3

CHAPTER

Research Designs:
Exploratory and Descriptive

Learning Objectives
By the end of the chapter, you should be able to:
answers to the research questions
Jdentify the framework or design you intend to use to arrive at
framed by you.
formulating the research design.
2. Appreciate the numerous options available to vou in
3 Understand the nature of exploratory and two-tiered research designs.
4 Understand the techniques and stages in descriptive studies.
5
Understand and interpret cross-sectional and longitudinal designs.

Buzy Bee (BB) home solution office at the dismal

As Anamika Rathore looked out from the 15th floor window of her Circus, it seemed that a similar fog had
January fog which was masking the bustling and cheerful view of Connaught
enveloped her normally decisive mind.
imported cabinets of all shapes and sizes,
The company had been set up two years back in this prime location. They promise and the pundits
made from superior quality buffed steel and aluminium. The product category showed great rise further by 11 per cent in
it to
had predicted an unparalleled growth of 28 per cent in the coming year and expected
Godrej and even regional and
the subsequent year. But somehow BB was not in the radar of the potential buyer. Kaffe,
unbranded manufacturers enjoyed better sales than BB.
Anamika had suggested that they study the buying behaviour of the residents of builder apartments and society flats
as they could be potential customers. The next step would be to identify the reasons for the lost opportunity. Anant
Chacko, the CEO, took her suggestion seriously and agreed to sponsor the survey. However, he asked her to present a
blueprint of the proposed investigation.
Ablueprint for ashort survey? Is that not making asimple thing so complicated? After all, it is not abuilding that
she intends to construct that he was asking for the architectural design. That's what happens with these aggresive
young people who have afancy, glitzy MBA from abroad. Then she suddenly remembered Nilesh, who was with a lo
cal market research fim, and immediately called him up. "Hi Nilesh, Anamika here, I need your help. Can you help me
design asurvey?" Hi Ani, sure. What kind of a design would you be looking at?" and he rattled off a set of names and
assumptions. Anamika was flummoxed, what had she let herself in for?

The CEO was right in the stipulation that he had made. In fact, most researches
lose out because either the research design was not conceptualized properly, or the
design formulated was weak. Daft (1995), while reviewing the academic articles for
the Academy of ManagementJournal and the AdministratiueScience Quarterly, states
INST:TUTE0F MANAGE*:#
2651
 Research Methodology
52

that 20 Der cent of the reasons for rejection was inadequate study design, Grunow:
(1995), urther coroborates andstates that this weak area was discovered in both e
ublished as wellas the unpublished articles that he analysed. For asingle research
Droblem, different design options might exist, however, they have to be carefully
selected based upon the deciding criteria and requirement of the study. This point
will be further elaborated when the criteria of a well-structured research design are
discussed in the chapter.
Thus, given certain preconditions, the researcher has multiple approaches to
study the same problem (Hitt et al., 1998). In fact, for the same research question
both qualitative and quantitative approach could be taken (Bartunek et al., 1993)
for example, toestablish the human development status of a country, we can look
at the quality of life (qualitative) that people enjoy or look at certain quantifiable
parameters like longevity, literacy and purchasing power parity (quantitative).
This is an approach that became acceptable only in the later half of the 20h
century, as the earlier school of thought was more based upon the objective nature
of theory building-the positivist paradigm. This only accepted designs which
called for an empirical observation and were followed by a certain level of statistical
analysis (Ackroyd, 1996). The constructivists, on the other hand, argue for more
divergent and behaviour specific techniques that are not a spillover from the natural
sciences, and thus, follow a more qualitative approach (Jorgensen, 1989; Atkinson
and Hammersley, 1994). However, what needs to be considered by the researcher is
what best suits and matches the research objectives;and only after that, he should
take a position and proceed with the choice of the study.

THE NATURE OF RESEARCH DESIGNS

LEARNING OBJECTIVE1 Once you have established the what of the study, i.e., the research problem, the
identifly the framework or next step is the how of the study, which specifies the method of achieving the stated
design you intend to use research objectives in the best possible manner.
to arive at answers to As stated earlier, different paradigms will guide the _election of the
the research questions gamut of
framed by you.
techniques available. These differences in approach have led to varying definitions
of what constitutes a research design.
Green et al. (2008) defines research designs as 'the
and procedures for acquiring the information needed. It isspecification
the
of methods
overall
pattern or framework of the project that stipulates what information is to be operational
from which sources by what procedures. If it is a good collected
design, it will insure that the
information obtained is relevant to the research questions and that it was
by objective and economical procedures! collected
Aresearh design is based Thyer (1993) states that, 'A
on aframework and provides plan for how aresearch study is traditional
to be
research design is a blueprint or detailed
adirection to the
being conducted ininvestigation
the most
can be measured, selecting a sample of
abasis for testing
completed-operationalizing variables so they
interest to study, collecting data to be used as
effiient manner. hypotheses, and
of the design is thus to provide a analysing the results. The essential requirement
the most efficient manner. Sellitz etframework and direction to the
al. (1962) states that 'A investigation in
research
arrangement conditions for collection and analysis of data ina design is the
of
to combine relevance to the research purpose manner that aims
with economy in
and holistic definition procedure'
"One of the most
Kerlinger (1995).He referscomprehensive
to a research design as, .. has been given by
of investigation so a plan, structure and
conceived as to obtain answers to research strategy
The plan is the complete scheme programme of the research. It
or questions or problems.
includes an outline
 Research Designs: Exploratory and Descriptive 53

of what the investigator will do from writing the hypotheses and their
operational
implications to the final analysis of data.'
the Thus, the formulated design must ensure three basic tenets:
Researchdesignis (a) Convert the research question and the stated
been
frameworkthathas
to operational variables that can be measured.
assumptions/hypotheses into
reatedtoseekanswers
On the
researchquestions. method
(b) Specify the process that would be followed to complete the above task, as
other hand, research the efficiently and economically as possible.
collect (C) Specify the control mechanism(s)' that would be used to ensure that the
s thetechnique to
informationrequired. effect of other variables that could impact the outcome of the study have
been controlled.
The important consideration is that none of these assumptions can be
foregone; all of them must be addressed succinctly and adequately in the design
for it to be able to lead on to the methods to be used for collecting the
problem
specific information. Thus, it follows the problem definition stage and precedes the
data collection stage. However, this is not an irreversible step. Sometimes when the
researcher is operationally defining the variables for study, it might emerge that the
research question needs to be restructured and consecutively the approach for data
collection also might oscillate from the quantitative to the qualitative or vice versa.
At this juncture, one needs to understand the distinction between research
design and research method. While the design is the specific framework that has
been created toseek answers to the research question, the research method is the
technique to collect the information required to answer the research problem, given
the created framework.
Thus, research designs have a critical and directive role to play in the research
process. The execution details of the research question to be investigated are referred
to as the research design.

FORMULATIONOF THE RESEARCH DESIGN:PRoCESS

LEARNING OBJECTIVE 2 Once the researcher has identified the research scope and objectives, he has
Appreciate the
also established his/her epistemological position. This could be positivistic--in
numerouS options which case the method of enquiry would necessarily be scientific and empirical.
available to you in Subsequently, this would require a statistical method of analysis (Ackroyd, 1996).
formulating the research The constructivists on the other hand argue for methods that are richer and more
design. HEL applicable to the social sciences, unlike the more pedantic experimnental approach.
Qualitative is a more definitive choice here than the quantitative (Atkinson and
Hammersley, 1994). Yet another approach is the principle of triangulation (Jick
1979), which advocates the simultaneous or a sequential use of the qualitative and
quantitative methodsof investigation. The proponents state that when the findings
from diverse methods are collated, then the results are richer, more wholistic anc
this, in turn, improves the sanctity of the analysis.
The principle of The formulated research questions are then, through a comprehensive
triangulation
advocates the simultaneouS or
asequential use of theoretical review, put into a practical perspective. The conceptual design thu
and quantitative qualitative developed requires and entails specifications of the variables under study as wel
methods of as approach to the analysis. This might in turn lead to a refining or rephrasing c
investigation. the defined research questions. Thus, the formulation of the research design is not
stagnant stage in the research process; rather it is an ongoing backward and forwar
integrated process by itself.
" An illustration: Let us take the example of the organic food study. The formulate
research problem was:
 54 Research Methodology

To investigate the consumer decision-making process for organic food products

according to the basket size.
and to segment the market that or
review of the literature, it was found sellec
On conducting an extensive choice; rather it could be the
consumption is not always a self-driven research design was formulated to O
influence the product choice. Thus, a
might However, once the design is selected an
the organic consumer's decision stages.
next step required is that the constris
proposed sampling plan is developed, theoperationalized. On defining the organi
and the variables to be studied must be psychographics of the individual-th
consumner, we realized the significance of the Thus, to get a wholistie
attitude, interest and opinion--which were extremely critical. consumer, as wel
view,one needs to look at the psychographic profile of the existing
mindset. This led to a revision of tho
as of the potential consumer with a similar
research problem:
products and
To investigate the consumer decision-making process for organic food
profle.
tosegment the market-existing and potential-according to their psychographic
CLASSIFICATION OF RESEARCH DESIGNS
LEARNING OBJECcTIVE 3
The researcher has a number of designs available to him for investigating the
Understand the nature
research objectives. There are various typologies that can be adopted for classifying
of exploratory and two them. The classification that is universally followed and is simple to comprehend is
tiered research designs. the one based upon the objective or the purpose of the study. Asimple classification
that is based upon the research needs ranging from simple and loosely structured
to the specific and more formally structured is given in Figure 3,1. This depiction
shows the two types of researchesexploratory and conclusive as separate design
options, with subcategories in each.
The demarcationbetween the designs in practice is not this compartmentalized.
Thus, a more appropriate approach would be to view the designs on a continuum
as in Figure 3.2. Hence, in case the research objective is diffused and reguires a
fine-tuning and refinement, one uses the exploratory design, this might lead to the
slightly more concrete descriptive design--here one describes all the aspectsof the
constructs and concepts under study. This leads to a more structured and controlled
causal research design.
The research design
dlassification that is universally In thischapter, exploratory and descriptive research designs are discussed in
detail. The causal design requires to be understood for its mathematical presumptions
followed and simple to and that would be dealt with in the next chapter.
comprehend is the one based
upon the objective or purpose
of the study. Exploratory Research Design
Exploratory designs, as stated earlier, are the simplest and most loosely structured
designs. As the name suggests, the basic objective of the study is to explore and
obtain clarity about the problem situation. It is flexible in its
involves a qualitative investigation. The sample size is notapproach and it mostily
and at times it might only involve unstructured strictly representative
experts. The essential purpose of the study is to: interviews with a couple of subject
Define and conceptualize the research problem to be
Explore and evaluate the diverse and multiple investigated
Assist in the development and formulation of the research opportunities
research
Operationalize and define the variables and constructs under hypotheses
Identify the possible nature of relationships that might exist study between the
variables under study
Explore the external factors and variables that might
impact the research
 Research Designs. Exploratory and Descriptive 55

FIGURE 3.1
Classificationof Research Design
résearchdestons

Exploratory
Research Design
Conclusive
Research Design

D Descriptive Causal
TResearch Research

Cross-sectional Longitudinal
Design siait Design

Single Cross Multiple Cross

sectional Design sectional Design

FIGURE 3.2
Research designs
a continuous process

Statistical
Analysis

Degree of Structure
 Research Methodology
56

For example, a university professor might decide to do an exploratory analysis

Exploratory research design utilized by the marketer
Is flexible in ity approach of the new channels of distribution that are being
this, a structured and def
and nvolves a qualitative promote and sell products and services. To accomplish tounderstand the
investigation in most cases. methodology might not be essential as the basic objective is findings are of intero
Itis the simplest and most paradigmsfor inclusion in the course curriculum. In case the or an annl:
academic, basic research
loosely structured design. the same may lead to a more structured,
different methods.
problem where one may want to establish the efficacy of
However, no matter what the scientific orientation and the research obiective
might be, the researcher can make use ofa wide variety of established methods and
techniques for conducting an exploratory research, like secondary data sources
unstructured or structured observations, expert interviews and focus oro
discussions with the concerned respondent group. Most of these techniques are
dealt with in detail in the subsequent chapters; however, we will discuss thenm in
brief in the context of their usage in exploratory research.

Secondary Resource Analysis

Secondary sources of data, as the name suggests, are data in terms of the details of
Secondary sources of previously collected findings in facts and figureswhich have been authenticated
data contain the details of and published. An added advantage of secondary data is that it can be represented in
previously collected findings a relatively easier way and is less expensive. Secondary data is a fast and inexpensive
and can be represented way of collecting information. The past details can sometimes point out to the
in a relatively easier and researcher that his proposed research is redundant and has already been established
inexpensive way. earlier. Secondly, the researcher might find that a small but significant aspect of the
construct or the environment has not been addressed and might require a full-ledged
research to explain some unpredictable results. For example, a marketer might have
extensively studied the potential of the different channels of communication for
promoting a home maintenance service' in Greater Mumbai. However, there is no
impact of any mix that he has tested. An anthropologist research associate, on going
through the findings, postulated the need for studying the potential of WOM (word
of mouth) in a close knit and predominantly Parsi colony where this might be the
most effective culture-dependent technique that would work. Thus, such insights
might provide leads for carrying out an experimental and conclusive research
subsequently.
Another valuable secondary resource isthe compiled and readily available
bases of the entire industry, business or construct. These might be available on data
and public domainsor through a structured acquisition process and cost. Thesefree are
both government and non-government publications and would have varying
of authentication and sanpling base. Based on the research levels
constraints
of accuracy required, the researcher might decide to make use of
and the level
them.
Comprehensive case method
Another secondary source which can serve as a
Comprehensive case technique for conducting an
method is intricately designed exploratory research is the case study method. It merits separate mention as it 1
and reveals a complete intricately designed and reveals a comprehensive and complete
presentation of facts as they facts, as they occur, in a single entity. This in-depth study is presentation o
occur in a single entity. It is of analysis. This unit could be an focused on a single unt
focused on asingle unit of
individual
or a complete country analysis might also be theemployee or a customer; an organizato
analysis. case of interest. They are by thet
nature, generally, post-hoc studies and report those
Occurred earlier. The scenario is reproduced based upon incidences which might have
the
and a primary recounting by those
involved in the occurrence.secondary intormauo
Thus, there might be
 Research Designs: Exploratory and Descriptive 57

an element of bias as the data, in most cases, become a judgemental analysis

thana simple recounting of events. rather
F'or exanmple, BCA Corporation wants to implement a
system in the organization and is performance appraisal
debating between the merits of a traditional
appraisal system and a 360" appraisal system. For a historical
two techniques, the HR director makes use of the understanding of the
theoretical works done on the
Constructs. However, the roll-0ut plans and repercussions and the
were not very clear. This could be better understood when they management issue
studies on Allied Association which had implemented traditional studied in-depth case
and Surakhsha International-360" systems. Thus, the two appraisal formats,
exploratory
carried out were sufficient to arrive at a decision in terms of what wouldresearcnes
for the organization. work best

Expert opinion survey

There might be a situation at times when the topic ofa research is such that there is
Expert opinion survey is no previous information available on it. Thus, in these cases, it is
Conducted when no previous advisable to seek
information or data is available
help irom the experts who might be able to provide some valuable insights based
upon their experience in the field or with the concept. This approach of collecting
on atopic ofresearch. It is particulars from significant and erudite people is referred to as the expert opinion
formal and structured in
survey. This methodology might be formal and structured and might be useful when
general.
being authenticated or supported by a secondary/primary research or it might be
fluid and unstructured and might require an in-depth interviewing of the expert.
For example, the evaluation of the merit of marketing organic food products in the
domestic Indian market cannot be done with the help of secondary data as no such
structured data sources exist. In this case the following can be contacted:
Doctors and dieticians as experts would be able to provide information
about the products and the level to which they would advocate organic
food products as a healthier alternative.
" Chefs whoare experimental and innovative and might look at providing
a better value to the clients. However, this would require evaluating their
level of awareness and perspective on the viability of providing organically
prepared dishes.
" Pragmatic retailers who are looking at new ways of generating footfalls
and conversions by offering contemporary and futuristic products. Again,
awareness about the product,past experience with selling healthier lifestyle
products would need to be probed to gauge their positive or negative
reactions to the new marketing initiatives.

It is advisable to quiz different

These could be useful in measuring the viability of the proposed plan.
expert sources as no expert, no
Discussions with knowledgeable people may reveal some information regarding
matter how learned or erudite., who might be considered as potential consumers. Secondly, the question whether
can be solely relied upon to a healthy proposition or a lifestyle proposition would work better to capture the
targeted consumers needs to be examined.
arrive at any condusions.
Thus, this method can play a directionalrole in shaping the research study.
Howeve, anote ofcautionisalso necessary asbyitsverynature,itisalooselystructured
and skewed method, thus supporting it with some secondary data or subsequendy
validating the presumptions through a primary research is recommended. Another
aspect to be kept in mind is that no expert, no matter how vast and significant his
experience is,can be solely relied upon to arrive at any conclusions, as in the example
stated above. It is also advisable to quiz different expert sources. Notwithstanding
these constraints, this technique is of great value to any researcher, no matter what
58 Research Methodology

more varied the perspective, more Gestaltian is the

interest is. The meaningful contribution to the
his/her area of
whichwill result in a field of
research approach,
study.
Focus group discussions
interviewing is to carry out discussions with
approach to under study. This
Another alternative associated with the problem
in all branches of
technique,
Focus group discussions significant individuals
rooted in sociology, is actively used management and behaviouralhere also it
technique is oriqinally rooted though originally has a special significance in
motivational
it
sciences. However, advocated consumer and research
insociology and is most and used for
staunchly advocated and used is most
staunchly
group, there is a
carefully selected small set of individuals
focus study. It is called a
respondent population topic for the duration focus
for consumer and motivational studies. In atypical under
of the larger of on
research studies, representative discuss the concerned
members individuale
group as the selected Usually the group comprises six to ten
two hours. throw enough
minutes to, sometimes, because less than six would not be able to
The number thus stated is there might emerge a one-sided
or a skewed
discussion and
perspectives for the
On the other hand, more than ten might lead to more
discussion on the topic. and that would be unwieldy to
manage
than any fruitful discussion observer
confusion rather
out in neutral settings bya trained
Generally, these discussions are carried moderator, in most cases, does not participate
also referred to as the moderator. The non-structured and
in the discussion. His primne objectiveprocess is to manage a relatively
and then maneuvers it to steer it ony
informal discussion. He initiates the
there is more than one observer to
to the desired information needs. Sometimes, discussion. The conduction and
record the verbal and non-verbal content of the understanding
behavioural
recording of the dialogue requiresconsiderable skill and product study, the
and the management of group dynamics. In the organic food consumers/buyers of
focus group discussions were carried out with the typical
health
grocery products. The objective was to establish the level of awareness about
hazards, environmental concerns and awareness of organic food products. A series
of such focus group discussions carried out across four metros-Delhi, Mumbai,
Bengaluru and Hyderabad--revealed that even though the new age consumer was
concerned about health, the awareness about organic products was extremely low
to non-existent.

Two-tiered Research Design

Once an exploratory study using aloosely structured exploratory design is over, the
The two-tiered research researcher would have agreater clarity and direction, leading subsequently to amore
design involves the formu structured research that he might undertake. Thus, he would manage to achieve the
lation of the research question
and the design framework.
following:
Acomprehensive and focused research question,which will clearly indicate
the orientation the study intends to take
Finding out through various sources as listed above that the need for a
Conclusive research study is not there and the decision-maker can make
use of the exploratory results to assist in the
Developing both the
decision making
general and the specific
certain trends or outcomeshypotheses presumptions
of the likelihood of or
Developing clarity on the framework and methodology best
achieve the suited to
This is/might beformulated research objectives
the first rung of a two-tiered
step is to formulate the research research design where the hrst
question and the second-tier is more formal and
 Research Dessgns Exploratory and Descriptive 59

1 What is the basic nature of

CONCEPT 2 Define exploratory researchresearch designs?
CHECK 3.
Ilustrate the importance of design.
4
What is meant by oomprehensive case method.
two-tiered research design?
structured and refers to the design framework defined
instances, the researchers avoid the first rung and move earlier in the chapter. inmost
the additional cOst and time involved. on to the second, due to
However, it is advocated strongly that the
exploratory
and redundant
stage can be extremely significant in
reducing the risks of ambiguous
research objectives.

LEARNING OBJECTIVE 4
Descriptive Research Designs
The second set of research designs,
tormal in nature. These are termed discussed
in the chapter, is more
as the descriptive designs. As thestructured and
Understand the
techniques and stages in the objective of these studies is to provide a name implies,
descriptive studies.
of the phenomena under study.The comprehensive and detailed explanation
intended
Give a detailed sketch or profile of the objective might be to:
respondent population being studied.
This might require a structured primary collation of the information to
understand the concerned population. For example, a marketer to design
nis advertising and sales promotion campaign for high-end watches, would
require a holistic profile of the population which buys high-end luxury
Descriptive designs provide products. Thus a descriptive study, which generates data on the who, what,
acomprehensive and detailed
when, where, why and how of luxury accessory brand purchase would be
explanation of the phenomena the design necessary to fulfil the research objectives.
under study. However, it lacks " There might be a temporal component to this design, that is, the description
the precision and accuracy of might be in a stagnant time period or be stretched across collecting the
experimental designs. relevant information in different stages in a stipulated timne period.
The studies are also carried out to measure the simultaneous occurrence
of certain phenomena or variables. For example, a researcher who wants to
establish the relationship between market flux and investment behaviour
might carry out a descriptive research to establish the correlation between
the two variables under study.

Conducting descriptive research

Descriptive research, as we stated earlier, is a framework used for a conclusive
research. It, however, lacks the precision and accuracy of experimental designs, yet it
lends itself to a wide spectrum of situations and is more frequently used in business
research. Based on the temporal collection of the research information, descriptive
research is further subdivided into two categories: cross-sectional studies and
longitudinal studies.
LEARNING OBJECTIVE 5
Understand and interpret Cross-sectional studies
cross-sectional and As the name suggests, the study involves a slice of the population just as in scientific
longitudinal designs. experiments one takes a cross-section of the leaf or the cheek cells to study the cell
Cros-sectional study structure under the microscope, similarly one takes a current subdivision of the
ivestigates aspecific churnk of population and studies the nature of the relevant variables being investigated.
the population under There are two essential characteristics of cross-sectional studies:
is study. It
scientific in its approach. The cross-sectional study is carried out at a single moment in time and
thus the applicability is most relevant for a specific period. For example,
 Research Methodology
attitude of Americans towards
60
cross-sectional study on the
post-9/11, was vastly
Asian-
different and a study done
ein
a behaviour towards the
Americans, pre- and attitude and population
reveal a different found earlier
2011 would absolutely in line with thatsection
of respondents
which might not bestudies are carried out on a
organizational from
Secondly, these (e.g., employees, voters,
population units under study sample is under consideration
and
the sectors). This
consumers, industry coordinate of the study.
time
under investigation only for the w¡ntedtofind out how to target the
cream company
case: A Danish ice
high-end ice creams.
Thus, they outsourced to alocal
" Illustrative in
consumer to indulge consumption habits of an upper class,
Indian find out the dessert March-May 2008 on 1,000
market research firm to study was conducted during
The
metro Indian consumer. the upper inconme bracket.
Indian metro
consumers in
revealed that most Indians have a sweet tol
conducted However, when th
The consumer survey regional concoctions at home.
specific desserts
and prefer to eat their
experimenting and generally look at exotic, foreign
are out, they love cream, especially in summer. The highlights of the
an ice
if lost for choice, opt for
findings were as follows: second
sample stated ice cream as thefirst plus the
" 92.6 per cent of the
choice.
cream as their first choice.
" 81per cent stated ice popular choice of most consumers.
" Regional brands were the however, only 15 per cent in the total
recall of foreign brands was,
" The
population.
globetrotters (who had made at least
The recall of foreign brands amongst two years)was 39 per cent.
five trips to a foreign country in the last person's social status is an
statement that a
" 92 per cent agreed with the
important determinant of who he/she is.
and 85 per cent believed that where
" 76 per cent believed, that what you eat
to.
youeat, are influenced by the social class you belong
fortnight, 72 per cent eat out once
" 83 per cent usually eat out once every
every weekend.
least once a week.
" 64 per cent eat an ice cream outside at
desserts, even if they
" 61.5 per cent were willing to experiment with exotic
were exorbitantly priced.
market, at least
The ice cream company concluded from the findings that the
better audience
Cross-sectional survey, in the metros, was ready. However, it was a niche segment and a
which is conducted on different base could be found amongst the savvy urban Indian traveller. Another conclusion
sample groups at different was that even though the ice cream was healthy and natural, it would have to take a
time intervals, is called cohort lifestyle positioning in order to melt the Indian heart.
analysis. There are also situations in which the population being studied is not of a
homogeneous nature and there is a divergence in the characteristics under study.
Thus it becomes essential to study the sub-segments independently. This variaton
of the design is termed as multiple cross-sectional studies. Usually this multi-sampe
analysis is carried out at the same moment in time. However, there might be instances
when the data is obtained from different samples at different time intervals and
then they are compared. Cohort analysis is the name given to such cross-sectional
surveys conducted on different sample groups at different time intervals. Cohorts
are essentially groups of people who share a time zone or have experienced an event
that took place at a particular time period. For example, in the 9/11 case, if westudy
and comnpare the attitudes of middle-aged Americans versus teenaged Americans
towards Asian-Americans, post the event, it would be a cohort analysis.
 Research Designs: Exploratory and Descriptive 61

The technique is especially useful in predicting election results, cohorts of

Imales-females, different religious sects, urban-rural or region-wise cohorts are
studied by leading opinion poll experts like Nielsen, Gallup and others.
Cross-sectionals studies are extremely useful to study current patterns or
behaviour or opinion. However, respondent's likelihood of future decisions or
delving too far in the past to determine the difference between the present and
the past behaviour is not a wise choice. In such cases, astudy that is anchored for
intormation collection at different moments in time is a better technique. The results
would be more reliable and valid, The advantage would be that rather than relying on
the respondent's memory or prediction,an actual monitoring of behaviour patterns
Would take place over tme.

Longitudinal studies
A
single sample of the identified population that is studied over a stretched period
Asinglesample of
the of time is termed as a longitudinal study design. Apanel of consumers specifically
identified population that chosen to study their grocery purchase pattern is an example of a longitudinal design.
is studied over a
stretched There are certain distinguishing features of the longitudinal studies:
period of time is termed as a " The study involves the selection of a representative panel, or a grOup of
longitudinal study design. individuals that typically represent the population under study.
group over
" The second feature involves the repeated measurement of the
made for the
fixed intervals of time. This measurement is specifically
variables under study.
" A distinguishing and mandatory feature of overthe design is that once the
sample is selected, it needs to stay constant the period of the study.
to be the same. Thus, in case
That means the number of panel members has
the panel, it is critical to replace
a panel member due to some reason leaves under study.
him/her with a representative member from the population components
in their temporal
Thus, the two descriptive designs basically differselection over time. However,
and secondly, in the stability of the sample unitobjectives. Also, though they are
which one is selected depends upon the research in practice, thetwo might merge
visualized conceptually as two endsof a continuum,
or complement each other in usage. that has just started a PGDM in
human
For example, a management school recruiters,
the stakeholders' (students,
resource management wants to ascertain programme structure and student quality
programme faculty) attitude toward the attitudes
monitor and alter the programme, relative to the changes in those
and to wants to measure this six-monthly,For at
Longitudinal studies are Specifically, suppose the B-school
often referred to as time
over time. months after the trainee has worked
on the job.
placements and six
series design due to the
the time of
design would be the longitudinal design. However, this might
repeated measurements taken this objective, the ideal but cannot be used for student
effectiveness as a
work for the recruiter population might not
Over time. year's pass outs would need to be studied. Thus, it
cross-section of that and instead
formulation ofa fixed panel of respondents for this purpose
require the for the post-training analysis. However, the
might be used
across-sectional sample fixed panel selected for monitoring the change over time.
faculty sample could be a consistency on the meásured variable over time,
change or
For determining alongitudinal toas the
studies. These are sometimes referred
the ideal design isthe repeated measurement overtime.
time-series design due to the
conducted?4 teogsi
How is it
CONCEPT 1. What is desciptive research? and longitudinal studies.
cross-sectional
CHECK 2. Differentiate between
 Research Methodoloqy
62
can be derived from the same
measurements, as stated above,
Repeated representative but different group selected
constant over time or on a would be under the domain of
sample, kept thoughthetwo collections
stage. Even
for everystudydesign, the obtained results andconclusions might be vastly different.
alongitudinal the illustrative case
given below.
This would be clear from
portfolio management
division of alarge private
case: The customer government
" Illustrative investment behaviour of bank customers inThis analbt
bank wanted to study the deposits.
securities, bullion and fixed
instruments,mutual funds and survey was done on
for a period of ive years. The
was done for every quarter in a year bank customers for each quarter and the resute
1,000
a different but stock sample of conclusions pertaining to the researcher's
obtained are shown in Table 3.1. Two with
emerged. First, government instruments were the most popular option,
attitude overall percentage of the division
approximately 45 per cent customers. Second, the
or less stable over time.
amongst the other three options is more
Quarter 2 Quarter 3 Quarter 4
TABLE 3.1 Quarter1
Use of
Resuts of longitudinal 43 45
Govt institutions 45 43
c3nk investment study
17 18 15
MF and others 21

22 21 19
Bullion 15

19 18 18 21
FD

Total 100 100 100 100

Another option that the bank had was to form a panel of the regular customers
and assess their periodic investments in these instruments; here the same group of
people would be interviewed in the five-year period. The findings and conclusions
obtained here would be slightly different, in case the sample remained the same.
Such a panel study, in addition to indicating an overall investment behaviour, would
have made it possible to monitor the options balanced between each other by the
same group over time, and also how overall the quarter still showed a uniform
pattern. This data will be available only if the customers studied remain constant at
each data collection phase.
To illustrate the advantage of longitudinal data, let us
consider two cases. The
results from the two are presented in Tables 3.2 and 3.3. In both the tables, the
the values under 'Row Total' represent the total figures,
investment made in the instrument
quarter 1 and the numbers under'Column Total' represent the
of quarter 2. The overall investment spread is the behaviour at the end
Thus, the results of the study as indicated earlier same at the end of each time period.
tables contain additional information still hold true. However, the twO
The first row of the numbers in about the movement of the decision taken.
Table 3.2 reveals that of the 45
invested in goverment securities in period 1, 25 consumers who
5 moved to mutual funds, 10 to invested in the same in quarter )
row of numbers in Table 3.3. bullion and 5 got FDs made. Now consider the irst
These
invested in government securities, 43numbers
still
reveal that of the 45 consumers wno
Atrue panel involves a money in mutual funds and one switched to invested in the same in period 2, lput
hs
committed sample group in the two cases can be bullion. The other investment options
that is more likely to Thus, in case one,similarly interpreted.
the investors who play safe
tolerate an extended or deposits more or less demonstrate the same and invest only in the nxeu
long data collecting fluctuate between options. In case two, behaviour. However, the other investors
sessions. conservative remain with the same
and however, the investors are more rigid and
options.
 Research Designs: Exploratory and Descriptive 63

Such longitudinal study using the same section of respondents thus provides
more accurate data than one using a series of different samples. These kinds of
panels are defined as true panels and theones using a different group every time are
called omnibus panels.
Advantages of a truepanel are that it has a more committed sample group that
is likely to tolerate extended or long data collecting sessions. Secondly, the profile
of time the
After acertainperiodchanged information is aone time task and need not be collected every time. Thus, a useful
panelmembers are respondent time can be spent on collecting some research-specific information.
perspectives can be
so that neW However, the problem is getting acommitted group of people for the entire
oblained
study period. Secondly, there is an element of mortality and attrition where the
members of the panel might leave midway and thereplaced new recruits might be
vastly different and could skew the results in an absolutely different direction. Athird
disadvantage is the highly structured study situation which might be responsible for
a consistent and structured behaviour, which might not be the case in the real or ield
conditions.
To deal with this, the research agencies making use of such panels try to mnake
certain that people behave normally and do not demonstrate exaggerated or artificial
behaviour. Also steps are taken to get new members who match the behaviour of
members are
the leaving members. Thirdly, after acertain period of time, the panel
changed so that new perspectives can be obtained.
designs, the
Thus, there are advantages and drawbacks in both the descriptive
and the degree
level of accuracy required, the nature of the monitored behaviour
variables determines the design
of influence of demographic and psychographic combination of the two for more
decision; or the researcher might decide to use a
accurate results.

TABLE 3.2 Customer investments Quarter 2

investment Customer Investments FD Row Total
Quarter 1 Government MF & Bullion
behaviour of regular instruments others
Consumers: Case 1
5 10 5 45
Govt institutions 25
4 21
MF & others
8 3 15
Bullion 4
13 19
FD 6
17 22 18 100
Column Total 43

Customer investments: Quarter 2

TABLE 3.3
Customer Investments FD Row Total
nvestment Government
Quarter1 MF & others Bullion
behaviour of regular instruments
CuStormers: Case 2 1 1 45
Govt institutions 43
3 2 21
16
MF & others
1 13 15
Bullion
5 14 19
0
FD
22 18 100
43 17
Column Total