Page 152 - [Uma_Sekaran]_Research_methods_for_business__a_sk(BookZZ.org)

136 THE RESEARCH PROCESS

Here, data were collected every month to assess whether UPS had regained
the business volume. Since data were collected at various points in time to
answer the same research question (have we regained lost ground?), the study is
a longitudinal one.

Example 6.27 A marketing manager is interested in tracing the pattern of sales of a particular
product in four different regions of the country on a quarterly basis for the next
2 years. Since data are collected several times to answer the same issue (tracing
pattern of sales), the study falls under the longitudinal category.

Longitudinal studies take more time and effort and cost more than cross-sec-
tional studies. However, well-planned longitudinal studies could, among other
things, help to identify cause-and-effect relationships. For example, one could
study the sales volume of a product before and after an advertisement, and pro-
vided other environmental changes have not impacted on the results, one could
attribute the increase in the sales volume, if any, to the advertisement. If there is
no increase in sales, one could conclude that either the advertisement is inef-
fective or it will take a longer time to take effect.
Experimental designs invariably are longitudinal studies since data are col-
lected both before and after a manipulation. Field studies could also be longitu-
dinal. For example, a study of the comparison data pertaining to the reactions of
managers in a company toward working women now and 10 years later will be
a longitudinal field study. Most field studies conducted, however, are cross-sec-
tional in nature because of the time, effort, and costs involved in collecting data
over several time periods. Longitudinal studies will certainly be necessary if a
manager wants to keep track of certain factors (e.g., sales, advertising effective-
ness, etc.) over a period of time to assess improvements, or to detect possible
causal connections (sales promotions and actual sales data; frequency of drug
testing and reduction in drug usage, etc.). Though more expensive, longitudinal
studies offer some good insights.

REVIEW OF ELEMENTS OF RESEARCH DESIGN

This concludes the discussions on the basic design issues regarding purpose of
the study, type of investigation, extent of researcher interference, study setting,
unit of analysis, and the time horizon. The researcher would determine the
appropriate decisions to be made in the study design based on the problem def-
inition, the research objectives, the extent of rigor desired, and cost considera-
tions. Sometimes, because of the time and costs involved, a researcher might be
constrained to settle for less than the “ideal” research design. For instance, the
researcher might have to conduct a cross-sectional instead of a longitudinal
study, do a field study rather than an experimental design, choose a smaller
rather than a larger sample size, and so on, thus suboptimizing the research
design decisions and settling for a lower level of scientific rigor because of