,
Journal of Educational Psychology, 74 (2), 16669.
Rust, Roland T., Donald R. Lehman, and John U. Farley (1990),
Estimating Publication Bias in Meta-Analysis, Journal of
Marketing Research, 27 (May), 22026.
Schmidt, Frank L. (1992), What Do Data Really Mean? American Psychologist, 47 (10), 117381.
and In-Sue Oh (2013), Methods for Second Order MetaAnalysis and Illustrative Applications, Organizational Behavior and Human Decision Processes, 121 (2), 204218.
Sethuraman, Raj, Gerard J. Tellis, and Richard A. Briesch (2011),
How Well Does Advertising Work? Generalizations from
Meta-Analysis of Brand Advertising Elasticities, Journal of
Marketing Research, 48 (June), 45771.
Sharpe, Donald (1997), Of Apples and Oranges, File Drawers

and Garbage: Why Validity Issues in Meta-Analysis Will Not
Go Away, Clinical Psychology Review, 17 (8), 881901.
Sheth, Jagdish N. and Rajendra S. Sisodia (1999), Revisiting
Marketings Lawlike Generalizations, Journal of the Academy
of Marketing Science, 27 (1), 71-87.
and (2002), Marketing Productivity Issues and
Analysis, Journal of Business Research, 55 (5), 34962.
Steenburgh, Thomas J. (2007), Measuring Consumer and
Competitive Impact with Elasticity Decompositions, Journal
of Marketing Research, 44 (November), 63646.
Stewart, David W. (1992), Speculations on the Future of Advertising Research, Journal of Advertising, 21 (3), 118.
and George M. Zinkhan (2006), Enhancing Marketing
Theory in Academic Research, Journal of the Academy of
Marketing Science, 34 (4), 47780.
Tierney, Jayne F., Mike Clarke, and Lesley A. Stewart (2000), Is
There Bias in the Publication of Individual Patient Data MetaAnalyses? International Journal of Technology Assessment in
Health Care, 16 (2), 65767.
Vakratsas, Demetrios and Tim Ambler (1999), How Advertising
Works: What Do We Really Know? Journal of Marketing, 63
(January), 2643.
Valente, Thomas W. and Everett M. Rogers (1995), The Origins
and Development of the Diffusion of Innovations Paradigm as
an Example of Scientific Growth, Science Communication, 16
(3), 24273.
Van Fraassen, Bas C. (1980), The Scientific Image. Oxford, UK:
Oxford University Press.
Verhoef, Peter C. and Peter S.H. Leeflang (2009), Understanding
the Marketing Departments Influence Within the Firm, Journal of Marketing, 73 (March), 1437.
Vogel, Verena, Heiner Evanschitzky, and B. Ramaseshan (2008),
Customer Equity Drivers and Future Sales, Journal of Marketing, 72 (November), 98108.
Wilkie, William L. and Elizabeth S. Moore (2003), Scholarly
Thought in Marketing: Exploring the 4 Eras of Thought
Development, Journal of Public Policy & Marketing, 22
(Fall), 11646.
Yadav, Manjit S. (2010), The Decline of Conceptual Articles and
Implications for Knowledge Development, Journal of Marketing, 74 (January), 119.
1
WEB APPENDIX
HAVE WE PROGRESSED MARKETING KNOWLEDGE?

A META-META-ANALYSIS OF EFFECT SIZES IN MARKETING RESEARCH
Martin Eisend
Web Appendix A
Included Meta-analyses
1.
Afef, Amami, Gharbi Jamel-eddine, and Frasson Claude (2010), "Meta-Analysis of

Incidental Advertising on Consumer Choice," International Journal of Business
Research, 10 (2), 172-84.
2. Aguirre-Rodriguez, Alexandra, Michael Bosnjak, and M. Joseph Sirgy (2012),
"Moderators of the Self-Congruity Effect on Consumer Decision-Making: A MetaAnalysis," Journal of Business Research, 62, 1179-88.
3. Amos, Clinton, Gary Holmes, and David Strutton (2008), "Exploring the Relationship
Between Celebrity Endorser Effects and Advertising Effectiveness. A Quantitative
Synthesis of Effect Size," International Journal of Advertising, 27 (2), 209-34.
4. Andrews, Kyle R., Christopher J. Carpenter, Allison S. Shaw, and Franklin J. Boster
(2008), "The Legitimization of Paltry Favors Effect: A Review and Meta-Analysis,"
Communication Reports, 21 (2), 59-69.
5. Argo, Jennifer J. and Kelley J. Main (2004), "Meta-Analyses of the Effectiveness of
Warning Labels," Journal of Public Policy & Marketing, 23 (2), 193-208.
6. Armstrong, J. Scott and Edward J. Lusk (1987), "Return Postag in Mail Surveys: A MetaAnalysis," Public Opinion Quarterly, 51 (2), 233-348.
7. Arts, Joep W. C., Ruud T. Frambach, and Tammo H. A. Bijmolt (2011), "Generalizations
on Consumer Innovation Adoption: A Meta-Analysis of Drivers of Intention and
Behavior," International Journal of Research in Marketing, 28, 134-44.
8. Assmus, Gert, John U. Farley, and Donald R. Lehmann (1984), "How Advertising Affect
Sales: Meta-Analysis of Econometric Results," Journal of Marketing Research, 21
(February), 65-74.
9. Beggan, James K. and Scott T. Allison (1997), "More There Than Meets Their Eyes:
Support for the Mere-Ownership Effect," Journal of Consumer Psychology, 6 (3), 28597.
10. Biswas, Abhijit, Elizabeth J. Wilson, and Jane W. Licata (1993), "Reference Pricing
Studies in Marketing: A Synthesis of Research Results," Journal of Business Research,
27 (3), 239-56.
2
11. Bowen, Frances E., Mahdi Rostami, and Piers Steel (2010), "Timing is Everything: A
Meta-Analysis of the Relationships Between Organizational Performance and
Innovation," Journal of Business Research, 63, 1179-85.
12. Brei, Vincius, Lvia D'Avila, Luis Felipe Camargo, and Juliana Engels (2011), "The
Influence of Adaptation and Standardization of the Marketing Mix on Performance: A
Meta-Analysis," Brazilian Administrative Review, 8 (3), 266-87.
13. Brown, J.R., R. Lusch, and L.P. Smith (1991), "Conflict and Satisfaction in an Industrial
Channel of Distribution," International Journal of Physical Distribution & Logistics
Management, 21 (6), 15-25.
14. Brown, Stephen P., Pamela M. Homer, and J. Jeffrey Inmann (1998), "A Meta-Analysis
of Relationships Between Ad-Evoked Feelings and Advertising Responses," Journal of
Marketing Research, 35 (February), 114-26.
15. Brown, Stephen P. and Robert A. Peterson (1993), "Antecedents and Consequences of
Salesperson Job Satisfaction: Meta-Analysis and Assessment of Causal Effects," Journal
of Marketing Research, 30 (February), 63-77.
16. Brown, Steven P. and Son K. Lam (2008), "A Meta-Analysis of Relationships Linking
Employee Satisfaction to Customer Responses," Journal of Retailing, 84 (3), 243-55.
17. Brown, Steven P. and Douglas M. Stayman (1992), "Antecedents and Consequences of
Attitude Toward the Ad: A Meta-Analysis," Journal of Consumer Research, 19 (June),
34-51.
18. Brudvig, Susan (2007), "The Relationship Between Trust and Commitment: A MetaAnalysis," Society for Marketing Advances Proceedings, 32-36.
19. Calantone, Roger J., Nukhet Harmancioglu, and Cornelia Droge (2010), "Inconclusive
Innovation 'Returns': A Meta-Analysis of Research on Innovation in New Product
Development," Journal of Product Innovation Management, 27, 1065-81.
20. Cano, Cynthia Rodriguez, Francois A. Carrillat, and Fernando Jaramillo (2004), "A
Meta-analysis of the Relationship Between Market Orientation and Business
Performance: Evidence from Five Continents," International Journal of Research in
Marketing, 21, 179-200.
21. Capella, Michael L., Charles R. Taylor, and Cynthia Webster (2008), "The Effect of
Cigarette Advertising Bans on Consumption: A Meta-Analysis," Journal of Advertising,
37 (2), 7-18.
22. Capella, Michael L., Cynthia Webster, and Brian R. Kinard (2011), "A Review of the
Effect of Cigarette Advertising," International Journal of Research in Marketing, 28,
269-79.
23. Carlson, Jay P., Leslie H. Vincent, David M. Hardesty, and William O. Bearden (2008),
"Objective and Subjective Knowledge Relationships: A Quantitative Analysis of
Consumer Research Findings," Journal of Consumer Research, 35 (February), 864-76.
24. Carrillat, Francois A., Fernando Jaramillo, and Jay P. Mulki (2007), "The Validity of
SERVQUAL and SERVPERF Scales," International Journal of Service Industry
Management, 18 (5), 472-90.
25. Carrillat, Francois A., Fernando Jaramillo, and Jay Prakash Mulki (2009), "Examining
the Impact of Service Quality: A Meta-Analysis of Empirical Evidence," Journal of
Marketing Theory and Practice, 17 (2), 95-110.
3
26. Chen, Chien-Chung (2012), "Detecting Moderator Effects on Construct Relationships in
Empirical Sales Research: A Meta-Analysis," Journal of Management and Marketing
Research, 11, 1-18.
27. Chen, Jiyao, Fariborz Damanpour, and Richard R. Reilly (2010), "Understanding
Antecedents of New Product Development Speed: A Meta-Analysis," Journal of
Operations Management, 28 (1), 17-33.
28. Cheon, Hongsik John (2004), "Forty-Four Years of Retrospective International
Advertising and Marketing Research: A Systematic Quantitative Meta-Analysis," PhD
Thesis, University of Florida.
29. Cheon, Hongsik John, Chang-Hoan Cho, and John C. Sutherland (2007), "A MetaAnalysis of Studies on the Determinants of Standardization and Localization of
International Marketing and Advertising Strategies," Journal of International Consumer
Marketing, 19 (4), 109-47.
30. Church, Allan H. (1993), "Estimating the Effect of Incentives on Mail Survey Response
Rates: A Meta-Analysis," Public Opinion Quarterly, 57 (1), 62-79.
31. Churchill, Gilbert A., Jr., Neil M. Ford, Steven W. Walker, and Orville C. Walker, Jr.
(1985), "The Determinants of Salesperson Performance: A Meta-Analysis," Journal of
Marketing Research, 22 (May), 103-18.
32. Clarke, Darral G. (1976), "Econometric Measurement of the Duration of Advertising
Effects on Sales," Journal of Marketing Research, 13 (November), 345-57.
33. Compeau, Larry D. and Dhruv Grewal (1998), "Comparative Price Advertising: An
Integrative Review," Journal of Public Policy & Marketing, 17 (2), 257-73.
34. Conchar, Margy P., Melvin R. Crask, and George M. Zinkhan (2005), "Market Valuation
Models of the Effect of Advertising and Promotional Spending: A Review and Metaanalysis," Journal of the Academy of Marketing Science, 33 (4), 445-60.
35. Costley, Carolyn L. (1988), "Meta Analysis of Involvement Research," in Advances in
Consumer Research, Michael J. Houston, ed. Vol. 15. Provo, UT: Association for
Consumer Research, 554-62.
36. Cox III, Eli P., Michael S. Wogalter, Sara L. Stokes, and Elizabeth J. Tipton Murff
(1997), "Do Product Warnings Increase Safe Behavior? A Meta-analysis," Journal of
Public Policy & Marketing, 16 (2), 195-204.
37. Crosno, Jody L. and Robert Dahlstrom (2010), "Examining the Nomological Network of
Opportunism: A Meta-Analysis," Journal of Marketing Channels, 17 (3), 177-90.
38. ---- (2008), "A Meta-Analytic Review of Opportunism in Exchange Relationships,"
Journal of the Academy of Marketing Science, 36, 191-201.
39. Cruz, Michael G. (1998), "Explicit and Implicit Conclusions in Persuasive Messages," in
Persuasion: Advances Through Meta-analysis, Mike Allen, ed. Cresskill, NJ: Hampton
Press, 217-30.
40. Cugelman, Brian, Mike Thelwall, and Phil Dawes (2011), "Online Interventions for
Social Marketing Health Behavior Change Campaigns: Meta-Analysis of Psychological
Architectures and Adherence Factors," Journal of Medical Internet Research, 13 (1),
e17.
41. Curtis, Tamilla, Russell Abratt, Dawna Rhoades, and Paul Dion (2011), "Customer
Loyalty, Repurchase and Satisfaction: A Meta-Analytical Review," Journal of
Consumer Satisfaction, Dissatisfaction & Complaining Behavior, 24, 1-26.
4
42. Dant, Rajiv P., Audhesh K. Paswan, and Patrick J. Kaufman (1996), "What We Know
About Ownership Redirection in Franchising: A Meta-Analysis," Journal of Retailing,
72 (4), 429-44.
43. Dato-on, Mary Conway and Robert Dahlstrom (2003), "A Meta-Analytic Investigation of
Contrast Effects in Decision Making," Psychology & Marketing, 20 (8), 707-31.
44. De Matos, Celso Augusto, Jorge Luiz Henrique, and Carlos Alberto Vargas Rossi (2007),
"Service Recovery Paradox: A Meta-Analysis," Journal of Service Research, 10 (1), 6077.
45. De Matos, Celso Augusto and Carlos Alberto Vargas Rossi (2008), "Word-of-Mouth
Communications in Marketing: A Meta-Analytic Review of the Antecedents and
Moderators," Journal of the Academy of Marketing Science, 36 (4), 578-96.
46. DelVecchio, Devon, David H. Henard, and Traci H. Freling (2006), "The Effect of Sales
Promotion on Post-Promotion Brand Preference: A Meta-Analysis," Journal of
Retailing, 82 (3), 203-13.
47. Eisend, Martin (2011), "How Humor in Advertising Works: A Meta-analytic Test of
Alternative Models," Marketing Letters, 22 (2), 115-32.
48. ---- (2004), "Is it Still Worth to be Credible? A Meta-Analysis of Temporal Patterns of
Source Credibility Effects in Marketing," in Advances in Consumer Research, Barabara
E. Kahn and Mary Frances Luce, eds. Vol. 31. Valdosta, GA: Association for Consumer
Research, 352-57.
49. ---- (2010), "A Meta-analysis of Gender Roles in Advertising," Journal of the Academy
of Marketing Science, 38, 418-40.
50. ---- (2009), "A Meta-analysis of Humor in Advertising," Journal of the Academy of
Marketing Science, 37 (2), 191-203.
51. ---- (2006a), "Source Credibility in Marketing Communication: A Meta-analysis,"
Marketing - Journal for Research and Management, 2 (1), 43-60.
52. ---- (2006b), "Two-Sided Advertising: A Meta-Analysis," International Journal of
Research in Marketing, 23 (2), 187-98.
53. ---- (2007), "Understanding Two-Sided Persuasion: An Empirical Assessment of
Theoretical Approaches," Psychology & Marketing, 24 (6), 615-40.
54. Eisend, Martin and Franziska Kster (2011), "The Effectiveness of Publicity Versus
Advertising: A Meta-analytic Investigation of its Moderators," Journal of the Academy
of Marketing Science, 39 (6), 906-21.
55. Ellis, Paul D. (2006), "Market Orientation and Performance: A Meta-Analysis and CrossNational Comparisons," Journal of Management Studies, 43 (5), 1089-107.
56. Estelami, Hooman and Donald R. Lehmann (2001), "The Impact of Research Design on
Consumer Price Recall Accuracy: An Integrative Review," Journal of the Academy of
57. Estelami, Hooman, Donald R. Lehmann, and Alfred C. Holden (2001), "Macro-economic
Determinants of Consumer Price Knowledge: A Meta-Analysis of Four Decades of
Research," International Journal of Research in Marketing, 18 (4), 341-55.
58. Evanschitzky, Heiner, Martin Eisend, Roger J. Calantone, and Yuanyuan Jian (2012),
"Success Factors of Product Innovation: An Updated Meta-analysis," Journal of Product
Innovation Management, 29 (S1), 21-37.
5
59. Fan, Xiaojun, Yi Qian, and Pei Huang (2012), "Factors Influencing Consumer Behaviour
Towards Store Brand. A Meta-Analysis," International Journal of Market Research, 54
(3), 407-30.
60. Farley, John U., Donald R. Lehmann, and Michael J. Ryan (1981), "Generalizing from
'Imperfect' Replication," Journal of Business, 54 (4), 597-610.
61. Farrell, Seonaid and A. Ralph Hakstian (2001), "Improving Salesperson Performance: A
Meta-Analytic Investigation of the Effectiveness and Utility of Personnel Selection
Procedures and Training Interventions," Psychology & Marketing, 18 (3), 281-316.
62. Fox, Richard J., Melvin R. Crask, and Jonghoon Kim (1988), "Mail Survey Response
Rate: A Meta-Analysis of Selected Techniques for Inducing Response," Public Opinion
Quarterly, 52, 467-91.
63. Franke, George R. (1999), "The Consistency of Individual's Ethical Perceptions: A MetaAnalysis," Working Paper.
64. Franke, George R. and Jeong-Eun Park (2008), "Mediated Effects of Self-Monitoring on
Personal Selling: A Meta-Analysis," Working Paper.
65. ---- (2006), "Salesperson Adaptive Selling Behavior and Customer Orientation: A MetaAnalysis," Journal of Marketing Research, 43 (November), 693-702.
66. Freling, Traci H., Leslie H. Vincent, Robert M. Schindler, David M. Hardesty, and Jason
W. Rowe (2012), "A Meta-Analytic Review of Just Below Pricing Effects," Working
Paper.
67. Gallagher, Kristel M. and John A. Updegraff (2012), "Health Message Framing Effects
on Attitudes, Intentions, and Behavior: A Meta-Analytic Review," Annals of Behavioral
Medicine, 43, 101-16.
68. Garlin, Francine V. and Katherine Owen (2006), "Setting the Tone With the Tune: A
Meta-Analytic Review of the Effects of Background Music in Retail Settings," Journal
of Business Research, 59, 755-64.
69. Gelbrich, Katja and Holger Roschk (2011a), "Do Complainants Appreciate
Overcompensation? A Meta-Analysis on the Effect of Simple Compensation Vs.
Overcompensation on Post-Complaint Satisfaction," Marketing Letters, 22, 31-47.
70. ---- (2011b), "A Meta-Analysis of Organizational Complaint Handling and Customer
Responses," Journal of Service Research, 14 (1), 24-43.
71. Gerwin, Donald and Nicholas Barrowman (2002), "An Evaluation of Research on
Integrated Product Development," Management Science, 48 (7), 938-53.
72. Geyskens, Inge, Jan-Benedict E. M. Steenkamp, and Nirmalya Kumar (1999),
"Generalizations About Satisfaction in Marketing Channel Relationships Using Metaanalysis," Journal of Marketing Research, 36 (May), 223-38.
73. ---- (1998), "Generalizations About Trust in Marketing Channel Relationships Using
Meta-Analysis," International Journal of Research in Marketing, 15 (3), 223-48.
74. Grewal, Dhruv, Sukumar Kavanoor, Edward F. Fern, Carolyn Costley, and James Barnes
(1997), "Comparative Versus Noncomparative Advertising: A Meta-Analysis," Journal
of Marketing, 61 (October), 1-15.
75. Grewal, Dhruv, Scott Motyka, Nancy M. Puccinelli, Anne L. Roggeveen, Ahmad
Daryanto, Ko De Ruyter, and Martin Wetzels (2010), "Understanding How to Achieve
Competitive Advantage Through Regulatory Fit: A Meta-Analysis," Marketing Science
Institute Working Paper Series.
6
76. Grinstein, Amir (2008a), "The Effect of Market Orientation and Its Components on
Innovation Consequences: A Meta-Analysis," Journal of the Academy of Marketing
Science, 36, 166-73.
77. ---- (2008b), "The Relationships Between Market Orientation and Alternative Strategic
Orientations: A Meta-Analysis," European Journal of Marketing, 72 (1/2), 115-34.
78. Guay, Serge, Jean Perrien, and Raoul Graf (2006), "Portraying a Business Relationship
With MacNeil's Contractual Norms: A Meta-Analysis of the Empirical Literature,"
Proceedings of the Conference of the European Marketing Academy (EMAC 2006).
79. Hawkins, Timothy, Michael G. Knipper, and David Strutton (2008), "Opportunism in
Buyer-Supplier Relations: New Insights From Quantitative Synthesis," Journal of
Marketing Channels, 16 (1), 43-75.
80. Heath, Timothy B. and Subimal Chatterjee (1995), "Asymetric Decoy Effects on LowerQuality versus Higher-Quality Brands: Meta-analytic and Experimental Evidence,"
Journal of Consumer Research, 22 (December), 268-84.
81. Henard, David H. and David M. Szymanski (2001), "Why Some New Products Are More
Successful Than Others," Journal of Marketing Research, 38, 362-75.
82. Hildebrand, Anja and Martin Klarmann (2012), "A Meta-Analysis of Salesperson
Innovation Adoption," Working Paper.
83. Hornik, Jacob, Joseph Cherian, Michelle Madansky, and Chem Narayana (1995),
"Determinants of Recycling Behavior: A Synthesis of Research Results," Journal of
Socio-Economics, 24 (1), 105-27.
84. Hornikx, Jos and Daniel J. O'Keefe (2009), "Adapting Consumer Advertising Appeals to
Cultural Values: A Meta-Analytic Review of Effects on Persuasiveness and Ad Liking,"
Communication Yearbook, 33, 39-71.
85. Huber, Frank and Marc Fischer (1999), "The Picture Communication Effect: A MetaAnalysis," Proceedings of the World Marketing Congress.
86. Hyodo, Jamie (2011), "Can Colors Make Me Happy? The Effect of Color on Mood: A
Meta-Analysis," Advances in Consumer Research, 39, 858-67.
87. Janiszewski, Chris, Hayden Noel, and Alan G. Sawyer (2003), "A Meta-Analysis of the
Spacing Effect in Verbal Learning: Implications for Research on Advertising Repetition
and Consumer Memory," Journal of Consumer Research, 30, 138-49.
88. Jaramillo, Fernando, Daniel M. Ladik, Greg W. Marshall, and Jay Prakash Mulki (2007),
"A Meta-Analysis of the Relationship Between Sales Orientation - Customer
Orientation (SOCO) and Salesperson Job Performance," Journal of Business and
Industrial Marketing, 22 (5), 302-10.
89. Jaramillo, Fernando, J. P. Mulki, and G. W. Marshall (2004), "A Meta-Analysis of the
Relationship Between Organizational Commitment and Job Performance: 25 Years of
Research," Journal of Business Research.
90. Johnson, Blair T. and Alice H. Eagly (1989), "Effects of Involvement on Persuasion: A
Meta-Analysis," Psychological Bulletin, 106 (2), 290-314.
91. Kareklas, Ioannis (2010), "A Quantitative Review and Extension of Racial Similarity
Effects in Advertising," PhD Thesis, University of Connecticut.
92. Keller, Punam Anand and Donald R. Lehmann (2008), "Designing Effective Health
Communications: A Meta-Analysis," Journal of Public Policy & Marketing, 27 (2), 11730.
7
93. Keller, Punam Anand, Donald R. Lehmann, and Katherine J. Milligan (2009),
"Effectiveness of Corporate Well-Being Programs," Journal of Macromarketing, 29 (3),
279-302.
94. Kirca, Ahmet H., Satish Jayachandran, and William O. Bearden (2005), "Market
Orientation: A Meta-Analytic Review and Assessment of Its Antecedents and Impact on
Performance," Journal of Marketing, 69 (April), 24-41.
95. Krasnikov, Alexander and Satish Jayachandran (2008), "The Relative Impact of
Marketing, Research-and-Development, and Operations Capabilities on Firm
Performance," Journal of Marketing, 72 (July), 1-11.
96. Krishna, Aradhna, Richard Briesch, Donald R. Lehmann, and Hong Yuan (2002), "A
Meta-Analysis of the Impact of Price Presentation on Perceived Savings," Journal of
Retailing, 78, 101-18.
97. Laer, Tom van, Ko de Ruyter, Luca M. Visconti, and Martin Wetzels (2012), "An
Extended Transportation-Imagery Model: A Meta-Analysis of the Antecedents and
Consequences of Consumers' Narrative Transportation," Working Paper.
98. Lawless, Michael W. and James E. Nelson (1994), "On Small Effects in Strategic
Management Research: Trends and Implications From Two Meta-Analyses,"
Proceedings of the Winter Educator's Conference of the American Marketing
Association.
99. Lewin, Jeffrey E. and Naveen Donthu (2005), "The Influence of Purchase Situation on
Buying Center Structure and Involvement: A Select Meta-Analysis of Organizational
Buying Behavior Research," Journal of Business Research, 58, 1381-90.
100.Lynn, Michael (1991), "Scarcity Effects on Value: A Quantitative Review of the
Commodity Theory Literature," Psychology & Marketing, 8 (1), 43-57.
101.Lynn, Michael and Michael McCall (2000), "Gratitude and Gratuity: A Meta-Analysis of
Research on the Service-Tipping Relationship," Journal of Socio-Economics, 29, 20314.
102.Manfreda, Katja Lozar, Michael Bosnjak, Jernej Berzelak, Iris Haas, and Vasja Vehovar
(2008), "Web Surveys Versus Other Survey Modes: A Meta-Analysis Comparing
Response Rates," International Journal of Market Research, 50 (1), 79-104.
103.Martin, Mary C. (1997), "Children's Understanding of the Intent of Advertising: A MetaAnalysis," Journal of Public Policy & Marketing, 16 (2), 205-16.
104.Martinez, J. Michael, Jeffrey L. Stinson, Minsoo Kang, and Colby B. Jubenville (2010),
"Intercollegiate Athletics and Institutional Fundraising: A Meta-Analysis," Sport
Marketing Quarterly, 19, 36-47.
105.Mata, Rui and Ludmila Nunes (2010), "When Less Is Enough: Cognitive Aging,
Information Search, and Decision Quality in Consumer Choice," Psychology and
Aging, 25 (2), 289-98.
106.Melnyk, Vladimir, Erica van Herpen, and Hans C. M. van Trijp (2010), "Determinants of
the Influence of Social Norms on Consumer Behavior: A Meta-Analysis," Working
Paper.
107.Montoya-Weiss, Mitzi M. and Roger J. Calantone (1994), "Determinants of New-Product
Performance: A Review and Meta-Analysis," Journal of Product Innovation
Management, 11 (5), 397-417.
8
108.Nelson, Jon P. (2011), "Alcohol Marketing, Adolescent Drinking and Publication Bias in
Longitudinal Studies: A Critical Survey Using Meta-Analysis," Journal of Economic
Surveys, 25 (2), 191-232.
109.Notani, Arti Sahni (1998), "Moderators of Perceived Behavioral Control's Predictiveness
in the Theory of Planned Behavior: A Meta-Analysis," Journal of Consumer
Psychology, 7 (3), 247-72.
110.O'Keefe, Daniel J. (1999), "How to Handle Opposing Arguments in Persuasive
Messages: A Meta-Analytic Review of the Effects of One-Sided and Two-Sided
Messages," Communication Yearbook, 22, 208-49.
111.---- (1998), "Justification Expliciteness and Persuasive Effect: A Meta-analytic Review of
the Effects of Varying Support Articulation in Persuasive Messages," Argumentation &
Advocacy, 35 (2), 61-75.
112.---- (1987), "The Persuasive Effects of Delaying Identification on High- and LowCredibility Communicators: A Meta-Analytic Review," Central States Speech Journal,
38 (2), 63-72.
113.---- (2002), "The Persuasive Effects of Variation in Standpoint Articulation," in Advances
in Pragma-Dialectics, F. H. van Eemeren, ed. Amsterdam: Sic Sat, 65-82.
114.O'Keefe, Daniel J. and Jakob D. Jensen (2006), "The Advantages of Compliance or the
Disadvantage of Noncompliance? A Meta-Analytic Review of the Relative Persuasive
Effectiveness of Gain-Framed and Loss-Framed Messages," Communication Yearbook,
30, 1-43.
115.---- (2009), "The Relative Persuasiveness of Gain-Framed and Loss-Framed Messages for
Encouraging Disease Detection Behaviors: A Meta-Analytic Review," Journal of
Communication, 59, 296-316.
116.---- (2007), "The Relative Persuasiveness of Gain-Framed and Loss-Framed Messages for
Encouraging Disease Prevention Behaviors: A Meta-Analytic Review," Journal of
Health Communication, 12, 623-44.
117.Orsingher, Chiara, Sara Valentini, and Matteo de Angelis (2010), "A Meta-analysis of
Satisfaction With Complaint Handling in Services," Journal of the Academy of
Marketing Science, 38, 169-86.
118.Palmatier, Robert W., Rajiv P. Dant, Dhruv Grewal, and Kenneth R. Evans (2006),
"Factors Influencing the Effectiveness of Relationship Marketing: A Meta-Analysis,"
Journal of Marketing, 70 (October), 136-53.
119.Pan, Yue, Simon Sheng, and Frank T. Xie (2012), "Antecedents of Customer Loyalty: An
Empirical Synthesis and Reexamination," Journal of Retailing and Consumer Services,
19, 150-58.
120.Pan, Yue and John R. Sparks (2012), "Predictors, Consequence, and Measurement of
Ethical Judgments: Review and Meta-Analysis," Journal of Business Research, 65, 8491.
121.Pan, Yue and George M. Zinkhan (2006), "Determinants of Retail Patronage: A MetaAnalytical Perspective," Journal of Retailing, 82 (3), 229-43.
122.Pattikawa, Lenny H., Ernst Verwaal, and Harry R. Commandeur (2006), "Understanding
New Product Performance," European Journal of Marketing, 40 (11/12), 1178-93.
123.Payan, Janice (1998), "Ethnic Differences in Self-Esteem: A Meta-Analysis,"
Proceedings of the Winter Conference of the American Marketing Association, 335-41.
9
124.Pentina, Iryna and David Strutton (2007), "Information Processing and New Product
Success: A Meta-Analysis," European Journal of Innovation Management, 10 (2), 14975.
125.Peter, J. Paul and Gilbert A. Churchill, Jr. (1986), "Relationships Among Research
Design Choices and Psychometric Properties of Rating Scales: A Meta-Analysis,"
Journal of Marketing Research, 23 (February), 1-10.
126.Peterson, Richard A. and A.J.P. Jolibert (1995), "A Meta-Analysis of Country-of-OriginEffects," Journal of International Business Studies, 26, 883-900.
127.Peterson, Robert A. (2000), "A Meta-Analysis of Variance Accounted for and Factor
Loadings in Exploratory Factor Analysis," Marketing Letters, 11 (3), 261-75.
128.Peterson, Robert A., Gerald Albaum, and Richard F. Beltramini (1985), "A MetaAnalysis of Effects Sizes in Consumer Behavior Experiments," Journal of Consumer
Research, 12 (June), 97-103.
129.Rajamma, Rajasree K., Mohammad Ali Zolfagharian, and Lou E. Pelton (2011),
"Dimensions and Outcomes of B2B Relational Exchange: A Meta-Analysis," Journal of
Business and Industrial Marketing, 26 (2), 104-14.
130.Rao, Akshay R. and Kent B. Monroe (1989), "The Effect of Price, Brand Name, and
Store Name on Buyers' Perceptions of Product Quality: An Integrative Review," Journal
of Marketing Research, 26 (August), 351-57.
131.Reilly, Michael D. and Jerry N. Conover (1983), "Meta-Analysis: Integrating Results
From Consumer Research," Advances in Consumer Research, 10, 509-13.
132.Rich, Gregory A., William H. Bommer, Scott B. MacKenzie, Philip M. Podsakoff, and
Jonathan L. Johnson (1999), "Apples and Apples or Apples and Oranges? A MetaAnalysis of Objective and Subjective Measures of Salesperson Performance," Journal of
Personal Selling and Sales Management, 19 (4), 41-52.
133.Riggle, Robert J., Diane R. Edmondson, and John D. Hansen (2009), "A Meta-Analysis
of the Relationship Between Perceived Organizational Support and Job Outcomes: 20
Years of Research," Journal of Business Research, 62, 1027-30.
134.Rubera, Gaia and Ahmet H. Kirca (2012), "Firm Innovativeness and Its Performance
Outcomes: A Meta-Analytic Review and Theoretical Integration," Journal of Marketing,
76 (May), 130-47.
135.Saeed, Khawaja A., Hwang Yujong, and Mun Y. Yi (2003), "Toward an Integrative
Framework for Online Consumer Behavior Research: A Meta-Analysis Approach,"
Journal of End User Computing, 15 (4), 1-26.
136.Scheibehenne, Benjamin, Rainer Greifeneder, and Peter M. Todd (2010), "Can There Be
Too Many Options? A Meta-Analytic Review of Choice Overload," Journal of
Consumer Research, 37 (3), 409-25.
137.Schepers, Jeroen and Martin Wetzels (2007), "A Meta-Analysis of the Technology
Acceptance Model: Investigating Subjective Norm and Moderation Effects,"
Information & Management, 44, 90-103.
138.Schwenk, Gero and Guido Mser (2009), "Intention and Behavior: A Bayesian MetaAnalysis With Focus on the Ajzen-Fishbein Model in the Field of Environmental
Behavior," Quality & Quantity, 43, 743-55.
139.Sheppard, Blair H., Jon Hartwick, and Paul R. Warshaw (1988), "The Theory of
Reasoned Action: A Meta-Analysis of Past Research With Recommendations for
10
Modifications and Future Research," Journal of Consumer Research, 15 (December),
325-43.
140.Shoham, Aviv (2002), "Standardization of International Strategy and Export
Performance: A Meta-Analysis," Journal of Global Marketing, 16 (1/2), 97-121.
141.Shoham, Aviv, Gregory M. Rose, and Frederic Kropp (2005), "Market Orientation and
Performance: A Meta-Analysis," Marketing Intelligence & Planning, 23 (5), 435-54.
142.Shoham, Aviv, Ayalla Ruvio, Eran Vigoda-Gadot, and Nitza Schwabsky (2006), "Market
Orientations in the Nonprofit and Voluntary Sector: A Meta-Analysis of Their
Relationships With Organizational Performance," Nonprofit and Voluntary Sector
Quarterly, 35 (3), 453-76.
143.Sierra, Jeremy J., Michael R. Hyman, and Robert S. Heiser (2012), "Ethnic Identity in
Advertising: A Review and Meta-Analysis," Journal of Promotion Management, 18 (4),
489-513.
144.Singer, Eleanor, John Van Hoewyk, Nancy Gebler, Trivellore Raghunathan, and
Katherine McGonagle (1999), "The Effect of Incentives on Response Rates in
Interviewer-Mediated Surveys," Journal of Official Statistics, 15 (2), 217-30.
145.Sivasubramaniam, Nagaraj, S. Jay Liebowitz, and Conway L. Lackman (2012),
"Determinants of New Product Development Team Performance: A Meta-Analytic
Review," Journal of Product Innovation Management, 29 (5), 803-20.
146.Smith, L.P. and H.F. Koenig (1985), "Meta-Analysis on the Relationships Between
Satisfaction and Manifest Conflict in Marketing Channels," Winter Educators'
Conference Proceedings of the American Marketing Association, 341-47.
147.Song, Michael, Ksenia Podoynitsyna, Hans van der Bij, and Johannes I. M. Halman
(2008), "Success Factors in New Ventures: A Meta-Analysis," Journal of Product
Innovation Management, 25, 7-27.
148.Spangenberg, Eric R. and Anthony G. Greenwald (1999), "Social Influence by
Requesting Self-Prophecy," Journal of Consumer Psychology, 8 (1), 61-89.
149.Steenkamp, Jan-Benedict E. M. and Inge Geyskens (2012), "Transaction Cost Economics
and the Roles of National Culture: A Test of Hypotheses Based on Inglehart and
Hofstede," Journal of the Academy of Marketing Science, 40, 252-70.
150.Sultan, Fareena, John U. Farley, and Donald R. Lehmann (1990), "A Meta-Analysis of
Applications of Diffusion Models," Journal of Marketing Research, 27 (February), 7077.
151.Swan, John E., Michael R. Bowers, and Lynne D. Richardson (1999), "Customer Trust in
the Salesperson: An Integrative Review and Meta-Analysis of the Empirical," Journal of
Business Research, 44 (2), 93-107.
152.Sweat, Michael D., Julie Denison, Caitlin Kennedy, Virginia Tedrow, and Kevin O'Reilly
(2012), "Effects of Condom Social Marketing on Condom Use in Developing Countries:
A Systematic Review and Meta-Analysis, 1990-2010," Bulletin of the World Health
Organization, 90, 613-22.
153.Szymanski, David M. and Paul S. Busch (1987), "Identifying the Generics-Prone
Consumer: A Meta-Analysis," Journal of Marketing Research, 24 (November), 425-31.
154.Szymanski, David M. and David H. Henard (2001), "Customer Satisfaction: A MetaAnalysis of the Empirical Evidence," Journal of the Academy of Marketing Science, 29
(1), 16-35.
11
155.Szymanski, David M., Michael W. Kroff, and Lisa C. Troy (2007), "Innovativeness and
New Product Success: Insights from the Cummulative Evidence," Journal of the
Academy of Marketing Science, 35, 35-52.
156.Taylor, David G. and David Strutton (2010), "Has E-Marketing Come of Age? Modeling
Historical Influences on Post-Adoption Era Internet Consumer Behaviors," Journal of
Business Research, 63, 950-56.
157.Tellis, Gerard J. and Birger Wernerfelt (1987), "Competitive Price and Quality Under
Asymmetric Information," Marketing Science, 6 (3), 240-53.
158.Tongren, Hale N. (1988), "Determinant Behavior Characteristics of Older Consumers,"
Journal of Consumer Affairs, 22 (1), 136-57.
159.Trappey, Charles (1996), "A Meta-Analysis of Consumer Choice and Subliminal
Advertising," Psychology & Marketing, 13 (5), 517-30.
160.Troy, Lisa C., Tanawat Hirunyawipada, and Audhesh K. Paswan (2008), "Crossfunctional Integration and New Product Success: An Empirical Investigation of the
Findings," Journal of Marketing, 72 (November), 132-46.
161.Tsai, Kuen-Hung and Chi-Tsun Huang (2012), "Technology Synergy, Product
Characteristics, and New Product Performance: A Meta-Analytic Review," Canadian
Journal of Administrative Sciences, 29, 336-47.
162.Van den Bulte, Christophe and Stefan Stremersch (2004), "Social Contagion and Income
Heterogeneity in New Product Diffusion: A Meta-Analytic Test," Marketing Science,
23 (4), 530-54.
163.van der Lee, Angela and Bas van den Putte (2001), "Fram Fame to Fortune: A MetaAnalytic Review of Credibility Endorsements in Persuasive Communication," Working
Paper.
164.Verbeke, Willem, Bart Dietz, and Ernst Verwaal (2011), "Drivers of Sales Performance:
A Contemporary Meta-Analysis. Have Salespeople Become Knowledge Brokers?,"
Journal of the Academy of Marketing Science, 39 (3), 407-28.
165.Verlegh, Peeter W. J. and Jan-Benedict E.M. Steenkamp (1999), "A Review and MetaAnalysis of Country-of-Origin Research," Journal of Economic Psychology, 20 (5),
521-47.
166.Vinchur, Andrew J., Fred S. Switzer III Schippmann, and Philip L. Roth (1998), "A
Meta-Analytic Review of Predictors of Job Performance for Salespeople," Journal of
Applied Psychology, 83 (4), 586-97.
167.Vlckner, Franziska and Julian Hofmann (2007), "The Price-Perceived Quality
Relationship: A Meta-Analytic Review and Assessment of Its Determinants," Marketing
Letters, 18, 181-96.
168.Walliser, Bjrn and Patrice Laroche (2004), "What Do We Really Know About
Sponsorship Memorization? A Meta-Analysis of the Determinants of Sponsors
Recognition," Working Paper.
169.Wang, Xuehua and Zhilin Yang (2008), "A Meta-Analysis of Effect Sizes in International
Marketing Experiments," International Marketing Review, 25 (3), 276-91.
170.Wellman, Robert J., David B. Sugarman, Joseph R. DiFranza, and Jonathan P. Winickoff
(2006), "The Extent to Which Tobacco Marketing and Tobacco Use in Films Contribute
to Children's Use of Tobacco," Archives of Pediatrics & Adolescent Medicine, 160
(December), 1285-96.
12
171.Wilson, Elizabeth J. and Daniel L. Sherrell (1993), "Source Effects in Communication
and Persuasion Research: A Meta-Analysis of Effect Size," Journal of the Academy of
172.Wood, John Andy, James S. Boles, Wesley Johnston, and Danny Bellenger (2008),
"Buyers' Trust of the Salesperson: An Item-Level Meta-Analysis," Journal of Personal
Selling and Sales Management, 28 (3), 263-83.
173.Wright, Malcolm and Murray MacRae (2007), "Bias and Variability in Purchase
Intention Scales," Journal of the Academy of Marketing Science, 35, 617-24.
174.Wright, Newell D. and Val Larsen (1993), "Materialism and Life Satisfaction: A MetaAnalysis," Journal of Consumer Satisfaction, Dissatisfaction & Complaining Behavior,
6, 158-65.
175.Yammarino, Francis J., Steven J. Skinner, and Terry L. Childers (1991), "Understanding
Mail Survey Response Behavior," Public Opinion Quarterly, 55, 613-39.
176.Zablah, Alex. R., George R. Franke, Tom J. Brown, and Darrell E. Bartholomew (2012),
"How and When Does Customer Orientation Influence Frontline Employee Job
Outcomes? A Meta-Analytic Evaluation," Journal of Marketing, 76 (May), 21-40.
13
Web Appendix B
TABLE B1
Correlation matrix for advertising (n = 294)
# Variable
1 ES
2 Time
3 Maturity
4 Intensity
5 Journal
6 Unpublished
7 Method
1
2
3
1.000
-.102 1.000
-.126 -.210 1.000
.076 -.028 -.096
.047 -.173 -.364
-.020 .066 .374
.041 .104 .022
1.000
.097 1.000
-.215 -.234 1.000
.240 -.030 -.414 1.000
TABLE B2
Correlation matrix for channels (n = 274)
# Variable
1 ES
2 Time
3 Maturity
4 Intensity
5 Journal
6 Unpublished
7 Method
1
2
3
4
5
6
1.000
-.156 1.000
-.160 .339 1.000
-.063 -.101 .043 1.000
-.023 -.576 -.094 .002 1.000
-.238 .574 .738 .165 -.367 1.000
-
TABLE B3
Correlation matrix for consumer behavior (n = 457)
# Variable
1 ES
2 Time
3 Maturity
4 Intensity
5 Journal
6 Unpublished
7 Method
1
2
3
1.000
.139 1.000
-.330 .038 1.000
.009 -.180 -.073
.066 -.176 -.416
-.110 -.385 -.030
-.045 -.163 .073
1.000
.030 1.000
-.113 .429 1.000
-.116 .242 .396 1.000
14
TABLE B4
Correlation matrix for method (n = 56)
# Variable
1 ES
2 Time
3 Maturity
4 Intensity
5 Journal
6 Unpublished
7 Method
1
2
3
1.000
.320 1.000
-.156 .155 1.000
.097 .127 -.302
.100 -.412 -.578
-.076 .570 .668
-.066 -.151 .253
1.000
.123 1.000
-.162 .560 1.000
-.969 -.042 .143 1.000
TABLE B5
Correlation matrix for new product development (n = 238)
# Variable
1 ES
2 Time
3 Maturity
4 Intensity
5 Journal
6 Unpublished
7 Method
1
2
3
4
5
6
1.000
-.239 1.000
-.038 -.129 1.000
.027 .314 -.332 1.000
.312 -.332 -.292 .128 1.000
.066 -.258 -.411 -.036 .556 1.000
-
TABLE B6
Correlation matrix for pricing (n = 57)
# Variable
1 ES
2 Time
3 Maturity
4 Intensity
5 Journal
6 Unpublished
7 Method
1
2
3
1.000
-.190 1.000
.056 -.154 1.000
-.047 .040 .078
-.044 -.698 .209
.290 -.540 .135
-.118 .261 .062
1.000
.255 1.000
-.727 .076 1.000
-.768 -.067 -.800 1.000
15
TABLE B7
Correlation matrix for sales (n = 362)
# Variable
1 ES
2 Time
3 Maturity
4 Intensity
5 Journal
6 Unpublished
7 Method
1
2
3
1.000
.138 1.000
-.210 -.771 1.000
.189 .098 -.270
.012 -.112 -.243
-.183 -.341 .329
.009 -.125 .074
1.000
.037 1.000
.211 -.157 1.000
.030 .093 .141 1.000
TABLE B8
Correlation matrix for strategy (n = 231)
# Variable
1 ES
2 Time
3 Maturity
4 Intensity
5 Journal
6 Unpublished
7 Method
1
2
3
4
5
6
1.000
.227 1.000
-.476 -.563 1.000
-.011 -.235 -.068 1.000
.088 .424 -.042 -.119 1.000
.131 .412 -.295 -.197 .659 1.000
-
16
Web Appendix C
FIGURE C
Relationship between effect size and time: LOWESS smoother
All effect sizes
Advertising
Lowess smoother
.2
.2
effect size
.4
.6
effect size
.4
.6
.8
.8
Lowess smoother
1960
1970
1980
1990
2000
average publication year of studies in the meta-analysis
1970
2010
1980
1990
2000
2010
bandwidth = .8
bandwidth = .8
Channels
Consumer behavior
Lowess smoother
.2
.2
effect size
.4
.6
effect size
.4
.6
.8
.8
Lowess smoother
1980
1995
1985
1990
bandwidth = .8
2000
1970
1990
2000
1980
bandwidth = .8
2010
17
Method
New Product Development

Lowess smoother
.1
.2
effect size
.2
.3
effect size
.4
.4
.6
.5
Lowess smoother
1970
1980
1990
1970
2000
2000
1980
1990
2010
bandwidth = .8
bandwidth = .8
Pricing
Sales
Lowess smoother
.2
.2
effect size
.4
effect size
.4
.6
.6
.8
.8
Lowess smoother
1975
1980
1985
1990
1995
2000
1960
1970
1980
1990
2000
bandwidth = .8
bandwidth = .8
Strategy
.2
effect size
.4
.6
.8
Lowess smoother
1985
1990
1995
2000
bandwidth = .8
2005
Copyright of Journal of Marketing is the property of American Marketing Association and its
content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email
articles for individual use.

Articulo Investigacion

Uploaded by

Copyright:

Available Formats

Articulo Investigacion

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Articulo Investigacion

Uploaded by

Copyright:

Available Formats

Martin Eisend

Have We Progressed Marketing

Online Supplement: http://dx.doi.org/10.1509/jm.14.0288

ince being recognized as a separate field of inquiry

How much do we know about marketing and how has

Martin Eisend is Professor of Marketing, European University Viadrina

2015, American Marketing Association

development that applies to marketing research and

Effect Sizes as a Measure of

A major function of science is to produce explanations of

24 / Journal of Marketing, May 2015

The value of scientific knowledge depends not only on

The Progress of Scientific

How does scientific knowledge develop over time?

lows a linear trend, takes the form of a logistic or quadratic

Scientific progress is continuous if the research field starts

Scientific Progress Is Discontinuous

As noted previously, the sociology of science proposes a

If no accepted paradigm guides research, a chaotic research

discontinuous but remains static (i.e., effect sizes remain

Subject Areas in Marketing

Marketing is a broad research field with many subject areas.

1The subject area diffusion of innovation was not considered,

Have We Progressed Marketing Knowledge? / 25

Typical Publication Outlets for Primary

International Journal of Advertising

Effect of advertising characteristics/stimuli on consumer

This classification scheme is debatable (as is any other

To investigate the amount and development of knowledge

26 / Journal of Marketing, May 2015

performedthat is, a meta-analysis of meta-analyses on

To locate published meta-analyses in marketing research,

tive review, synthesis, and generalization, combined

2The journals are European Journal of Marketing, International

Kayande and Bhargava [1994]), the primary meta-analysis

At least one meta-analytic effect size was selected from

Have We Progressed Marketing Knowledge? / 27

because no difference between raw mean and weighted

Coding of Independent and Control Variables

The main independent variable is time, which explains the

Effect size (ES)

Explains the size of the effect as a

Min = .00, max = .91,

Min = .18, max = 218.17,

28 / Journal of Marketing, May 2015

and presumably influential control variables in addition to

The effect sizes were modeled as a linear function of the

Figure 1 presents the frequency distribution of the observed

subject area + time +

ESi + 11maturityi + 12 intensity i + 13 journali

ESi + 14 manuscript statusi + 15 method typei + ei ,

where ESi is the meta-analytic effect size, b0 is a constant,

Table 4 provides the correlation matrix of the variables. For

How Much Do We Know About Marketing? The

Have We Progressed Marketing Knowledge? / 29

30 / Journal of Marketing, May 2015

aThe number of meta-analyses/effect sizes does not

Have We Progressed Marketing Knowledge? / 31

success). The mean effect size in this meta-meta-analysis