This document summarizes two major project failures: the healthcare.gov website launch and the Denver International Airport baggage handling system project.
The healthcare.gov website launch in 2014 failed due to an overly aggressive schedule that did not allow time to meet customer needs or properly test the site's capacity. The site crashed upon launch due to being overwhelmed by high traffic. The Denver airport baggage system project in the 1990s was plagued by dysfunctional decision making, unrealistic timelines, and lack of expertise to manage such a complex automated system. The project went over budget, delayed the airport's opening by over a year, and never functioned properly despite 10 additional years of work. Both projects failed due to multiple issues including poor planning
1. BBA 3626, Project Management Overview 1
Course Learning Outcomes for Unit I
Upon completion of this unit, students should be able to:
1. Identify project management concepts.
Reading Assignment
Chapter 1: Introduction to Project Management, pp. 2-19, and
23
Unit Lesson
Unit I covers basic project management concepts. This summary
is going to emphasize some major project
failures. Let’s start out with a quote that relates to this subject
by Albert Einstein, “Insanity: doing the same
thing over again and expecting different results” (Nelson, 2007,
p. 1).
One of the project failures of 2014 was the healthcare.gov
failure which caused a major crisis for this
mandated program. This failure was due to an over aggressive
2. schedule and not meeting the customer
needs. Too many people trying to get access to a site that could
not handle the capacity helped the failure of
heathcare.gov. It appears that possibly the site was not tested to
see if the capacity was correct or could
handle the load. The healthcare.gov site was not ready to go
when it went into implementation. According to
Matthew Heusser (2013), this was “the most public software
project failure of the decade” (para. 2).
Healthcare.gov used the agile approach to software development
in the terms of a sprint. A sprint is a session
every two weeks where the team goes over their progress and
shows what works or does not work. Then the
team plans what they will do next. It is a storyboard type of
development that is the new buzz word for
software development. Basically, each sprint is an iteration or a
completed chunk of software development to
be designed. The theory is that the software code is designed,
coded, and fully tested from end-to-end before
moving on to the next chunk. The development may have been
going along fine but the system was not ready
to be fully implemented as a finished product. The system was
not tested to its capacity if at all. The project’s
failure was monumental and systemic. Multiple failures
occurred on multiple levels, but no one really knows
exactly what happened. Healthcare.gov was only in the beta
testing phase which means it was not ready for
production (Heusser, 2013).
Media sites such as the New Yorker, Washington Post, and
MedCity News claimed healthcare.gov failed
because agile development was not utilized. Unfortunately, that
was false because healthcare.gov’s front-end
GUI and back-end data services hub were utilizing agile
processes. Evidence clearly showed that sprints,
3. user stories, and incremental testing were occurring during the
development process. Possibly in this case,
requirements decomposition should have been used instead of
user stories. A good design document may
have helped this project be successful. There is one thing for
certain, bad project management skills were
alive and well in this project (Daconta, 2013).
Now, let’s go back in time to the 1990s and discuss another
huge project failure. The Denver Airport Baggage
System project in the 1990s was a disaster. Dysfunctional
decision making aided in the demise of this project.
The world’s largest automated airport baggage handling system
was a project gone wrong. The goal of the
project was to have a turnaround time of about 30 minutes once
the plane landed. Denver needed a larger
airport so this plan was within the larger airport plan. The plan
became too complex and the opening of the
airport added an additional 16 months. The delay cost the city
of Denver $1.1 million per day throughout the
delay. Even when the airport finally did open, the baggage
system was not at full capacity. Only a single
concourse, with a single airline, and for outbound traffic only
worked on the first day of opening. All three
UNIT I STUDY GUIDE
An Introduction to Project Management
BBA 3626, Project Management Overview 2
4. UNIT x STUDY GUIDE
Title
concourses should have been an integrated system. This project
was worked on another 10 years and it
never worked well. The project was terminated in 2005
(Calleam Consulting, 2008).
Some of the failures of this Denver project included an
underestimate of the complexity involved, changes in
strategy, decision to proceed, schedule, scope, budget
constraints, acceptance of change requests, design of
the physical building structure, and the decision to seek a
different path. This project was compounded by so
many project failure issues that it never recovered. Other failure
points included risk management failures
when an electrical system suffered from a power surge that
crashed the system before the required filters to
prevent this were installed. If the team had focused on risk
management procedures, this failure would have
been avoided. Also, one of the key leaders passed away and a
new sponsor took over who lacked the
knowledge to understand the system. The system was plagued
with architectural and design issues because
the system had over 100 individual PCs networked together.
Failure of one PC could result in an outage since
no automatic backups for failed components were considered.
The PCs were distributed all over the airport
causing even more problems when a problem arose. The system
was unable to detect jams when they
occurred and suitcases just kept piling up. Schedule pressure
may have been a factor for the design
problems. The system was not fault tolerant in any aspect and
was a total failure. Knowledge and expertise
5. played a key role in the failure of this project. The leaders did
not have the expertise to deal with such a
sophisticated automated baggage system which was new at the
time (Calleam Consulting, 2008).
Projects fail for many reasons such as: little or no support from
senior management, absence of user
participation throughout the project, poorly or ill-defined scope,
inadequate project leadership, lack of good
requirements, no definitive methodology, unrealistic marketing
deadlines versus technology implementation
requirements, no communication, no formal change or
configuration control management, inadequate quality
assurance verification and validation, lack of good testing
techniques or none at all, lack of knowledge
transfer, inexperienced IT professionals, and employee and
experienced IT professionals burn out. Both of
these projects discussed suffered from multiple reasons listed
above. Throughout this course, you will learn
how to manage a project effectively in order to avoid becoming
a statistic as listed in the failed projects of this
unit summary.
References
Calleam Consulting. (2008). Case study- Denver International
Airport baggage handling system: An
illustration of ineffectual decision making. Retrieved from
http://calleam.com/WTPF/wp-
content/uploads/articles/DIABaggage.pdf
Daconta, M. C. (2013). Media got it wrong: Healthcare.gov
6. failed despite agile practice. GCN. Retrieved from
http://gcn.com/blogs/reality-check/2013/11/healthcare-
agile.aspx
Heusser, M. (2013). 6 software development lessons learned
from Healthcare.gov’s failed launch. Retrieved
from http://www.cio.com/article/2380827/developer/6-software-
development-lessons-from-healthcare-
gov-s-failed-launch.html
Nelson, R. (2007). IT project management: Infamous failures,
classic mistakes, and best practices. MIS
Quarterly Executive, 6(2).
Suggested Reading
Nelson, R. (2007). IT project management: Infamous failures,
classic mistakes, and best practices. MIS
Quarterly Executive, 6(2).
MAR 3231, Marketing Research 1
Course Learning Outcomes for Unit I
7. Upon completion of this unit, students should be able to:
1. Discuss marketing research’s primary business functions.
1.1 Describe the differences between market research and
marketing research.
1.2 Evaluate the importance of marketing research in
organizations.
1.3 Examine how marketing research helps managers make
major marketing decisions.
Course/Unit
Learning Outcomes
Learning Activity
1.1
Unit Lesson Presentation; “Marketing Research: Meaning,
Definitions, and
Objectives – Explained!;” “The Declining Use of the Term
Market Research;”
Article Review
1.2
Unit Lesson Presentation; “Market Research;” “Consumer
Market Research is
the Systematic Collection of Data Regarding Customers’
Preference for Actual
and Potential Products/services;” “The Value of Marketing
Research to Your
Business;” “Researching Consumers in Multicultural Societies:
Emerging
Methodological Issues”
8. 1.3
Unit Lesson Presentation; “IDC Research Shows How Buyer
Preferences
Influencing Wearable Designs;” “Marketing Research is
‘Critical’”
Required Unit Resources
Click here to access a resource on consumer market research.
In order to access the following resources, click the links
below:
Chand, S. (n.d.). Marketing research: Meaning, definition, and
objectives – Explained! Retrieved from
http://www.yourarticlelibrary.com/marketing/marketing-
research-meaning-definition-and-objectives-
explained/25862/
Fellman, M. W. (1999). Marketing research is ‘critical.’
Marketing Research, 11(3), 4–5. Retrieved from
https://libraryresources.columbiasouthern.edu/login?url=http://s
earch.ebscohost.com/login.aspx?direc
t=true&db=bth&AN=2554392&site=ehost-live&scope=site
Films Media Group (Producer). (2009). Market research [Video
file]. Retrieved from
https://libraryresources.columbiasouthern.edu/login?auth=CAS
&url=http://fod.infobase.com/PortalPla
9. ylists.aspx?wID=273866&xtid=42019&loid=89527
Click here to access a transcript for this video.
Nunan, D. (2016). The declining use of the term market
research. International Journal of Market Research,
58(4), 503-522. Retrieved from
https://libraryresources.columbiasouthern.edu/login?url=http://s
earch.ebscohost.com/login.aspx?direc
t=true&db=bth&AN=117204670&site=ehost-live&scope=site
Pronto Marketing. (2016). The value of marketing research to
your business. Retrieved from
https://www.prontomarketing.com/2016/12/the-value-of-
marketing-research-to-your-business/
UNIT I STUDY GUIDE
The Role and Value of
Market Research
https://online.columbiasouthern.edu/CSU_Content/Courses/Busi
ness/MAR/MAR3231/17G/UnitI_ConsumerMarketResearch.pdf
http://www.yourarticlelibrary.com/marketing/marketing-
research-meaning-definition-and-objectives-explained/25862/
http://www.yourarticlelibrary.com/marketing/marketing-
research-meaning-definition-and-objectives-explained/25862/
https://libraryresources.columbiasouthern.edu/login?url=http://s
earch.ebscohost.com/login.aspx?direct=true&db=bth&AN=2554
392&site=ehost-live&scope=site
https://libraryresources.columbiasouthern.edu/login?url=http://s
11. Slater, S., & Yani-de-Soriano, M. (2010). Researching
consumers in multicultural societies: Emerging
methodological issues. Journal of Marketing Management,
16(11–12), 18. Retrieved from
https://libraryresources.columbiasouthern.edu/login?url=http://s
earch.ebscohost.com/login.aspx?direc
t=true&db=bth&AN=54302234&site=ehost-live&scope=site
Unit Lesson
The unit lessons for this course are presented through
interactive presentations.
Unit I lesson
Unit I lesson PDF
Suggested Unit Resources
12. In order to access the following resource, click the link below:
This is a very quick read on a very costly mistake Coca-Cola
made because of a marketing research error. It
shows the importance of accurate marketing research.
Smith, S. (2013). Coca-Cola lost millions because of this
market research mistake. Retrieved from
https://www.qualtrics.com/blog/coca-cola-market-research/
https://libraryresources.columbiasouthern.edu/login?url=http://s
earch.ebscohost.com/logi n.aspx?direct=true&db=bth&AN=1215
21303&site=ehost-live&scope=site
https://libraryresources.columbiasouthern.edu/login?url=http://s
earch.ebscohost.com/login.aspx?direct=true&db=bth&AN=1215
21303&site=ehost-live&scope=site
https://libraryresources.columbiasou thern.edu/login?url=http://s
earch.ebscohost.com/login.aspx?direct=true&db=bth&AN=5430
2234&site=ehost-live&scope=site
https://libraryresources.columbiasouthern.edu/login?url=http://s
earch.ebscohost.com/login.aspx?direct=true&db=bth&AN=5430
2234&site=ehost-live&scope=site
https://online.columbiasouthern.edu/CSU_Content/Courses/Busi
ness/MAR/MAR3231/17G/Captivate%20Presentations/Unit_I/m
ultiscreen.html
https://online.columbiasouthern.edu/CSU_Content/Courses/Busi
ness/MAR/MAR3231/17G/Captivate%20Presentations/Unit_I/M
AR_3231_Unit_I.pdf
https://www.qualtrics.com/blog/coca-cola-market-research/
https://online.columbiasouthern.edu/CSU_Content/Courses/Busi
ness/MAR/MAR3231/17G/Captivate%20Presentations/Unit_I/m
ultiscreen.html
13. In te rn a tio n a l J o u rn a l o f M a rk e t Research Vol. 58
Issue 4
The declining use o f the term
market research
An empirical analysis
Daniel Nunan
Birkbeck, University o f London
This paper analyses the use of the term ‘market research’ in a
contemporary
context. Although the term is well established as an industry
definition, its use
and meaning have become increasingly contested. This study
brings together
empirical data from a range of sources that reflect key
stakeholders within the
market research sector. Findings suggest that the term ‘market
research’ has
become increasingly marginalised amongst these key
stakeholders. Few of the
leading research firms use this term to describe their core
activity, and data suggest
that wider use of the term has declined over the past decade.
Where ‘market
research’ is used, the term is typically demoted to describing a
set of skills rather
than a strategic concept around adding value. A number of
explanations for this
are explored, including isomorphism among research firms, the
role of research
in generating value, and the broader economic context in which
15. and practitioner market researchers (Boddy & Croft 2007), the
impact of
new technology (Cooke &c Buckley 2008) and a lack of impact
of research
among clients (Lewis 2012). However, there is nothing new
about this
self-evaluation, nor are the themes that are being evaluated new
within the
market research literature. Reviewing journals from 70 years
ago a similar
set of issues emerge. These include concerns over the shortage
of suitably
qualified people entering the field of market research (Arnold
1947),
precipitous declines in response rates and quality (Warrington
1940), and
questions over whether researchers are focusing on the most
important
and relevant problems of the day (Hirose 1945). Of course, the
contexts of
these concerns were very different. Warrington was concerned
that it was
no longer possible to achieve 90% response rates on survey,
while Hirose
highlighted the needs of researchers to address the reality of the
post-war
economy.
This paper follows in the tradition of such analysis, but
addresses a
question that is central to the concept of market research yet has
evaded
recent study: how is the term ‘market research’ itself used? To
be clear,
the paper does not seek to call into question the future of
research itself.
16. Collecting, analysing and understanding customer data,
whatever it is
labelled, has become a key part of contem porary corporate
strategy.
Beyond this there has been increased interest from government
in the
potential of data as a tool through which to gain economic
advantage. All
these trends suggest that, to borrow the tagline of the M arket
Research
Society, more people than ever believe that evidence matters.
Rather,
the purpose of this paper is to study in a very specific way the
use and
meaning of the term ‘market research’. The paper does not seek
to be
an opinion piece, but serves as an empirically informed analysis
of the
concept of market research in 2016. As the MRS approaches its
70th
anniversary, and in the light of the significant changes that have
been
encountered within the research world over the past decade,
there is
considerable merit in visiting the question. While analysing the
label
used to describe an industry may seem unim portant, or even
indulgent,
what we call ourselves has a key role in the development of
professional
identify. As such, any mismatch between the work that
researchers
undertake and the labels ascribed to the profession are likely to
impact
upon this identity (Pratt et al. 2006).
17. This paper is structured as follows. First, an overview of the
varying
current and historical uses of the term market research is
provided. Second,
5 0 4
International Journal of Market Research Vol. 58 Issue 4
analysis of data from a range of research stakeholders is used to
map the
current usage of the term. Finally, a number of potential
explanations and
implications for these findings are presented.
W h a t d o w e m ean w h en w e say m a rk e t research?
It is a characteristic of the history of commerce that the
historical narrative
stretches back only a little further than personal memories. In
one sense
the professional history of market research could be defined by
the
establishment of professional associations such as the MRS and
ESOMAR
in the late 1940s. However, the term market research was
already in
wide enough use in 1950 for a history to be published in the
Journal o f
Marketing (Lockley 1950), which was then, and arguably still is
now, one
of the leading marketing journals. This paper identified that
opinion polls
in the US can be traced back to the 1820s and that
18. questionnaires were
being widely used to gauge consumer opinions of advertising as
early as
the 1890s. In terms of a more formal acceptance of market
research as a
field, rather than of use of specific techniques, Lockley notes
that, ‘It was
not until 1910 that evidences of market research become
frequent enough
to indicate that a new field of business activity had made a
serious start.
The idea of what later was called market research became
endemic to the
period between 1910 and 1920 (1950, p. 733).
There is some agreement (Blankenship 1949; Lockley 1950)
that, for
practical purposes, the market research sector was well
established by
the early 1930s. Despite the very limited number of marketing
journals
being published at the time there were more than 150 articles
discussing
either market research or marketing research published between
1930
and 1950, reflecting the centrality of the term to marketing
thought.
Inevitably, there soon emerged a discussion over the definition
of terms,
and the difference between market research and marketing
research, as
well as the difference between a survey-research firm and
market-research
firm. Blankenship questioned the need for so many definitions
due to the
risk of confusing research users: ‘let’s give the businessman
19. some simple
definition of marketing research which he can use with full
understanding’
(1949, p. 305). This is evidently a debate that has yet to be
resolved,
with much the same sentiment being communicated by Poynter
(2014)
more than 60 years later when he argues that, in practical terms,
they
mean the same thing. Unfortunately for Poynter, and perhaps
common
sense, at the time of writing there are more than 5,000 websites
claiming
to provide some form of clarification of the definition of the
two terms.
5 0 5
The declining use of the term market research
Definitions typically suggest that marketing research is a
broader term
implying understanding of customers that happens within
organisations
and marketing functions. One the other hand, market research is
a more
macro analysis reflecting work done by research firms. It is
notable that
while professional associations, such as the AMA or ESOMAR,
provide
their own (lengthy) definitions, these seem designed more as a
means of
marking professional territory than providing clarity to
researchers or
20. users of research. In practice the delineation might also be as
much a
difference between the use of market research in the UK and
marketing
research in the US.
In the late 1990s an additional term, ‘insight’, began to emerge
in the
literature, both in addition to and as an alternative to market
research.
This can be seen as a response to the need to generate greater
influence
among stakeholders, particular among senior management. The
term
customer insight emerged not from the field of market research
but from
the emerging field of customer relationship management
(Hirschowitz
2001), where proponents were aware that they needed to sell
both the
technology and the benefits it would bring to the bottom line.
Its adoption
by market researchers can be related to the declining influence
of the
marketing function within marketing as a whole since the 1990s.
In an
influential paper, Webster et al. (2005) highlight the scope of
the problem:
‘many elements of the central marketing function have been
“centrifuged”
outward and embedded in functions as diverse as field sales and
product
engineering that are closer to customers. Today, marketing in
many large
companies is less of a department and more a diaspora of skills
and
21. capabilities spread across and even outside the organization.’
Thus, as marketing transforms from a function to a set of skills,
so
market research becomes pushed further from the centre of
decision
making. The clients for market research become more detached
from
the core decision-making process, putting the onus on market
research
organisations to demonstrate their value more clearly. However,
the
definition of customer insight as ‘knowledge about the customer
that
is valuable for the firm’ (Smith et al. 2006) is a broad one and
market
research becomes just one of many sources of data that can be
used as part
of the process of generating insight (Said et al. 2015).
Method
To analyse the use of the term ‘market research’ I looked for
data
representing views of a number of stakeholders. Such key
stakeholder
506
I n t e r n a t i o n a l J o u r n a l o f M a r k e t R e s e a r c h
V o l. 5 8 Is s u e 4
groups are identified as market research firms, professional
bodies,
academic researchers and the more general public perception of
22. research.
Although clients are clearly a valuable stakeholder group, they
were not
included in this selection due to lack of availability of suitable
data sources.
Having said that, it is possible to infer some of the client
position through
analysis of data on research firms. In terms of time period, the
focus of the
report was on data from the beginning of 2005 to the end of
2015. There
is also an advantage in terms of method as this encompasses the
period for
which reliable data are available from internet sources. Data
were gathered
from the sources described below.
Analysis o f industry league table data
First, the most recently available league table from the MRS at
the time
of analysis (January 2016) was used.1 An alternative source
would have
been the American Marketing Association Top 50 global market
research
firms table, but in choosing the MRS table the study was able to
apply a
consistent analysis on a single country. A comparison of the
data from the
UK firms in the AMA table shows that the reported earnings are
similar.
The MRS league tables contain both ‘largest’ and ‘fastest
growing’ and,
within these categories, there are two further divisions into
‘consolidated’
and ‘individual agencies’. This reflects the nature of business
23. models
with the marketing services sector where agencies are often part
of a
group ownership structure under larger holding companies.
Given the
mix of companies, these league tables present an inevitable
compromise
in drawing a line between individual and parent ownership.
Given the
complexity of this exercise the question of parent company is
not always
consistent. For example, the parent company of Millward Brown
is listed
as Kantar rather than WPP, while Ipsos is not listed as the
parent of Ipsos
MORI. More pragmatically, Tesco is not listed as the parent
company of
dunnhumby. This creates another challenge in that UK-located,
or focused,
firms may carry out overseas work from their UK offices.
Nevertheless,
the MRS league table can be regarded as the closest thing to a
definitive
account of firm size that is available within the UK market, and
similarities
with the similar AMA league table can be taken as an indicator
of reliability.
To ensure the maximum value of findings in this study, both the
parent
company and individual agency rankings were analysed. For
example, it
might be expected that market research could exist as a business
unit within
1 The MRS 2014 Research and Insight Industry League Tables
24. UK, compiled in October 2015.
5 0 7
T he d e c lin in g use o f th e te r m m a rk e t research
a larger firm, therefore the individual agency rankings would
contain firms
that make far more use of the term market research. On the
other hand,
as the larger list is more comprehensive it could be seen as
being a more
accurate reflection of the overall market.
The consolidated league table contains 86 firms with revenue
ranging
from £3.2 million to £464 million, while the independent firm
table contains
50 firms with revenue ranging from £11.2 million to £227
million. For the
purposes of this study we analysed the 50 largest firms in the
consolidated
league table as these provided nearly 95% of all revenue from
the group
of firms on the list.
For each of these firms’ public-facing websites the following
were
analysed:
• presence of the term ‘market research’ in a description of the
firm in the
page title (i.e. the text that shows in search results) or on the
home page
25. • presence of the term ‘market research’ in description of the
companies’
core activities on an ‘about’ page, or similar page explaining
core
company activities
• presence of the term ‘market research’ in a business unit or set
of
product offerings
• alternative phrases used instead of, or in addition to, market
research
to describe core business activities.
A n a lysis o f use o f a c a d e m ic lite ra tu re o n m a rk e
t research
To analyse the academic research there are a number of
challenges in
identifying the boundaries of the data set. For example, there is
the
question of what counts as academic research. This would
normally be
defined as including articles in peer reviewed journals, but there
are also
the cases of working papers, books, monogr aphs, conference
proceedings
and the grey literature. In going back over historical data there
is a further
challenge in that both the volume and coverage of academic
literature
within online databases has changed significantly over the past
ten years.
For example, while the Google Scholar service contains a
probably
26. unmatched database of recent academic publications it has a
bias towards
publication formats that are available publicly and online,
increasing
the volume of publications in recent years. This is compounded
by the
large increase in the number of academic publications within
the field of
management, many of which lack effective peer review or
quality control
- often described as predatory journals.
5 0 8
In tern ation al Journal o f M arket Research Vol. 58 Issue 4
Judging the academic use of ‘market research’ also required
access to a
number of granular search options. Simply searching for the
term ‘market
research’ within the text of papers would not be effective as the
term is
likely to appear in the context of many discussions around
marketing. It
might also appear as a citation to a journal that has ‘market
research’ in the
title. On the other hand, simply searching for the term in the
title would
be insufficient as not all the subject keywords relevant to the
paper need
to be in the title.
To effectively address these issues, the EBSCO Business Source
Complete
27. database was used. This is a widely used database of business
and
professional articles that indexes more than 5,000 business
journals. While
not as broad as Google Scholar it provides a range of advanced
search
capabilities and also contains appropriate metadata, allowing
searches
within specific sections of the article such as title or abstract.
For each year
from 2005 to 2015 the total number of articles featuring the
term ‘market
research’ in either the abstract or title were selected. Only
English-language
articles were selected, although in practice only a very small
number of
articles were published in languages other than English given
the focus on
publishing in English within the international marketing
literature. Finally,
only articles that were published in peer reviewed journals were
selected.
Analysis o f public search trends
Analysis of search trends gives some measure of the popularity
of a concept,
although actual search volume data are closely guarded by
search companies.
However, Google provides a means of getting a close
approximation of this
data through its Google Trends tool. Google Trends is a web
service that
provides some estimation of the volume of search terms on the
Google
Search Engine over time. Two caveats needed to be provided
28. with use of
this data. First, only a normalised value is provided, giving a
number that
reflects a proportion of searches over time rather than an
absolute value.
This is important as the value for a particular search term may
fall over
time even if the absolute number of searches increases, due to
an increase
in overall search volumes. Second, when searching for the term
‘market
research’, some form of fuzzy matching is provided within the
algorithm to
associate misspelling and synonyms with the term. As such,
Google Trends
also provides some contextual information when searching for
words to
ensure that an appropriate match is found. In this case two
categories were
returned: ‘market research’ [search term] and ‘market research’
[industry].
Both terms were therefore used in this analysis.
5 0 9
The declining use o f the term market research
Issues in analysis o f historical data
The challenge of longitudinal analysis is in controlling for the
many
significant variables that could have influenced the analysis.
While
longitudinal data were available for the analysis of publications
29. and search
trends, it was not possible to analyse information available on
market
research company websites. Although services such as the
Internet Archive
enable the access of historical information, data on many firms
is missing
or incomplete. With changes in ownership, rebranding and
technical
changes to websites, the availability of sufficiently comparable
data was
seen as too limited to allow for this analysis.
Findings
Analysis o f leading firms
While summary data are contained in this section the full data
set of
firms listed and analysis of the core characteristics stated
previously
are contained in Appendix 1 (consolidated data set) and
Appendix 2
(individual agency data set). Table 1 shows the percentage of
firms in
the list for which the term market research featured on the
appropriate
corporate website in relation to the criteria. An additional set of
data is
given in the fourth column, showing the percentage of ‘Top 5 0
’ research
firms that use the term market research in a context that meets
any of
the criteria. One unexpected result was that the term market
research
was slightly more likely to be used on the websites of parent or
30. holding
companies than those of individual agencies. One explanation
for this
is that holding company sites are more likely to be orientated
towards
a different set of stakeholders, e.g. job seekers and investors
rather than
potential clients and customers.
T a b le 1 Usage o f th e te rm 'm a rk e t research' b y le a d
in g firm s
% o f firm s in % o f firm s in
C o n s o lid a te d Table In d iv id u a l T a ble
C riteria (n = 50 ) (n = 50 )
Presence o f th e te rm 'm a rke t research'in a d e scription
o f th e firm on th e title page o r hom e page
28 32
Presence o f th e te rm on a b o u t/c o re a ctivities page 30
24
Presence o f th e te rm in p ro d u c t d e scription 34 30
Presence o f th e te rm in any o f the above criteria 56 50
5 1 0
Intern ation al Journal o f M arket Research Vol. 58 Issue 4
These data provide only a partial picture in that they measure
31. the
presence or absence of the use of ‘market research’ in certain
contexts.
Taking the most positive view, around half of firms use the term
market
research in some context to describe the activities of their firm.
Bearing in
mind an uneven distribution of revenue between firms, Table 2
indicates
the distribution if weighted for company size.
Although there is some variance within results, the data could
be
summarised by saying that around half of the leading market
research
firms do not mention the term market research in a core way on
their
website. Additionally, no more than a third of leading firms,
representing
perhaps one-fifth of sector revenue, make prominent use of the
term
market research. However, it is necessary to acknowledge that,
on its own,
this is a crude classification. By looking simply at mentions of
the term
‘market research’, the context in which the term was used is not
captured,
nor is the frequency of use.
Additional analysis was carried out to dig deeper into these
findings.
Individual firms are not highlighted as it is not the intent of this
paper to
‘call out’ firms for the way in which the term ‘market research’
is used.
First, for those firms that did not meet any of the criteria, a
32. more general
search (using the Google site, Search Operator) was carried out
to find
out if the term appeared anywhere on their site. For all but three
firms the
term was identified on the website, and in the cases where it
was not this
is most likely to be a technical restriction where the full content
of the site
was not indexable, rather than a reflection of site content.
Where market
research was mentioned it appeared in three categories, all of
which were
secondary to the core services provided by the firm:
1. in the context of detailing the Market Research Society Code
of
Conduct in the terms and conditions or other ‘small print’
T a b le 2 Usage o f th e te r m 'm a r k e t re s e a rc h 'b y le
a d in g firm s (w e ig h te d fo r re venue)
% of firms
% o f firms in turnover in
Consolidated Table Individual Table
Criteria (n = 50) (n = 50)
Presence o f th e te rm 'm a rke t research'in a d e scription
o f th e firm on th e title page o r ho m e page
26% (£642m) 40% (£925m)
Presence o f th e te rm on a b o u t/c o re a ctivitie s page 21%
(£539m) 22% (£500m)
Presence o f th e te rm in p ro d u c t d e scription 27%
(£625m) 28% (£650m)
33. Presence o f th e te rm in any o f the above criteria 41%
(£1035m) 54% (£1.255bn)
511
The declining use o f the term market research
2. in the context of a press release or announcement about being
nominated
for or winning a Market Research Society or other sector award
3. in the context of a job advertisement where prior skills in
market
research had been identified as part of the requirement for a
role.
Second, for those firms where market research was mentioned,
the
prominence or circumstances where it was mentioned were
analysed. Of the
17 firms that met the first criteria of appearing on the home or
title page,
only seven used the term market research in the title or meta-
data rather
than being visible. This suggests that the use of the phrase
market research
was for the purposes of search engine ranking and optimisation
rather than
a key description of business. Similarly, of the 17 firms that
used the term
market research in their ‘about’ page or as a product
description, in only
eight cases was it used as a primary identifier of the firm’s
business.
34. Finally, we present a summary set of phrases that are used on
websites
as alternatives to the term market research (Table 3). A number
of
duplicates have been removed from this list.
Table 3 D escriptions o f core business a ctivity
Research, data and insight
Commercial intelligence, research and consulting
Customer science
Business intelligence
Technology research
Intelligence and analytics
Marketing analytics
Analytics and advice
Market insights and intelligence
Strategic insight
Social research
Understanding o f consumer decision making
Generating insight
Insight
Research
Consumer insight
Emerging markets research
Survey research
35. Market intelligence
Customer experience
Usage o f the term in academic research
The data from use of the term in academic research are
highlighted in
Figure 1. This chart provides data on peer-reviewed articles that
were
published between 2005 and 2015. Although this shows a slight
upward
trend, it can be explained by an increase in the publication
frequency of
the International Journal o f Market Research. In most years the
range of
publication is between 80 and 100 articles, representing a total
of 980
articles over the period. Through this 11-year period, 166
articles (17% of
51 2
International Journal o f Market Research Vol. 58 Issue 4
137.5
110
Figure 1 Number o f articles published in academic journals that
feature the term 'market
research'(or equivalent) in the title
the total) were published in the International Journal o f Market
36. Research
and, with the exception of Qualitative Market Research: An
International
Journal (19 articles), the remaining articles were spread across
more than
100 journals, most with fewer than five articles published in
each. While
the search database allowed for the selection of content that
appeared only
in peer reviewed journals, it did not have the ability to delineate
between
peer reviewed and non-peer reviewed sections of articles. As a
result much
editorial, book review and non peer-reviewed content, such as
‘viewpoint’
articles, are also included, inflating the count. More than half of
the articles
assigned to the International Journal o f Market Research are in
this category.
This issue did not appear to impact other journals. As with the
analysis of
text on company websites a challenge was created where the
presence of
text does not necessarily infer the same meaning of text. For
example, many
of the papers published used the term ‘market research’ in the
abstract even
though the paper was tangential to the subject matter of market
research,
suggesting that it was a strategic choice to signal to an editor
the relevance
of the paper to their journal. One final piece of additional
analysis was
the number of articles with the term market research in the title,
the total
coming to 132 over the 11-year period. Of these articles, 47
37. were published
5 1 3
The d e c lin in g use o f th e te rm m a rk e t research
in the two ‘main’ market research journals, International
Journal o f Market
Research and Qualitative Market Research: An International
Journal, with
the remaining number spread across a large number of journals,
most with
no more than one or two publications per journal.
Overall, this suggests that the use of the term ‘market research’
within
academic research has been relatively static over the past
decade and
concentrated within the two journals that have the phrase
‘market research’
in their titles. While the number of articles appears steady in
absolute terms
this should be seen in the context of the relatively small number
of articles
published that have market research as a core premise in their
titles (c.10
per year, not dissimilar to the number published in the 1940s)
and the very
large increase in the number of management journals over this
period. For
example, the Association o f Business Schools Academic
Journal Quality
Guide increased the number of journals it ranked by 88% from
745 in
38. 2010 to 1,401 in 2015 (ABS 2015). In addition to this, there are
many
thousands of ‘unranked’ and lower-quality journals. This
suggests that the
numbers presented above should be seen in the context of the
many tens of
thousands of articles published in management journals each
year. In other
words, in terms of academic publishing, market research as a
distinct focus
can be seen as an increasingly niche activity.
Search trend analysis
Trend analysis from Google Trends is shown in Figure 2. The
solid line
refers to searches for the specific term ‘market research’, while
the dashed
line refers to the more general category of market research
[industry] that
matches similar search terms. While absolute data aren’t
provided, the
numbers for relative interest can be determined, with the market
research
industry at 100 in February 2005 and 17 in December 2015,
meaning that
relative search interest in December 2015 was approximately
one-fifth of
that ten years previously.2
As the data are relative, a more useful comparison might be
with some
similar terms. For the purposes of comparison we also present a
second
graph (Figure 3), for the terms ‘analytics’ (in solid tint) and
‘Big Data’ (in
39. dashed tint).
Although caution must be exercised, given the lack of absolute
data, this
does give some indication of a significant, relative decline in
the use of the
term market research in searches.
2 Data from this search can be generated from the Google
Trends site at www.google.co.uk/trends.
5 1 4
http://www.google.co.uk/trends
Intern ation al Journal o f M arket Research Vol. 58 Issue 4
«
/
Figure 2 G o o g le T r e n d s a n a ly s is o f t h e t e r m m
a r k e t re s e a rc h
Figure 3 G o o g le T r e n d s c o m p a r a t iv e a n a ly s is
Discussion
Overall the findings point to a picture where the majority of
firms in
the study, at least when communicating to external
stakeholders, seek to
define themselves as something other than a market research
firm. The
question is, why? The data suggest that this trend is
unambiguous, but it is
also something that has emerged rather than been planned. As
40. the purpose
of this paper is to generate discussion rather than to pass
judgement on
the way that firms describe their business activity, a number of
potential
suggestions are now proposed.
Isomorphism
Isomorphism (Di Maggio & Powell 1983) refers to the
phenomenon
whereby pressures within industries cause firms within the same
industry
to adopt similar behaviours. In this study the use of data from
websites
5 1 5
The declining use o f the term market research
shows only one aspect of the way in which firms wish to portray
themselves and, as we are not dealing with e-commerce firms,
the customer
relationship is likely to be primarily formed by other types of
face-to-face
interaction. Therefore, the website could be seen as performing
a signalling
role whereby smaller firms mimic the language and tone of
larger firms in
order to demonstrate that they are competitive. Such techniques
have been
shown to have an important role in the way that the positioning
of firms is
judged (Oberg et al. 2009). A qualitative observation from this
41. analysis is
that there is significant homogeneity within the ways that firms
described
themselves on their websites, as well as the structure, tone and
overall
approach to the brand. Given the financial dominance of a small
number
of larger firms it is possible that only a smaller shift away from
using the
term market research by larger firms may have created wide
isomorphic
pressure.
Creating and demonstrating value
In relation to the question of value, it is possible that the
research sector
has been a victim of its own success. The concept of evidence-
based
decision making has spread throughout management, at least as
an idea
even if it is not widely put into practice. This is a parallel to the
changes
that have impacted marketing as a whole, as highlighted by the
earlier
comments on the diffusion of marketing (Webster et al. 2005).
How many
CEOs today would stand up and admit that they were more
interested
in operations and production than in customers? Some, perhaps,
but
not the majority. As the marketing concept has become more
diffuse the
professional identity of marketing has weakened. This is not by
any means
unique to marketing. Much contemporary management practice
42. is built
around the idea of agile organisations that are not hindered by
previous
organisational, professional and operational silos. What were
previously
thought of as departments or functions, and by extension career
paths,
are now sets of professional skills. As a consequence, the role
of research
as a creator of value has been subsumed within a more complex
notion
of insight.
The changing economic context
One very practical explanation might be found in the economic
context
surrounding market research. In the first half of the 20th
century
manufacturing was often concerned with producing enough to
meet
5 1 6
I n t e r n a t i o n a l J o u r n a l o f M a r k e t R e s e a r c h
V o l. 5 8 Is s u e 4
demand. The need for new markets was limited by the absence
of sufficient
production volume to meet the needs of existing markets.
During the
post-war period an improvement in production efficiency, and a
nascent
growth in the ability to export finished products rather than just
43. raw
materials, led to the pursuit of new markets becoming a key
strategic goal.
Thus, the term market research was closely aligned with the
strategic goals
of organisations. As the scope and function of research has
increased,
and shifted, the term market research has become overly
associated with
corporate strategies that have themselves become dated. The
focus of the
economic context within strategic decision making has therefore
moved
away from markets and towards individual consumers, and the
change in
language used by firms reflects this.
Conclusion an d fu tu re qu estion s
Short sighted managements often fail to understand that there is
no such thing
as a growth industry ... The reason they defined their industry
wrong is because
they were product oriented rather than customer oriented.
(Levitt 1960, p. 45)
In his classic paper ‘Marketing myopia’, the late Harvard
academic Ted
Levitt (1960) highlighted the strategic risks in being unable to
answer the
question ‘What industry am I in?’ More than half a century and
thousands
of citations later, the central point of this paper remains clear
and prescient.
Managers who are unable to address this question define their
industries
44. too narrowly, become complacent over success, and fail to
respond and
redefine their industries in the face of external change. Of
course, there is
a paradox in using this example in the context of market
research. Levitt’s
paper has been said to have heralded the start of the modern
marketing
revolution - the idea that customer orientation was of strategic
importance
to organisations. More than anything, it might be said that this
message
enabled the widespread growth of market research through
persuading
senior managers of the importance of understanding customer
needs and
looking towards new markets.
What is its relevance in 2016? The evidence suggests that the
use of
the term market research today is as a placeholder or a
descriptor of a
certain set of skills, and it appears to be largely used within the
industry.
The potential for research in commerce has, arguably, never
been greater.
However, the term market research is at the intersection of a
number of
declining factors, not least a perception of the decline in the
value it can
generate for its customers. This raises a number of important
questions
5 1 7
45. The d e c lin in g use o f th e te r m m a rk e t research
that need to be addressed by key stakeholders in research. These
include
the following.
• Should universities continue to offer courses that are labelled
‘market
research’?
• Should new graduates and other young professionals seek
skills to
develop a career in market research?
• Should market research journals and professional associations
consider
renaming themselves?
At this stage it would be tempting to make a recommendation
that the
market research world needs to reinvent itself, but the data
analysed in
this paper suggest that the industry is already doing so. In
conclusion, the
market research ‘industry’ has moved on and is increasingly
defining itself
in other ways - and, in showing this flexibility, it helps to
secure its future
success.
A p p e n d ix 1: C o n s o lid a t e d lis t o f la r g e s t m a r
k e t r e s e a r c h f ir m s
2014
ranking
46. Company
name
2014
(£m)
MR on
home
p ag e /title
MR on
'about'
page
MR in
product
description
1 K a n ta r 46 4 .1 7 2 X X X
2 W o o d M a c k e n z ie Research & C o n s u ltin g 227.421
X X X
3 D u n n h u m b y 191.553 X X X
4 IpsosM O RI 165.200 Y Y Y
5 G a rtn e r 138.487 X X X
6 G fk 136.579 X X Y
7 N ielse n 120.715 Y X X
47. 8 E u ro m o n ito r 88.295 Y Y Y
9 M in te i G ro u p 74.940 Y Y Y
10 O m n ic o m /D A S 74.021 X X Y
11 E b iq u ity 68.452 X X X
12 In fo rm a 66.932 X X X
13 I n fo rm a tio n R esources 39.107 X X X
14 C e llo G ro u p 38.001 X Y Y
15 IMS H e a lth 36.167 X X X
16 R esearch N o w 34.463 Y X X
(c o n tin u e d )
5 1 8
Intern ation al Journal o f M arket Research Vol. 58 Issue 4
2014
ranking
Company
name
2014
(£m)
MR on
48. home
p ag e/title
MR on
'about'
page
MR in
product
description
17 NatCen 34.154 X X X
18 MMR Group 28.100 Y X X
19 D ouble Helix 27.500 Y X X
20 BDRC Group 24.654 Y X X
21 Creston Insight 23.127 X Y X
22 The Research Partnership 21.017 Y Y Y
23 Toluna 20.991 X X X
24 W orldO ne Research 20.164 X y Y
25 Experian Consum er Insight 19.645 X X X
26 YouGov 19.400 X Y Y
27 Chim e 18.897 X X X
28 Frost & Sullivan 18.824 X Y Y
49. 29 SPA Future T h in kin g 16.896 X X X
30 C o n firm it 16.354 Y Y Y
31 Forrester Research 16.080 X X Y
32 Hay G roup Insight 15.000 X X X
33 IDC 14.790 X X X
34 MaritzCX 13.727 X X X
35 comScore 13.500 X X Y
36 M arkit Economics 12.950 X X X
37 Incite M arke tin g Planning 12.620 X X X
38 Business Research G roup 12.230 Y Y X
39 Populus Group 12.187 X X X
40 ORC International 12.119 X X X
41 M arket Force In fo rm a tio n 11.550 X Y Y
42 Q uadrangle 11.509 X X X
43 Firefish 11.237 X X X
44 G allup O rganisation 11.117 Y X X
45 The Planning Shop In te rn a tio n al 11.079 X Y X
46 Footfall 10.630 X X X
50. 47 Strategy Analytics 10.138 X Y Y
48 TTi G lobal Research 9.082 Y X Y
49 Brainjuicer 8.917 Y Y Y
50 KPMG N u n w o o d C onsulting 8.422 X X X
519
The declining use o f the term market research
Appendix 2: List of largest individual agencies
2014
ranking Company name
2014
(£m)
MR on
home page/
title
MR in
'about'
page
MR in
product
description
51. 1 Wood Mackenzie Research &
Consulting
227.421 X X X
2 TNS UK 195.334 Y X X
3 Dunnhumby 191.553 X X X
4 IpsosMORI 165.200 Y Y Y
5 Gartner 138.487 X X X
6 ACNielsen 101.534 Y X X
7 Millward Brown 101.101 X X X
8 Gfk NOP 94.096 Y X Y
9 Euromonitor 88.295 Y Y Y
10 Mintel Group 74.940 Y Y Y
11 Ebiquity 68.452 X X X
12 Kantar Media 52.730 X X X
13 Precise Media M onitoring 50.886 X X Y
14 Informa Telecoms & Media 42.104 X X X
15 Information Resources 39.107 X X X
16 IMS Health 36.167 X X X
17 Lightspeed Research 35.869 Y Y X
52. 18 Research Now 34.463 Y X X
19 NatCen 34.154 X X X
20 Hall & Partners 27.727 X X X
21 Double Helix 27.500 Y X X
22 Datamonitor 24.848 X X X
23 Flamingo Research 24.044 X X X
24 Gfk Retail & Technology 23.828 X X Y
25 The Research Partnership 21.017 Y Y Y
26 Toluna 20.991 X X X
27 MMR Research W orldwide 20.200 Y X X
28 WorldOne Research 20.164 X y Y
29 Experian Consumer Insight 19.645 X X X
30 YouGov 19.400 X Y Y
31 Frost & Sullivan 18.824 X Y Y
32 SPA Future Thinking 16.896 X X X
(continued)
S20
53. International Journal of Market Research Vol. 58 Issue 4
2014
ranking Company name
2014
(£m)
MR on
hom e page/
title
MR in
'about'
page
MR in
product
description
33 Confirmit 16.354 Y Y Y
34 Cello Health Insight 16.228 X Y Y
35 Forrester Research 16.080 X X Y
36 Hay Group Insight 15.000 X X X
37 I DC 14.790 X X X
38 MaritzCX 13.727 X X X
39 Nielsen Media Research 13.671 Y X X
54. 40 comScore 13.500 X X Y
41 BDRC Continental 13.155 Y X X
42 Markit Economics 12.950 X X X
43 Incite Marketing Planning 12.620 X X X
44 Business Research Group 12.230 Y Y X
45 Populus Group 12.187 X X X
46 ORC International 12.119 X X X
47 Market Force Information 11.550 X Y Y
48 Quadrangle 11.509 X X X
49 2CV 11.359 Y X X
50 Firefish 11.237 X X X
R e f e r e n c e s
Arnold, P. (1947) Woman’s role in market research .Journal o f
Marketing, 12, 1, pp. 87-91.
Association of Business Schools (ABS) (2015) Academic
Journal Guide 2015. Available online
at: http://charteredabs.org/academic-journal-guide-2015/
(accessed 13 January 2016).
Blankenship, A. (1949) Needed: a broader concept of marketing
research .Journal o f
Marketing, 13, 3, p. 305.
Boddy, C. & Croft, R. (2007) The strength of British market
55. research is British market
researchers: a reply to Piercy. International journal o f Market
Research, 49, 3, pp. 287-298.
Cooke, M. & Buckley, N. (2008) Web 2.0, social networks and
the future of market research.
International Journal o f Market Research, 50, 2, 267-292.
DiMaggio, P. & Powell, W. (1983) The iron cage revisited:
institutional isomorphism and
collective rationality in organizational fields. American
Sociological Review, 48, pp. 147-
160.
Freidson, E. (2001) Professionalism: The Third Logic. Chicago,
IL: University of Chicago
Press.
Hirose, A. (1945) Market research as a practical help in
reconversion problems. Journal o f
Marketing, 9, 4, p. 342.
Hirschowitz, A. (2001) Closing the CRM loop: the 21st century
marketer’s challenge:
transforming customer insight into customer value. Journal o f
Targeting, Measurement and
Analysis for Marketing, 10, 2, pp. 168-178.
521
http://charteredabs.org/academic-journal-guide-2015/
The declining use o f the term market research
56. Le Farm, J. (2011) The Rise and Fall o f Modern Medicine.
London: Abacus.
Levitt, T. (1960) Marketing Myopia. Harvard Business Review,
38, 4 (July-August), pp. 45-56.
Lewis, I. (2012) The future of market research. International
Journal o f Market Research, 54, 1,
pp. 11-13.
Lockey, L. (1950) Notes on the history of marketing research.
Journal o f Marketing, 14, 5,
pp. 733-736.
Oberg, A., Schoellhorn, T. & Woywode, M. (2009) Isomorphism
in organisational self-
representation in the World Wide Web? Institutionalisation
process regarding internet
presentation of organisations. Academy o f Management
Proceedings, 1, pp. 1-6.
Poynter, R. (2014) Marketing research or market research? An
unhelpful distinction. NewMR
Blog. Available online at: http://newmr.org/blog/marketing-
research-or-market-research-a-
unhelpful-distinction/ (accessed 13 January 2016).
Pratt, M., Rockmann, K. & Kaufmann, J. (2006) Constructing
professional identity: the
role of work and identity learning cycles in the customization of
identity among medical
residents. Academy o f Management Journal, 49, 2, pp. 235-
262.
Said, E., Macdonald, E., Wilson, H. & Marcos, J. (2015) How
organisations generate and use
customer insight. Journal o f Marketing Management, 31, 9-10,
57. pp. 1158-1179.
Smith, B., Wilson, H. & Clark, M. (2006) Creating and using
customer insight: 12 rules of best
practice. Journal o f Medical Marketing, 6, 2, pp. 135-139.
Tallis, R. (2004) Hippocratic Oaths. London: Atlantic Books.
Warrington, A. (1940) Diminishing returns in market research.
Journal o f Marketing, 5, 2,
p. 167.
Webster, F., Maker, A. & Ganesan, S. (2005) The decline and
dispersion of marketing
competence. MIT Sloan Management Review, 46, 4, pp. 34-43.
A b o u t th e a u th o r
Daniel Nunan is a lecturer in management at Birkbeck,
University of
London. He has published in international journals including the
journal o f
Business Ethics, N ew Technology, Work and Employment, the
International
Journal o f Market Research and the International Journal o f
Retail and
Distribution Management. He is a member of the editorial board
of the
International Journal o f Market Research and was nominated
for the
Market Research Society Silver Medal in 2012 and 2014.
Address correspondence to: Daniel Nunan, Department of
Management,
Birkbeck, University of London, Torrington Square, London,
WC1E 7JL.
58. Email: [email protected]
5 2 2
http://newmr.org/blog/marketing-research-or-market-research-a-
unhelpful-distinction/
http://newmr.org/blog/marketing-research-or-market-research-a-
unhelpful-distinction/
mailto:[email protected]
Copyright of International Journal of Market Research is the
property of Warc LTD 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.
Researching consumers in multicultural societies:
Emerging methodological issues*
Stephanie Slater, Cardiff University, UK
Mirella Yani-de-Soriano, Cardiff University, UK
Abstract The paper presents a critical review of the main studies
in cross-cultural
consumer behaviour and marketing research, and identifies the
methodological
issues that frequently undermine the quality of research in this
area. The paper
offers suggestions for addressing these issues, which are
becoming even more
59. complex due to growing Internet-based marketing research and
increasingly
multicultural societies. The authors discuss the relevance of
cross-cultural
marketing research and the challenges associated with it in the
context of a
changing global environment, and explain how, by
understanding and addressing
these concerns, marketers will be able to achieve superior
marketing research
findings through improved validity of results.
Keywords cross-cultural methodology; consumer research
Introduction
In order to understand the methodological problems associated
with cross-cultural
marketing research, the relationship between research design
and culture needs to be
investigated. All too often scholars and practitioners highlight
the importance of
understanding culture when making marketing decisions, but
fail to take account of
cultural differences when they design their marketing research
project.
We know that the theories and models relating to marketing and
consumer behaviour
have been mainly developed in an Anglo-Saxon context, notably
the United States and
the UK. However, these theories have rarely been tested in
cultures having different
languages and traditions and confronted with diverse
environmental conditions, such as
Asia, Africa, and Latin America. The reason for the lack of this
60. type of research lies
fundamentally in the difficulties arising when any method or
theory in the behavioural
sciences is applied outside its area of origin (Brislin, Lonner, &
Thorndike, 1973).
Boddewyn (1981) provides clear evidence for this argument in
his study of the first 25
years of comparative marketing. He found that in consumer
behaviour studies,
researchers used the research designs tested in the United States
without taking into
consideration the circumstances particular to other cultures.
*Both authors contributed equally to this article
ISSN 0267-257X print/ISSN 1472-1376 online
# 2010 Westburn Publishers Ltd.
DOI: 10.1080/0267257X.2010.509581
http://www.informaworld.com
Journal of Marketing Management
Vol. 26, Nos. 11–12, October 2010, 1143–1160
During the 1980s, a healthy trend towards asserting the
generalisability of the
existing consumer behaviour theories and findings emerged in
the United States.
During this period, most theory-based empirical consumer
behaviour studies
involved testing of the external validity of findings (Aulakh &
Kotabe, 1992). This
61. trend continued into the 1990s. Cross-cultural marketing
research gained great
importance due to the globalisation of marketing activities and
the cross-cultural use
of advertisements (Malhotra, Agarwal, & Peterson, 1996;
Manrai & Manrai, 1996).
Nevertheless, at the end of this decade, the number of studies in
cross-cultural
marketing research remained limited (Sin, Cheung, & Lee,
1999).
In the twenty-first century, the role of cross-cultural marketing
research has become
increasingly critical in guiding business decision making. The
increased pace of
globalisation and the dramatic advancements in technology are
creating new
challenges for firms and researchers. From a managerial point
of view, firms need to
understand the cultural variation between countries and cultures
to be able to devise
sound marketing strategies based on these variations if they are
to succeed in the global
marketplace (Craig & Douglas, 2001; Malhotra, 2001). From a
theoretical
perspective, researchers need to conduct cross-cultural
marketing research to
establish the validity and generalisability of marketing and
consumer behaviour
theories and models. The argument that it is crucial to establish
whether a theory or
model can be used outside its area of origin has been raised for
many years and in a
number of studies. More than 40 years ago, Whiting (1968)
argued that since most
sociopsychological studies were undertaken within the
62. framework of Western
European cultures, it was not possible to be certain whether the
discovered
relationships were universally valid.
In the last 20 years, significant progress has been made
regarding conceptual/
theoretical issues and empirical findings. Methodologically,
however, recurrent and
emerging issues are a key concern. To enable comparability of
data and measurement
equivalence, scholars need to establish a framework for
evaluating the validity and
rigour of data. This is paramount so that we can better
understand and address the
problems of conducting business and marketing research in
global markets, which are
often characterised by increasingly multicultural societies.
The objective of this paper is to draw closer attention to the
main methodological
issues confronting cross-cultural marketing research and the
need to address them
competently, since failure to do so could lead to confounding
alternative explanations,
limiting the usefulness of the marketing research project
(Malhotra et al., 1996). The
paper starts with a discussion of the two major general concerns
in cross-cultural
research and moves on to discuss more specific issues, both
recurrent, which are
unsolved or simply ignored, and emergent. The paper provides
guidelines for
addressing these problems. A summary of the main
methodological concerns is
presented in Table 1.
63. The cultural context
This section discusses the two major general concerns in cross -
cultural research: the
emic–etic issue and the independence of sample issue. These are
important problems
that are closely connected and need to be considered if
methodological developments
in cross-cultural research are to demonstrate cross-cultural
generalisability.
1144 Journal of Marketing Management, Volume 26
Table 1 Literature review summary of cross-cultural
methodological concerns.
Methodological
concern Rationale Key studies
Cultural
dimension
Highlights differences in social
values and culture and provide us
with an explanation as to why
companies who wish to exploit
market potential need to consider
cross-cultural influence.
Chang (2003); Chong and Park
(2003); Geletkanycz (1997);
Hofstede (1980, 1991); Malhotra
(2001); McCort and Malhotra
64. (1993); Saka (2004);
Trompenaars and Hampden
Turner (1997);
Generalisability To achieve generalisability of
existing psychological knowledge,
theory, laws and propositions.
Berry (1980); Berry, Poortinga,
Segall, and Dasen (1992); Evans
et al. (2004); Haeckel (2001);
Mitchell (2001); Sekran (1983);
Triandis (1980); Triandis,
Malpass, and Davidson (1972);
Woodside (2005)
Comparability To ensure valid comparisons of
datasets across countries/
cultures.
Aulakh and Kotabe (1992); Bhalla
and Lin (1987); Boddewyn (1981);
Douglas and Craig (1983, 2007);
Malhotra (2001)
Metric
equivalence
The psychometric properties of two
or more data sets from two or
more cultural groups exhibit
essentially the same coherence of
structure.
Berry (1980); Davis, Douglas, and
Silk (1981); Heeler and Ray
65. (1972); Sekaran (1983)
Functional
equivalence
Similar activities should have
similar functions in different
societies if parameters are
comparable.
Berry et al. (1992); Soriano and
Foxall (2002)
Conceptual
equivalence
Subjects need to have an equal
understanding or interpretation of
the meaning of behaviour,
product, or stimuli.
Barnard (1982); Berry (1980);
Brislin, Lonner, and Thorndike
(1973); Streiner and Norman
(1995); Malhotra et al. (1996);
Small et al. (1999)
Translation
equivalence
The scales and other verbal stimuli
should be translated so that they
can be understood in different
cultures and have equivalent
meaning.
66. Response bias The instrument should produce
answers that are free of response
set bias.
Baumgartner and Steenkamp
(2001); Caruanna et al. (1998);
Heide and Gronhaug (1992)
Return rate Internet survey return rate could be
lower than traditional methods as
respondents become more
apathetic to these studies.
Backmann et al. (2000)
(Continued )
Slater and Yani-de-Soriano Researching consumers in
multicultural societies 1145
The emic–etic issue
Cross-cultural research has two objectives. One is to describe a
culture by studying
specific behaviours or concepts from within such culture; this is
the emic approach.
The other objective aims at theory building, that is, to make
generalisations across
cultures that take into account all human behaviour; this is the
etic approach (Berry,
1990; Douglas & Craig, 1983; Triandis, Malpass, & Davidson,
1971). The
importance of comparative marketing studies is that they have
the potential for
67. enriching the understanding of what marketing is all about by
helping to refine and
check the generality of our existing marketing concepts,
hypotheses, and theories
(Berry, 1980; Boddewyn, 1981). Triandis (1980, p. 3) stated
that ‘it is imperative to
establish cross-cultural generalities in order to understand
whatever cultural
differences are observed’. When researchers apply an etic
approach, they are
imposing constructs developed in one culture to all cultures
(Berry, 1990; Douglas
& Craig, 1983; Triandis et al., 1971). When constructs are
imposed in this way, they
are referred to as the ‘imposed etic’ (Berry, 1969, 1990) or
‘pseudo etic’ (Triandis et al.,
1971). This is the case when theories and models developed in
the United States are
applied outside their area of origin. The problem is that the
concepts that are been
tested may have different meanings in non-Western cultures. A
derived etic approach is
essential to enable understanding of cultural variations, and it is
achieved when
researchers adapt their constructs to fit the relevant culture
under study (emic
approach) (Rogoff, 2003).
Personal construct theory (Kelly, 1955) offers support to the
claim that individuals
perceive their environment differently and that this has
implications for construct
elicitation. Translation from one language to another may not
mean that words are
perceived by the recipient as having the same meaning. Etic
characteristics have also
68. Table I. (Continued)
Methodological
concern Rationale Key studies
Ethical issues in
preventing
subject fraud
Subject fraud can be avoided by
installing cookies. in respondents’
browsers or by using passwords,
but these practices would
compromise confidentiality and
anonymity (unethical).
Siah (2005)
Sample
representation
Respondents not representative of
target population.
Spyridakis et al. (2005)
Multicultural
issues
Research in multicultural contexts
should be sensitive to groups’
ethnic or racial differences to
understand better the meaning of
their consumption behaviour.
69. Cleveland and Laroche (2007);
Fletcher (2006); Jamal (2003);
Nevid and Sta. Maria (1999);
Villegas and Shah (2005);
Emerging
issues in
international
research
Emerging issues in cross-cultural
marketing research, particularly
in developing environments.
Craig and Douglas (2001);
Malhotra and Peterson (2001);
Stening and Zhang (2007)
1146 Journal of Marketing Management, Volume 26
been found to influence behavioural intent. For example,
Triandis et al. (1971, p. 30)
describe the way ‘superordination–subordination’ mimic
behaviours that show status
whereas ‘intimacy–formality’ are symbolic of interpersonal
familiarity. In cultures
such as Japan, status has historically been influenced by
Confucian philosophy
(Chen, 1995) and will therefore impact on the way Japanese
society perceive and
respond to certain questionnaire constructs. Cultural variables
impact on scale
reliability. The meaning associated with the data set in question
becomes invalid
70. because of differences in the way the two cultures associate
meaning.
The emic–etic dilemma asks whether behaviour has to be
understood in the context
of the culture in which it occurs (emic approach) or whether
cultural differences can be
conceived of as variations of a common or universal theme (etic
approach) (Berry, 1990;
Douglas & Craig, 1983; Triandis et al., 1971). The emic–etic
dilemma continues to
challenge the field of psychology by posing the central question
on how generalisable
psychological findings are (Lonner, 1999). It also raises
important questions about the
design of a measurement tool that has universal application
across the cultures under
investigation. A number of papers address this problem,
offering potential solutions.
Osgood suggests one way of dealing with semantic differentials
is to use a universal
construct and then use emic constructs to measure it (Osgood,
1965; Triandis et al.,
1971, p. 7). Tucker (1966) also offers a potential solution to the
emic–etic dilemma.
He proposed that one way to alleviate bias is to adopt a three-
mode factor analysis to
study the intercorrelation patterns of sample respondents. The
emic–etic approach
requires ‘respondent factors’ to be analysed as part of the
research process, as this
provides the researcher with information that functions to
identify the behavioural
orientation of respondents (Triandis et al., 1971).
71. Another solution to the emic–etic apparent contradiction
suggests that the emic
approach could become a rich potential source for
understanding the etic
underpinnings in cross-cultural research, that is, formulating
universalistic theories
and hypotheses that can be tested. This presupposes that each
culture is not so unique
that comparison among cultures is futile or totally meaningless
(Sekaran, 1983, p. 65).
Finally, Berry (1990) suggests combining the two approaches
rather than applying
emic dimensions of one culture to other cultures, which requires
researchers to get
familiar with the relevant differences in each culture, putting
aside their own cultural
biases (Berry, 1990).
Whilst there is no ‘one size fits all’ approach to address the
emic–etic issue, if cross-
national etic factors and culture-specific emic factors are not
given appropriate attention
in the design process, data interpretation is likely to yield
results that lack meaning.
When rigour and relevance are lost, the value of the issues
under examination becomes
questionable. The next section discusses another major concern
in cross-cultural
research – the independence of sample issue.
The independence of sample issue
The cultural studies conducted by scholars such as Hofstede
(1980, 1991) and
Trompenaars and Hampden-Turner (1997) highlight differences
in social values and
72. culture, and provide us with an explanation of why companies
who wish to exploit
market potential at a global level need to consider cross-cultural
influence. However,
independence of samples in relevant cultures represents another
major challenge for
cross-cultural researchers. The problem happens when values
and behaviours become
transfused among cultures to the degree that it is difficult to
differentiate the emic
from the etic (Sekaran, 1983). This problem is particularly
relevant in the present age
Slater and Yani-de-Soriano Researching consumers in
multicultural societies 1147
of globalisation and rapid technological advances, as people
around the globe have
been adopting similar values and behaviours (Nasif, Al -Daeaj,
Ebrahimi, &
Thibodeaux, 1991). The borders among cultures are becoming
blurred due to
cultural diffusion (or cultural convergence), and therefore the
samples taken from
different cultures might not be independent, leading to biased
results (Yeganeh, Su, &
Chrysostome, 2004).
Sir Francis Galton was one of the first scholars to question the
validity of statistical
inference from cross-cultural surveys (Naroll, 1961; Strauss et
al., 1995; Tylor, 1889).
His case was based on the non-independence of observations in
the sample, leading to
73. spurious correlation because of autocorrelation. Galton argued
that if sample
independence could not be achieved or methods of correction
were not applied,
then the research findings were of no value. The non-
independence of observations
in this and other non-experimental questionnaire-based research
became known as
Galton’s problem.
In the context of cross-cultural research relating to heredity,
Galton argued that the
cultural similarity observed in a sample could be related to
cultural diffusion. For
instance, asking two people from the same family the same
question does not provide
the researcher with responses that are statistically independent.
Similarly, cross-
cultural sample clusters may not control for factors such as
borrowing and common
descent. Galton’s problem is that the issue of sample group
independence is essential
for valid cross-cultural research, but is never achievable in
practice. Galton’s problem,
therefore, is often cited as a criticism of empirical comparative
studies of culture: the
results are inherently uninterpretable. It undermines all cross-
cultural research,
forcing researchers to make important methodological
decisions.
Whilst a number of scholars agree with the need to achieve
sample group
independence, there are conflicting views as to whether or not
Galton’s problem is
solvable. A number of studies have put forward solutions,
74. arguing that by using
appropriate method design techniques, it is possible to control
for diffusion
(Denton, 2007; Naroll, 1961; Naroll & D’Andrade, 1963;
Schaffer & Riordan,
2003; Strauss et al., 1975; Vandenber g & Lance, 2000). The
Bimodal Sift method
(Naroll, 1961) and the Interval Sift method (Naroll &
D’Andrade, 1963) propose two
solutions to the problem. These methods allow the researcher to
treat the sample as
independent ‘once the validity of the sift’ has been confirmed.
The cluster method
(Naroll & D’Andrade, 1961, p. 1054) and the matched pair
method (Naroll &
D’Andrade, 1961, p. 1054), whilst ‘statistically less flexible’
than the sifting
methods, offer an alternative. They constitute a fairly ‘rigorous
way’ (Naroll &
D’Andrade, 1961, p. 1054) of measuring diffusion because they
enable the
researcher to distinguish between both the functional and
historical associations of
the culture and yield information about the importance of each
association.
Galton’s original studies were important because they
introduced the importance of
sample independence in cross-cultural surveys. However, future
cross-cultural
research needs to extend the issues raised by Galton in the
context of a global
marketing environment. A key challenge for the future is how
diffusion might be
measured in multicultural societies. The issues that emerge from
acculturation and the
75. globalisation of consumer behaviour mean that culture may no
longer be an
appropriate unit of analysis for ‘cross-cultural’ surveys. The
next section discusses
the main methodological issues confronting cross-cultural
marketing research and
offers suggestions for addressing them.
1148 Journal of Marketing Management, Volume 26
The issue of comparability of data and measurement
equivalence
While the paper by Malhotra et al. (1996) offers an excellent
account of the
methodological issues associated with cross-cultural marketing,
the approach is
somewhat biased towards survey techniques that use numbers
and can therefore
demonstrate statistical relevance. A key observation of the
Malhotra et al. paper is
the North American emphasis that is associated with the
manuscript’s research
orientation. In fairness to the authors, the emphasis on North
America was probably
more a reflection of when the paper was written than of the
author’s intention to
restrict sample randomness to a certain geographical region.
Nevertheless, the point is
worthy of attention now that current marketing theory and
business decisions have
become more global in their market orientation.
The issue of comparability is critical to cross-cultural research.
76. Comparability has
been defined by Douglas and Craig (1983, pp. 131–132) as ‘data
that have, as far as
possible, the same meaning, interpretatio n, and the same level
of accuracy, precision of
measurement and reliability in all countries and cultures’. They
add that
‘comparability of data is important irrespective of whether
research is conducted in
a single country, for it is important to bear in mind that research
relating to a similar
problem may subsequently be conducted in another country’.
Bhalla and Lin (1987)
concurred with these authors in that the need for comparability
is a key issue
confronting marketing researchers, regardless of whether the
research is conducted
in one country or a number of countries simultaneously.
The major methodological challenges of cross-cultural research
have been
underscored by Sekaran (1983) to ensure functional
equivalence, problems of
instrumentation, data-collection methods, sampling design
issues, and data analysis.
Likewise, Parameswaran and Yaprak (1987) acknowledged the
need to establish
construct, functional, conceptual, instrument, translation, and
sampling equivalence
of research measures in cross-cultural research before inferring
statistical and practical
conclusions. Although ‘comparability’ and ‘equivalence’ are
often used interchangeably,
comparability is the more generic term; equivalence refers more
precisely to
measurement. Lack of data equivalence could lead to wrong
77. conclusions. Observed
differences in measures might be attributed to true differences
in the latent variables,
although they are solely caused by differential response
behaviour; or true differences
might be masked by differential response behaviour and thus
remain undetected,
causing an uncontrollable increase in type one and type two
errors (Salzberger &
Sinkovics, 2006, p. 392).
Functional equivalence
Functional equivalence refers to the fact that similar activities
should have similar
functions in different societies if their parameters are to be
compared. According to
Sekaran (1983, p. 62), ‘valid cross-cultural behaviour
comparisons can be made only
when the behaviour in question has developed in the different
cultures in response to
similar problems shared by the different social or cultural
groups’. For instance,
bicycles are used in the Netherlands as a means of
transportation, whereas in
Venezuela they are used primarily for recreation or sport. In
this case, functional
equivalence cannot be achieved because the goals of behaviour
towards the same
product are different across these two cultures.
Slater and Yani-de-Soriano Researching consumers in
multicultural societies 1149
78. Conceptual equivalence
Conceptual equivalence refers to the fact that subjects have an
equal understanding or
interpretation of the meaning of behaviour, product, or
marketing stimuli. As an
example, price markdowns are regular events in the United
States. However, they
might be seen with suspicion by consumers in developing
countries. Similarly, Soriano
and Foxall (2002) found that saving-up behaviour in Venezuela
and England was
conceptually non-equivalent. While saving up is seen as a
positive behaviour in
England, in Venezuela, due to high levels of inflation, it is seen
negatively or as a
not-desired behaviour.
Due to these types of problems, Berry, Poortinga, Segall, &
Dasen (1992, p. 237)
have suggested that ‘a close scrutiny of each stimulus is
necessary to identify positive
peculiarities in meaning or other reasons that a stimulus might
be inappropriate in a
particular culture and should not be used’. Furthermore,
problems of conceptual
equivalence are not limited to countries with different
languages, but they also affect
countries who share the same language. Therefore, this paper
argues against the
practice of directly transferring an instrument developed in the
UK to the United
States or from Spain to Latin America, for example. Instead, it
is vital to make
necessary adjustments and pretests, despite the common
language base. Clinton and
79. Calantone (1996) found conceptual problems between English-
speaking countries.
Equally, it is important to note that cross-cultural research
issues are not limited to
research in different countries but can arise in research within
the same country.
Advertising studies on African American and white Americans
highlighted major
differences in the way these two groups perceived and reacted
to advertising
campaigns (Bush, Smith, & Martin, 1999). A study by Soley and
Reid (1983)
showed similar findings. African Americans were more satisfied
with the advertising
content that was presented in TV commercials and magazines
than white Americans.
Given that the two groups were not separated by languages or
significant distances, we
can conclude that the differences observed were more distinctly
cultural, rather than
linguistic or geographical. Given that countries around the
world are becoming
increasingly multicultural, this problem receives special
attention in the emerging
methodological issues section of this paper.
Translation equivalence
A way to operationalise conceptual equivalence is to test for
translation equivalence.
This can be achieved using translation and back translation of
instruments (Brislin
et al., 1973). Berry (1980, p. 10) explained translation
equivalence as ‘a technique
which involves an initial translation to a target language by one
bilingual person, and a
80. back translation to the original language by another;
discrepancies will often indicate
the presence of conceptual non-equivalence’. Sekaran (1983)
has added that the
equivalence of source and target version of the instrument can
be ensured with good
back translations conducted by people who should be not only
proficient with the
different languages involved but are also familiar with the
cultures in question and
with the usage of the concepts and their meanings in such
cultures. The problem of
translation equivalence (linguistic) is a difficult one as some
linguistic concepts do not
translate directly into other tongues and important nuances of
meanings can be put at
risk (Barnard, 1982). Erkut, Alarcon, Garcia-Coll, Tropp, and
Vazquez-Garcia (1999)
suggest a concept-driven rather than translation-driven approach
to creating bilingual
measures, which requires a bilingual/bicultural research team
with indigenous
1150 Journal of Marketing Management, Volume 26
researchers from the cultures being studied, in order to reduce
researchers’ cultural
bias and achieve both conceptual and linguistic equivalence.
Metric equivalence
Metric equivalence exists when psychometric properties of two
or more sets of data
from two or more cultural groups exhibit essentially the same
81. coherence of structure.
There are two requirements for this type of equivalence: first,
that statistical
relationships remain fairly constant among independent and
dependent variables;
second, that statistical relationships among dependent variables
should be patterned
similarly in two or more cultural groups before comparisons are
allowed. Similarity in
correlation matrices or common factor structures can serve to
demonstrate this type of
equivalence (Berry, 1980, p. 10). Metric equivalence, however,
is no guarantee of
invariance. Byrne and Campbell (1999) have warned against the
presumption of
equivalent measurement and theoretical structure in cross-
cultural comparisons
because although the factorial structure of an instrument may
replicate across
cultures, this is no guarantee that the item measurements and
theoretical structure
are invariant across groups. They showed that, in fact, item
score data can vary across
cultures, despite measurements from an instrument for which
the factorial structure
has been equivalently specified in each group. Therefore,
researchers and practitioners
should also question the philosophical and conceptual
appropriateness of an
instrument that has been developed in a culture that differs from
the one in which it
is to be used.
Fontaine, Poortinga, Delbeke, and Schwartz (2008, p. 363)
emphasise that researchers
need to interpret equivalent measurements appropriately,
82. arguing that ‘such an
assessment is more than a technical, psychometric exercise’.
Based on their findings
from evaluating the structural equivalence of the values domain
across 38 countries,
they demonstrated that patterns of non-equivalence are not
sampling fluctuations, but
can be attributed to meaningful variations, which generate
insights into the
relationship between the structure of the values domain and
other aspects of society
and culture. One of their findings revealed that the size of the
structural deviations
related more strongly to societal development in the teacher
than in the student
samples, which means that it is critical to include non-student
samples in cross-
cultural research studies.
Complementary approaches to improve comparability
Pretests
In addition to the methods already discussed, other approaches
should be applied to
improve comparability of data, including pretests, sample
equivalence, sampling
methods, and data-collection procedures. Brislin et al. (1973)
state that multiple
techniques should be used in all cross-cultural research, since
the back translation as
a single method is no panacea; all materials should be pretested
with respondents
similar to those of the proposed main sample, since there will
always be items that
simply do not work well in actual use (Brislin, 1986, p. 161).
83. Douglas and Craig (2007)
found that, in marketing, back translation is still, by far, the
main method used to
check translation accuracy. Because the translation process is
complex, a single back
translation does not ensure a valid and reliable instrument in the
target language, since
the objective is not that the words translate literally but that
they have the same
Slater and Yani-de-Soriano Researching consumers in
multicultural societies 1151
meaning across cultures. Therefore, a team approach is required
to minimise bias.
Unfortunately, although the use of pretests is particularly useful
in cross-cultural
research, it is not widely being employed, with less than one-
third of studies
conducted between 1993 and 2005 shown to have used it
(Douglas & Craig, 2007).
Sample equivalence
It is important to achieve sample equivalence, given that
homogeneity of the sample
reduces alternative explanations of the results (Lonner & Berry,
1986) and therefore
the sample should be chosen to maximise equivalence rather
than, for example,
representativity. Sample equivalence can be satisfied if
respondents in each study
share similar demographic and socio-economic characteristics.
However, Sin et al.
84. (1999) warns that unless the researcher knows well the cultures
under study,
sometimes subjects coming from the same sampling frame may
not guarantee
sampling equivalence, such as in the case of undergraduate
students. In addition,
sample equivalence facilitates neutral response styles and
validates cross-national
comparisons between countries. One problem with this type of
study is that it
ignores the potential cumulative effect of efforts when targeting
behaviour and leads
to ethnocentric evaluations (Triandis et al., 1971). It can be
argued that in the Hong
Kong American study, dependent variables of interest might
well be factored out when
using differential sampling systems. For these reasons whilst
scholars need to eliminate
non-equivalence variables, methodological design procedures
need to ensure that
when controlling for non-equivalence researchers do not
neutralise the key variables
of interest to users of the research given an important aim of
market research is to
probe for market differences.
Sampling methods and data-collection procedures
Sampling methods and procedures for data collection are also
essential aspects in
improving comparability of data. In some developed countries,
particularly in the
United States where consumer research is a major function of
business, the collection
of data tends to be highly professional. However, this is not true
in developing
85. countries, where, in addition to the lack of a research tradition,
cultural factors play
an important role in the collection of information. This includes
attitudes towards
security and privacy (e.g. people are wary of strangers and
might not feel comfortable
providing information about themselves) and issues of
accessibility (e.g. boundary
walls).
Quota sampling in marketing research has been used widely
both in developed and
developing countries (Malhotra & Peterson, 2001) and can be
combined with other
methods. Webster (1966) studied five Western European
countries, employing
probability sampling in four countries from readily available
lists, and quota
sampling in one country where random sampling needed the
purchase of a special
list at a very high price. She argued quota sampling could
produce acceptably correct
and meaningful results comparable with those obtained by
probability sampling in
other countries, and that sameness of method of collection by no
means assures
comparable data for analysis. Webster (1966) stated that data
are comparable if they
have the same degree of reliability and one has used the most
efficient rather than the
same method of data collection in each country.
Dunn (1974) extended Webster’s (1966) arguments by pointing
out that if the local
alternative methods are all equally good, one should use the
same methods in one or
86. more countries, but if they may lead to some bias, one might
deliberately choose
1152 Journal of Marketing Management, Volume 26
different methods to check whether and how potential bias
operates. A crucial point
has been made by Osgood, May, and Miron (1975) in that the
purpose of the research
determines the sampling strategy, that is, whether the
investigator wants
representativeness within each country or equivalence across
countries. They warn
that maximising representativeness within a country usually
means minimising
equivalence between countries and vice versa. In a cross-
cultural study, it is possible
to use one method such as quota sampling in a developing
country, and another
method such as a random walk in a developed country. An
advantage is achieved
when comparing results between the two studies, as ‘the
differences in sampling
methods can also be utilised to provide a check on the
reliability of results and the
potential bias inherent in different methods’ (Douglas & Craig,
1983, p. 219). If
different methods in different countries produce similar results,
it means that the
measurement instrument utilised is reliable. Hence, it may be
not only possible but
also desirable to use different sampling techniques in different
countries to achieve
sample equivalence and representativeness (Malhotra &
87. Peterson, 2001). This is
particularly relevant today, as marketing researchers must
develop the capability to
conduct research that spans diverse research environments to
benefit from growth
opportunities outside developed countries (Craig & Douglas,
2001).
Emerging trends and methodological issues
Internet-based research
The accelerated growth of the Internet has opened many new
opportunities for
academic and marketing research reaching populations all
around the world.
Research has shown that web surveys and experiments are
faster, easier, cheaper,
and more flexible compared to traditional methods
(McCullough, 1998; Pitkow &
Recker, 1995). However, there are methodological and ethical
issues that must be
addressed if the medium is to provide meaningful information.
The main issues
include sampling and generalisability of results, participant
behaviour (such as drop-
out rates for panels, incomprehension, response set bias, low
response rate, and subject
fraud), data integrity (e.g. caused by technical problems), and
ethics (keeping privacy,
anonymity, confidentiality, and avoiding stress related to
sensitive questions).
The Internet, as a research device, is growing, and the quality of
the research
depends on the perceived credibility and trust both participants
88. and researchers have
in this medium (Montgomery & Richie, 2002; Siah, 2005;
Spyridakis, Wei, Barrick,
Cuddihy, & Maust, 2005).
There is a lack of consensus on the pre-eminence of the Internet
as a research
medium in the future. Wilson and Laskey (2003) found that
most of the companies in
the UK had used online studies in the previous year, but only as
an additional
supporting methodology rather than as a substitute to traditional
approaches in
certain special types of research (e.g., website evaluation). They
also found that
research companies had concerns about the weaknesses already
mentioned, in
particular, the issue of representativeness, given the nature of
Internet sample frames.
Recent research in social marketing studying UK adolescents
argues that Internet-
based research has the potential to generate data that are
comparable to those
generated by conventional research methods and with improved
efficiency in terms
of time, cost, quality, and quantity of responses (De Meyrick,
2007). Tingling, Parent,
and Wade (2003) believe that the Internet survey is an under -
exploited resource,
Slater and Yani-de-Soriano Researching consumers in
multicultural societies 1153
89. arguing that it can be used to reduce potential bias by randomly
assigning choices along
the screen and/or by generating more complex randomised
display patterns than
allowed in paper surveys. The Internet offers a whole new way
of engaging with
consumers. It is therefore not surprising that in twenty-first
century marketing, well-
known brands such as British Airways, Philips, and O2 are
launching their own
community websites to engage consumers and gauge their
responses to new
products, designs, and advertisements (Benady, 2008).
However, website content,
design, and structure needs to take into account cultural
sensitivity to improve its
communication effectiveness both at global and multicultural
levels (Fletcher, 2006).
Diverse marketing-research contexts
Technological advances improve the basic infrastructure of
developing nations
opening up new marketing opportunities for global companies.
Emerging economies
such as China and India offer immense market potential.
Marketing researchers need a
clear understanding of these potential markets in order to be
able to conduct valid
research studies. They should also be careful not to interpret
and generalise results
based on their own cultural experience in developed countries.
Chinese, for example,
think in a holistic way, whilst Westerners think in an analytical
way (Needham, 1978;
Nisbet, 2003) posing a challenge to data interpretation (Stening
90. & Zhang, 2007).
Consumer behaviour needs to be studied and interpreted by
researchers from the
culture being studied, who deeply understand the context of
consumption.
The global and multicultural perspectives of consumer
behaviour research
The twenty-first century is seeing a growing trend towards more
countries becoming
increasingly ethnically diverse. For example, citizens in the
United States and the UK
are multicultural in their ethnicity. Although there is a great
emphasis on international
marketing, little research has been conducted to understand
diversity within a specific
country (Nevid & Sta. Maria, 1999). Cleveland and Laroche
(2007) argue that it is no
longer appropriate to use countries as the cultural unit of
analysis or market
segmentation, since most of the world’s countries are already
highly multicultural.
Globalisation thus seems to reduce the homogeneity of
consumer behaviour within
countries, while increasing commonalities across the globe.
Findings from recent multicultural research have shown
interesting results. Two
Hispanic groups (Cuban American and Mexican American),
which traditionally have
been treated as one group, have shown differences affecting
advertising effectiveness
(Villegas & Shah, 2005). Jamal (2003) found groups of different
ethnic backgrounds
in the UK engage in culture swapping to taste different cultures.