C 5 T R P: E O R D: Instructional Goals

CHAPTER 5
THE RESEARCH PROCESS: ELEMENTS OF

RESEARCH DESIGN
Instructional Goals
1. To clarify the various components of research design as indicated in Figure 5.1 of the
book.
2. To highlight obvious combinations in research design choices; for example, a lab or field
experiment is invariably conducted to establish cause and effect relationships; a lab
experiment is always done in a contrived setting with maximal researcher interference
and in a longitudinal fashion.
3. To stress the importance of making optimal research design choices aimed at balancing
scientific rigor and research costs (and feasibility).
Discussion Questions
1. What are the basic research design issues? Describe them in some detail.
Basic research design issues are primarily a function of the purpose of the study
(whether it is exploratory, descriptive, or hypothesis‐testing), and relate to such
aspects as the type of study to be done (causal or correlational), the setting in which
it will be done (natural or contrived), how much researcher control will have to be
exercised (very little in the case of field studies, to very much in the case of
experimental designs), how many times data will have to be collected (one shot
versus longitudinal), and the unit of analysis – i.e. the level at which data will be
aggregated. For most correlational studies, the field setting with minimal researcher
influence will be the choice. Most field studies are generally crosssectional, though
some could be longitudinal. Longitudinal studies, though better for understanding the
dynamics of the situation fully, also consume more time and resources. Thus, the
costs of a study also determine some of the design choices. The unit of analysis
depends on whether the research question focuses on individuals, dyads, groups, or
entire systems.
© 2009 John Wiley & Sons Ltd.

www.wileyeurope.com/college/sekaran 45
2. Why is it important to consider basic research design issues before conducting the
study and even as early as at the time of formulating the research question?
The research design ensures that the purpose for which a study is conducted is
effectively addressed. Some studies are focused on finding results which will be highly
reliable, whereas others might be interested in “getting some idea” of what is going on,
rather than wanting to know the “absolute truth”. If the rigor needed can be achieved at
only high cost (e.g. more manipulation and control, conducting the study over an
extended period of time with a big sample, etc), and if this cost is considered too much,
then the goals of research might have to be revised or even completely changed. Thus,
thinking through the research design issues at the early stages helps in averting several
problematic issues later.
3. Is a field study totally out of the question if one is trying to establish cause and effect
relationships?
Generally lab and field experiments, rather than field studies, are useful for
establishing greater internal validity, or faith in cause and effect relationships. However,
when longitudinal data are collected from field studies, or when particular types of
statistical analyses are done on cross‐ sectional data gathered from field studies, it is
sometimes possible to establish the existence of causal relationships. More specifically,
cross‐lagged correlation analysis (not discussed in the book) using data collected at
more than one point in time, can indicate causal relationships.
Even using cross‐sectional data (i.e. with data collected at one time only) statistical
techniques, such as path analysis, can identify cause and effect relationships. However,
in such cases, the data and the theory will have to lend themselves to certain
assumptions which are not always easily met.
4. “An exploratory study is just as useful as a predictive study”. Discuss this statement.
Without exploratory studies, new areas of knowledge cannot be developed. Every
theory we now have must at one time have started as an exploratory research study.
Exploratory research lays the groundwork for understanding new phenomena, and
developing new measures to test these. The exploratory studies of today blossom into
the predictive studies of tomorrow, and in that sense both types of study are useful and
essential to generate knowledge, understanding, and prediction of business
phenomena.
5. Why is the unit of analysis an integral part of the research design?
The unit of analysis is an important issue to be considered to find the right answers
to the research questions posed. The unit of analysis also determines the sample size.
For example, if one is interested in researching the factors that influence the stock
market in three different European countries, it is the behavior of stock markets in those
three countries that are of central interest to the study, and not the individual stock
market within each country. At the time of data analysis, the data gathered from each of
the stock markets within each country will somehow have to be meaningfully
aggregated, and only those three data points, which will form the three samples, have
to be taken into consideration. Thus, the unit of analysis is a function of the research
question posed, and is an integral part of the research design. As will be seen later,
research design decisions relating to sampling also depend on the unit of analysis. Let us
say a researcher decides to have a sample size of 30 for a study. Sampling 30 individuals
in an organization when the unit of analysis is individuals, is not as problematic as
sampling 30 organizations when the unit of analysis is organizations, or sampling 30
countries when the unit of analysis is countries. Thus, the unit of analysis influences
other decisions such as the sampling design, the sample size, data collection methods,
etc.
6. Discuss the inter‐relationships among: non‐contrived setting, purpose of the study,
type of investigation, researcher interference, and time horizon of the study.
A non‐contrived setting simply indicates that the research is conducted where the
flow of events normally takes place – i.e. in the natural system. Such research could be
either field studies or field experiments. The purpose of the field study could be for
exploration and understanding of phenomena, for describing phenomena, or for
hypothesis‐testing. The field study is a correlational study, and not usually initiated to
explore cause and effect relationships. Researcher interference in field studies is
minimal. Field studies could be either cross‐ sectional or longitudinal. Field experiments,
on the other hand, are undertaken to establish cause and effect relationships, are
longitudinal in nature, and researcher interference in field experiments is much more
than in field studies, since the independent variable will have to be manipulated. These
are mostly hypothesis‐testing studies, though they could be exploratory in nature, as
well.
7. Below are three scenarios. For each, indicate how the researcher should proceed with
the following, giving reasons:
a. The purpose of the study
b. The type of investigation
c. The extent of researcher interference
d. The study setting
e. The time horizon for the study
f. The unit of analysis.
Scenario A
Ms. Joyce Lynn, the owner of a small business (a woman’s dress boutique), has
invited a consultant to tell her how she is different from similar small businesses within a
60‐mile radius, in regard to her usage of the most modern computer technology, sales
volume, profit margin, and staff training.

Answer
This will be a descriptive study (describing how she compares with the others).
Data will be collected from small businesses on the use of computers, sales volume,
profit margin, and training programs and comparisons made. Some of these descriptions
might be qualitative, as for example, in describing training methods. This will be a field
study (using many similar small businesses), and researcher interference will be
minimal. It will be a one‐shot study, and the unit of analysis will be small business
systems.
Scenario B
Mr. Paul Hodge, the owner of several restaurants is concerned about the wide
differences in the profit margins of the various restaurants. He would like to try some
incentive plans for increasing the efficiency levels of those restaurants that are lagging
behind. But, before he introduces this, he would like to be sure that the idea will work.
He asks a researcher to help him on this issue.
Answer
This would be a causal, hypothesis‐testing study. It will be a field experiment,
using one or two restaurants to manipulate incentive plans and see if this causes an
increase in profit margins in those restaurants. Because of the manipulation, there will
be some researcher interference with the natural flow of events. The time horizon for
the study would be longitudinal since data will be obtained on sales, profits, etc., both
before and after the manipulation. The unit of analysis would be the restaurant.
Scenario C
A manager is intrigued why some people seem to derive joy from work and get
energized by it, while others find it troublesome and frustrating.
This would be a correlational, hypothesis‐testing, field study with minimal
researcher interference. It will be a one‐shot study and the unit of analysis will be
individuals.
Exercises
Doing the exercises in the chapter will help the student to apply the basic
elements of research design to different scenarios. More in‐class exercises are offered at
the end of this chapter (with answers), that can be used either in the classroom or as
exam questions.
Answers to Exercises in the Book
Exercise 5.1
A foreman thinks that the low efficiency of the machine tool operators is directly
linked to the high level of fumes emitted in the workshop. He would like to prove this to
his supervisor through a research study.
1. Would this be a causal or a correlational study? Why?
2. Is this an exploratory, descriptive, or hypothesis‐testing (analytical or predictive) study?
Why?
3. What kind of a study would this be: field study, lab experiment, or field experiment?
Why?
4. What would be the unit of analysis? Why?
5. Would this be a cross‐sectional or a longitudinal study? Why?
1. This would be a causal study because the operator wants to prove to the supervisor that
the fumes are causing operators to be low in their efficiency. In other words, the
machine tool operator is trying to establish the fact that fumes cause low efficiency in
workers.
2. This is an analytical study because the machine tool operator wants to establish that
fumes cause low efficiency and convince his workshop supervisor through such analysis
(i.e. establish cause and effect relationship).
3. This would be a field experiment. Though the study would be set up in the natural
environment of the workers where the work is normally done, the amount of fumes will
have to be manipulated while other factors, such as atmospheric pressure, may have to
be controlled. Because of the location of the study, it will be a field experiment.
4. The unit of analysis would be the individual operators. The data will be collected with
respect to each operator and then the conclusion will be made as to whether the
operators are less efficient because of the fumes emitted in the workshop.
5. This would be a longitudinal study because data will be gathered at more than one
point in time. First, the efficiency of the operators would be assessed at a given rate of
fume emission. Then the fumes emitted would be manipulated to varying degrees, and
at each manipulation the efficiency of the workers would again be assessed to confirm
that the high rate of fume emission causes a drop in operators’ efficiency.
Exercise 5.2
Many were concerned about the operations of the infamous BCCI, the
international banking institution. If the Government had desired to probe into the
details, would this investigation have called for:
1. A causal or correlational study? Why?
2. An exploratory, descriptive, or hypothesis‐testing study or case analysis? Why?
3. A field study, lab experiment, or field experiment? Why?
4. A cross‐sectional or longitudinal study? Why?
1. This would have been a correlational study since the details desired are to be probed
into without trying to establish cause and effect relationships.
2. It would be a descriptive study describing the operations of BCCI.
3. This would be a field study since data will be collected in the natural environment.
4. This will be a cross‐sectional study since the data will be collected at only one point in
time.
Exercise 5.3
You want to examine how exposure to thin or heavy models in advertisements influences
a person’s self‐esteem. You believe that the effect of exposure to models in
advertisements depends on the extremity of the model's thinness or heaviness.
Discuss the design decisions that you as a researcher will make to investigate this issue,
giving reasons for your choices.
An experimental study (causal study) will be needed since the goal of the study is to test
a cause‐and‐effect relationship.
A pretest will help you to select advertisements containing female models for the
following four conditions: moderately thin, extremely thin, moderately heavy, and
extremely heavy;
- participants view an advertisement booklet containing ads with female models.
- participants rate each model in terms of size (‐5 =extremely overweight, +5 =
extremely thin) and attractiveness (‐5 = extremely unattractive, + 5 = extremely
attractive).
Based on these scores, you will be able to select advertising models in each condition.
Note that models should differ from each other in terms of size but not in terms of
attractiveness (you have to perform post hoc comparisons to test for this).

In the actual experiment, participants must be randomly assigned to one of the four
conditions of the 2 (model size: thin vs. heavy) and (extremity of model size: moderate
vs. extreme) between‐subjects design. Participants are then provided with a booklet
containing ads: ads with models, pertaining to their condition, and filler ads with no
models (randomize the order of the ads). Eventually (after one or more filler tasks)
participants complete a self‐esteem scale such as for instance the Appearance
SelfEsteem Scale (Heatherton and Polivy 1991).
Exercise 5.4
You want to investigate the specific effects of specific emotions on customers'
behavioral responses to failed service encounters across industries.
Discuss the design decisions that you as a researcher will make to investigate this
issue, giving reasons for your choices.
Because it is difficult (and probably unethical) to induce specific emotions in an

experiment a field study seems to be the more logical choice. Note however, that when
you use a field study, it is very important to control for variables that might potentially
bias your results such as complaint success likelihood and switching costs. The critical
incident technique is a method that may help you to collect a wide variety of negative
experiences with service organizations.
Exercise 5.5
War on Cancer
Dr. Larry Norton of Memorial Sloan‐Kettering Cancer Center predicts that cancer
treatment will undergo major changes. Several drugs are being developed to battle
cancer without harming healthy tissues. It is a question of discovering which of
these drugs does the job best.
Design a study that would help find which drug would do the trick.
An experimental study (causal study) of the Solomon Four‐Group design will be
needed. A count of the healthy tissues in cancer patients at more or less the same stage
of cancer progress will first be taken, with about 20 patients in each group. Each group,
but the control group, will be treated with the different new drugs. The cancer cell
count will be taken both before and after treatment for a period of time (say, 3 to 6
months) to see which has the most beneficial effect. Thus it would be a longitudinal
study with some manipulation (drug variation) and control (matched or randomized
groups). The unit of analysis will be groups.

More in‐Class Exercises on Research
Design (with Answers)
Several situations are given below. For each situation, respond to the following research
design questions.
a. What would be the nature of the study that would be designed ‐ Exploratory,
descriptive, or hypothesis‐testing?
b. Give reasons for your answer to (a) above.
c. Would it be a Field study, Comparative study, Lab experiment or Field experiment?
d. Would it be a causal or correlational study?
e. Justify the reasons for your answer to (d).
f. Would it be a longitudinal or cross‐sectional study?
g. How would you explain your answer to (f) above? h. What would be the unit of
analysis? i. Defend your answer to (h) above.
1. A researcher wants to test the theory that low levels of oxygen in the mines is the
single most important factor that tires miners easily.
ANSWER:
a & b: Hypothesis‐testing, because the researcher wants to analyze and test if low
levels of oxygen would, in fact, cause tiredness in miners. c: A Field Experiment, because
there would be manipulation of the levels of oxygen
in the natural environment of the workers. d & e: Causal study, because the
researcher wants to see if low levels of oxygen
cause tiredness in miners. He wants to establish a cause and effect relationship
between the independent and the dependent variables. f & g: It would be a
Longitudinal study because data on tiredness will be collected
from the same batch of workers before and after each manipulation of the extent of
oxygen injected into the environment. h & i: The unit of analysis will be individual
miners because the data will not be aggregated across groups, but each individual
miner’s level of tiredness will be recorded and used in data analysis.
2. An Auditor is interested in knowing the relationship among the three variables –
depreciation, assets accounting, and taxes paid.
ANSWER:
a & b: Hypothesis‐testing, because the manager wants to analyze the correlations
among the three variables. c: Comparative Field study since the data will be
gathered from several firms. d & e: It will be correlational since the
relationship (and not cause and effect
relationships) among the variables is being studied. f & g: Cross‐
sectional since the data will be gathered only once. h & i: Unit
of analysis will be the firms.
3. The head‐nurse in a hospital would like to have a profile of the patients that come to
her ward.
ANSWER:
a & b: Descriptive since the head nurse only wants a profile – a mere description of
the type of patients who come to her ward. c: Field study, since the people coming to
her ward (natural setting) will be studied. d & e: It is a descriptive study, using
merely frequency distributions and means and
standard deviations. It is thus neither a causal nor a correlational study. f & g: Cross‐
sectional, since data will be gathered from each person just once to
get the profile. h & i: Unit of analysis will be individual patients since individual
profiles will be
studied.
4. A basic researcher wants to engage in a research project which will enable her to
establish the relationship between certain personality characteristics and the risk
taking tendencies of stock brokers.
ANSWER:
a & b: Hypothesis‐testing, since the researcher would have formulated a
conceptual framework of the type of personality characteristics that would induce
stock brokers to take more risks, and would want to test her hypotheses. c: Field
study since subjects will be studied in the natural environment in which
they usually operate. d & e: Correlational study since the relationships
among the personality
characteristics and risk‐taking are studied. f & g: Cross‐sectional since data will
be gathered at only one point in time.
h & i: Unit of analysis will be individuals (stock brokers) since the data analysis will
be done on each individual’s responses without any aggregation across groups.

Note: This can also be set up as a Lab experiment in which Finance majors are put in
a contrived setting and offered some investment assignments that call for risk taking.
Their personality characteristics can be obtained through a questionnaire survey, and
their risk taking behavioral tendencies assessed by manipulating the level of financial
risk involved in the investments.
5. Bob’s Nissan and Toyota dealership is suffering from a severe decline in the number of
car sales. Bob decides to test two approaches to sales. He randomly assigns his 12
salesmen to two groups. He then trains one group to use the “hard sell” approach and
the other to use the “soft sell” approach so that he can determine which method is
more effective.
ANSWER:
a & b: Predictive, hypothesis‐testing study since Bob wants to be able to test which
one of the two sales approaches will help sell more cars (i.e. predict which sales
approach is effective). c: Field experiment since the independent variable “sales
approach” is
manipulated to be either hard or soft sell and tried in the natural environment in which
sales usually take place and not in an artificial lab setting. d & e: Causal study, since Bob
wants to know which approach would cause sales to
increase. f & g: Longitudinal, because the sales levels, both before the
manipulation and
after, will be measured. h & i: Unit of analysis will be groups since the sales of each
of the two groups – “soft sell” and “hard sell” salesmen – will be aggregated and
compared.
6. A production manager wants to see if, by providing enhanced automation technology,

the productivity of the workers in the plants can be raised. However, before trying this
with all the five plants, she would like to try this with a group of apprentices who
work in one plant.
ANSWER:
a & b: Hypothesis‐testing, analytical study, since the manager wants to be able to
analyze the cause and effect relationship between enhanced automation and worker
productivity. c: Field experiment, since the study will be conducted in the natural
environment
where events usually take place. d & e: Causal study, since causal relationships
are trying to be established. f & g: Longitudinal, since productivity measures
will be taken from the same workers before and after the introduction of the

enhanced automation. h & i: Individuals will be the unit of analysis since
there will be no aggregation of
data across group members.

CHAPTER 8
DATA COLLECTION METHODS
Instructional Goals
1. To develop interviewing skills in students;
2. To help students become proficient in designing questionnaires;
3. To introduce students to different methods of data collection
including observational surveys, electronic surveys, and collecting information from
unobtrusive sources;
4. To emphasize the advantages of obtaining data from multiple
sources, and multiple methods of data collection;
5. To offer an idea of which data collection method would be
appropriate during different stages of research.
1. Describe different data sources, explaining their usefulness and disadvantages.
Secondary sources of data constitute all available sources from which existing
information can be extracted. This could be published or unpublished information in the
Library, in Government or Company records, in data tapes, on the internet, etc. Primary
sources of data can be any source which provides information elicited directly for the
first time. This could be organizations, individuals, trash cans, or whatever source from
which the desired information can be obtained.
Secondary data sources are useful because they offer a base on which to build
further research without having to reinvent the wheel. Past findings can spur new
research; past data can also offer opportunities to analyze them from a different
perspective or angle. Under certain circumstances, however, secondary data might be
misleading if one does not take the changed circumstances into consideration. For
example, if old archives of national data are utilized to examine and draw conclusions
about gender differences at the workplace, this might become an exercise in futility. The
workplace has changed much now compared to 20 years ago – there are more educated

career women, their aspirations are different, their current expectations of life are in
contrast to what they were two decades ago, etc. As such, one has to be cautious while
using secondary data sources.
Primary data will offer more correct and up‐to‐date information when phenomena
are investigated. However, it costs more to collect fresh data, access to organizations is
not always readily obtained, and people knowledgeable in research have to be coopted.
Generally, data collected through unobtrusive methods are more reliable than data
gathered through surveys.
2. As a manager, you have invited a research team to come in, study, and offer
suggestions on how to improve the performance of your staff. What steps would you
take to allay their apprehensions even before the research team sets foot in your
department?
As the manager, I would inform the employees of the Department, in a meeting
called for the purpose, that unless we can turn around and make some profits, the
organization may have to lay off people. Since management does not see laying off
personnel as an acceptable solution, at least at this stage, we are inviting outside
experts to examine our operations and give us some ideas on how we may be able to
increase our profits. If we all cooperate with the research team that will be coming in
shortly, and provide them with whatever information they might need, it would help the
company to implement the steps and processes necessary to increase output, make a
profit, and retain the current employees.
3. What is bias and how can it be reduced while interviewing?
Bias is any error that creeps into the information gathered. Biases occur whenever
incorrect information is provided to the researcher by whoever offers the information.
Sometimes, genuine mistakes might be made and hence incorrect information could be
provided. However, biases frequently occur because the researcher phrases the
question in a particular way, inflects the voice in a particular manner, or simply because
the respondent, zealous of pleasing the researcher, provides an answer that the
individual believes is what the researcher would like to hear, even though it is not
factual.
Bias can be reduced in interviewing by choosing the appropriate environment in
which the interview is conducted, by establishing trust with the respondent, by phrasing
questions so that they do not lead the interviewee to respond in a particular manner, by
clarifying issues when the respondent seems confused, by not coming across as
evaluative or judgmental, and by not placing special emphasis on specific words.
4. Explain the principles of wording, stating how these are important in questionnaire
design, citing examples not in the book.
a. The wording of the question and the level of sophistication of the language used are
important. These are important because, if the respondent does not understand the

words, obviously, he or she is not going to be able to respond to the questions. An
underground coal miner and a top business executive may need to be asked some of
the same issues differently. Thus, using words that fit the level of sophistication of
understanding of the respondent is necessary. To take an extreme example, to assess
the experienced quality of family life, a poorly educated farm hand may be first asked to
list the activities he normally engages in when he gets home from work, and then asked
to rate the extent of satisfaction gained from each of these activities. An index of quality
of family life might then be calculated by examining the family‐related activities listed
by the respondent. With a business executive, on the other hand, the same concept can
be tapped by asking the individual to rate his or her satisfaction to questions such as:
Balance between work and family life; the impact of travelling on experiencing quality
time with the family, and the like.
b. Both positively and negatively worded questions need to be asked in the questionnaire
so that the respondent does not mechanically answer the items without much mental
engagement. For example, if one question asks for the level of excitement that a
particular advertisement evoked, another could ask the extent to which that same
advertisement seemed irrelevant. Obviously, when both are tapped on the same scale,
the response to both cannot be the same. This break in the way the questions are
worded, both positively and negatively, helps to enhance the attention span of the
respondents who might otherwise respond mechanically.
c. Double‐barreled questions should be avoided. If one asks the question, “Do you value
and manage a diverse workforce well”? there may be no clear response given to the
question because one may value a diverse workforce, but hardly know how to manage
it! In such cases, it would be better to ask two separate questions rather than one.
d. Ambiguous questions should also be avoided. Questions such as “Do you discuss your
work with the president regularly”? or “Do you go to the movies frequently”? are
ambiguous, because the terms regularly and frequently are left open to the
interpretation of each respondent. One individual might discuss work with the president
on a daily basis, and another every six months. Both are “regularly” (or at regular
intervals) discussing work! But is this what the researcher is looking for? Similarly,
frequently going to the movies might mean one movie every week to one person, one
movie every month to another, and three times a week to a third person!
Since these terms are not defined, biased data will be procured from the respondents.
e. Recall‐dependent questions will also introduce biases. For example, not many people
remember when exactly they started smoking, or why they quit a particular school and
joined another when they were 10 years old.
f. Leading questions also result in biased responses because the interviewee might be led
to believe that a particular type of answer is sought and might try to oblige the
interviewer, even if the response is not what the individual believes to be true. An

example of a leading question is: Don’t you think that more women should be promoted
to decision‐making line positions in organizations? Such a question is likely to elicit the
obvious response, “Surely, Yes”!
g. Loaded questions should also be avoided because of the emotionality it invokes, thus
eliciting biased responses. As an example: “Don’t you think that the L.A.P.D. is biased
against blacks?” is a question with racial overtones which will bring forth emotional,
rather than objective responses to the question.
h. Questions invoking social desirability will also produce biased responses. An example
would be to ask if anyone felt that workforce diversity is bad for the system. In the
context of the changing demographics of the workforce, even if some felt that diversity
at the workplace is disastrous, they would tend not to disagree because it is not a
socially acceptable response.
i. It is also important to avoid lengthy questions which might confuse the respondent and
introduce several types of bias. An example of a long question is: To what extent would
asking for responses from different groups of employees, such as for instance,
production, marketing, R & D, advertising, and sales, on such issues as the policy
governing the organization, the organizational culture, staff discipline and other related
matters, provide valid data to the president, to enable him to make some critical
decisions on how to run the company effectively and efficiently?
5. What are projective techniques and how can they be profitably used?
Projective techniques help the respondent to offer answers to questions which
might otherwise be rather difficult or cumbersome to respond to. Projective methods
which include such techniques as word association, sentence completion, thematic
apperception, and ink‐blot tests, usually tap the deep‐seated motivations of the
respondent and provide relevant answers to difficult questions. However, the responses
have to be analyzed by individuals well trained in interpreting the answers if they are to
be meaningfully utilized for research purposes. Marketing research can benefit
substantially by using projective techniques, as for example, in developing products,
designing ads, and selecting appropriate media.
6. How are multiple methods of data collection from multiple sources related to
reliability and validity of the measures?
If there is convergence or strong correlation among the data obtained from different
data collection methods on the same variable, then one can establish convergent
validity because the data converge even when collected by different methods. The same
holds true when there is a strong and significant correlation between the data obtained
on the same variable from different sources. Though reliability cannot be directly
assessed from multi‐methods or multi‐sources of data collection, usually, since a valid
instrument is reliable (though the converse may not necessarily be true), one can
presume reliability if validity is established through the multi‐trait, multi‐matrix method.

7. Every data collection method has its own built‐in biases. Therefore, resorting to multi‐
methods of data collection is only going to compound the biases. Critique this
statement.
It is true that most data collection methods in the social sciences have their own
built‐in biases. For instance, interviews are susceptible to interviewer and respondent
biases, observational data will have observer and actor biases, and so on. But when data
are collected through multiple methods and compared, if the correlations among the
several types of responses received through the different data collection methods are
high, we can be fairly confident that the data have good validity and reliability –
i.e. they are “good.” If the correlations are low, however, we would start wondering
about the goodness of our data. We would then pay closer attention to the methods
used. For instance, if respondents say that they experience a high level of
meaningfulness in their jobs when they are interviewed, and when the concept is
measured through several items in the questionnaire, we would be inclined to treat
these as acceptable data even though both may have a low correlation to data obtained
from a sentence completion motivational research. If, however, there is no correlation
at all among any of the methods, the goodness of the data collected will be highly
suspect.
Thus, multi‐methods of data collection help us to make certain judgments about the
goodness of our data and serves to make decisions on which sets of data may perhaps
be more acceptable than others.
8. One way to deal with discrepancies found in the data obtained from multiple sources
is to average the figures and take the mean as the value on the variable. What is your
reaction to this?
Wherever possible “objective” data, through unobtrusive methods, need to be
obtained. For instance, if an employee’s performance is to be measured, it is better to
go through the person’s records and evaluation forms of the superiors (in case more
objective measures of output are not available). However, such records are usually
confidential in nature and are not likely to be made accessible to the researcher. Hence
data will be sought from multiple sources–for example, from the subject, the co‐
workers, the immediate supervisor, other superiors, and perhaps even the
subordinates. If there is not much consistency in the data obtained through these
various sources, there is no option left but to average these. However, the researcher
should point out the biases inherent in this and acknowledge this limitation in the
report.
9. How has the advancement of technology helped data gathering?
Hand‐held computers used in field interviews are very useful for collecting and
recording information from many individuals. The recorded information can be checked
for accuracy the same day.

Electronic surveys by way of interviews or questionnaires, have the advantage of self
correction in data, in the sense that out of range responses etc., are automatically
corrected on the spot, and skip patterns can be conveniently applied (e.g. if answer is
‘No’ to question # 5, proceed to question # 19).
The only drawback is that people might ignore and not respond to electronic
surveys, and some may not have the requisite facilities.
10. How will you use the data from observational study to reach scientific conclusions?
Observational studies have been profitably used for providing descriptive
information. However, by categorizing the responses in a meaningful way, and also
developing ordinal or Likert‐type scales therefrom, (e.g. High Frequency to Low
Frequency, etc.), scientific conclusions can be reached applying parametric or
nonparametric statistics, as appropriate.
11. The fewer the biases in measurement and in the data collection procedures, the more
scientific the research. Comment on this statement.
This observation is correct. When reliable and valid measures are used to tap
concepts, and when the data collection methods have less built‐in biases, we would
have good data, provided that, of course, the sample is representative. Thus,
replicability, accuracy and precision, and generalizability, become possible. The research
then becomes more scientific.
Exercises
Exercise 8.1
A production manager wants to assess the reactions of the blue‐collar workers
in his department (including foremen) to the introduction of computerintegrated
manufacturing (CIM) systems. He is particularly interested to know how they would
perceive the effects of CIM on:
a. their future jobs
b. additional training that they will have to receive
c. future job advancement.
Design a questionnaire for the production manager.
Questionnaire
Cria Software Enterprises
4571 Southwood Ave
Cupertino, California 94024

Date
Dear Employee,
As we had discussed in our meetings, Computer Integrated Manufacturing (CIM)
will form a part of our operations in the future. We would like to know how you
visualize certain aspects of the future environment as we introduce the changes. Please
take a few minutes to complete this short questionnaire and return it to the locked box
with the slit on the top, in the front office. Thank you for responding within the next five
days.
sd
George Coelho
Production Manager
1. Personal Information (place a mark on the appropriate box)
JOB STATUS NUMBER OF YEARS WORKED IN THE SHIFT CURRENTLY

DEPARTMENT WORKING IN
□ Machinist □ Less than 1 □ Day
□ Fitter □ 1‐3 □ Evening
□ Loader □ 3‐5 □ Night
□ Inspector □ 5‐10
□ Foreman □ Over 10 years
□ Surveyor
□ Other
2. Your Opinions regarding the following.

Please circle the appropriate number for each of the following items using the
scale below.
STRONGLY AGREE 1 AGREE 2 NEITHER DISAGREE 4 STRONGLY

AGREE NOR DISAGREE 5
DISAGREE 3
1. I will need additional training to 1 2 3 4 5 work in the

changed environment.
2. The new system will offer me 1 2 3 4 5 better
opportunities for advancement.

3. The opportunities for training will 1 2 3 4 5
have to be enhanced with CIM.
4. I am not sure if CIM will need all 1 2 3 4 5 the people
we now have in this department.
5. I feel that most of us may not have 1 2 3 4 5 better
opportunities for future promotions in the new manufacturing environment.
6. Most of us will need special 1 2 3 4 5 training to
work with CIM.
7. I am sure the future looks bright for 1 2 3 4 5 most of us here.
Items 4 and 7 measure opinion about their future jobs

1, 3, and 6 measure perceived training needs
2 and 5 measure job advancement
Exercise 8.2
Answers may vary.
Exercise 8.3
Answers may vary.
Exercise 8.4
Answers may vary. The questionnaire should include valid and reliable scales measuring service quality,
customer satisfaction, and customer loyalty.
Exercise 8.5
Design an interview schedule to assess the Intellectual capital, as perceived by
employees, in an organization after identifying the dimensions and elements.
The dimensions of intellectual capital would primarily comprise, among other
things, the following:
The knowledge‐base of the employees; Their
know‐how
The patents owned by the organization;
The secret formulas utilized by the company (special formulas / manufacturing process,
and the like).

The extent to which the company has an edge over others, in terms of specialized
knowledge and skills.
The Unstructured Interview Schedule would basically ask for the employees’
ideas as to what the term means and what they consider are the essentials of
intellectual capital.
The Structured Interview Schedule will be on the following lines:
1. Department:
2. Nature of job:
3. Age:
4. Designation:
5. To what extent do you think your organization depends on the knowledge of operations
of workers for its success?
6. To what extent do you think the success of this organization is due to the specialized
knowledge and skills possessed by employees?
7. Should some of the employees leave this organization, how difficult will it be to recruit
others with the same or similar knowledge?
8. To what extent is the know‐how relating to certain operations confined to particular
individuals who alone know how to do their job?
9. How many patents have been owned by your company in the last 5 years?
10. To what extent would you say your company has far more valuable patents than your
competitors in the industry?
Additional Exercises in Questionnaire Design

Two simple exercises on Questionnaire Design follow, and these can be used for
class discussion after the students have individually worked on them as homework
assignments. Different students can then be asked to write on the chalk board different
parts of the questionnaire – for example, the introduction, instructions to respondents,
measures for different concepts with scaling, and the end part of the questionnaire. The
class as a whole can then critique the exercise on the board. See also Exercises 10.1 to
10.5 in the Book. These can also be used for class discussions including discussions on
possible ways of establishing reliability and validity for the measures.
1. The Marketing Director of a pharmaceutical company wants to examine the
effects of number of sales personnel, population density, the average sales calls made to
the doctors in the region, and the average monthly sales calls made on the chemists in the
region, on the sales of the pharmaceutical products of the company. She has six particular

regions in mind. Design a questionnaire to do the study. Also discuss what sources of data
you will gather information from.
Population density can be obtained directly from the Census Bureau’s
Directory. All the other information should be readily available from the Regional Sales
Headquarter Records, or from the local offices. If necessary, a simple questionnaire
asking direct questions of the Regional Director to tap the variables could be used.
2. An R&D Project Manager wants to elicit the reactions of the machine
operators in the Production Department to three different types of machines currently in
use – Type A, Type B, and Type C. If he knows how the machinists experience these three
types in terms of ease of handling, sturdiness, flexibility for different adaptations, and
maintenance, he would be able to decide whether or not to develop a new prototype he has
in mind. Design a questionnaire for the purpose.
This is also a relatively simple questionnaire that will be administered to the
machinists, asking for their views on the three types of machines, aiming the questions
to particular aspects of ease of handling, sturdiness, flexibility for different adaptations,
and maintenance.
Next to Chapter 3, this Chapter is perhaps the one which would demand more in‐
class exercises and discussions to ensure that students have understood and can
incorporate the learnings into their research project work. When the group puts
together a draft of the questionnaire for administering it to the organization when they
do their class project, the instructor can play a key role in offering constructive criticisms
and suggestions, and all in all, help students to understand how to design reasonably
good questionnaires.

CHAPTER 10
SAMPLING
Instructional Goals
1. To impress upon the students that the hallmarks of science discussed at the beginning of
the course, i.e. generalizability, replicability, and precision and confidence are mainly
ensured by the sampling design chosen, including the sample size.
2. To explain how the probability sampling designs are representative of the population.
3. To highlight the advantages and the drawbacks of different types of probability sampling
designs.
4. To make students realize that probability sampling designs are not always feasible, and
sometimes non‐probability sampling designs are the only viable alternative available.
5. To elucidate that either too large or too small a sample size could be detrimental for
drawing meaningful and valid conclusions.
6. To explain the notions of efficiency in sampling and the trade‐off between precision and
confidence.
7. To clarify the choice points in sampling design shown in Figure 11.3 in the textbook and
provided in Powerpoint.
8. To introduce the need for identifying and articulating the sample characteristics in terms
of gender, age, job level, tenure, etc., for purposes of replicability, and for comparing the
results of different research studies.
Note: The above aspect in item 8 is brought out later in the book, but is worth stating now.
However, sensitizing students about sample characteristics at this stage will help them to
understand why a good description of sample characteristics is important and why they
should pay attention to it in their final report.
In‐Class Exercises on Sampling Designs

What kinds of sampling designs would be used for the following?:
a. A study to get a quick idea of the medical acceptability of a new aspirin substitute
which cannot be dispensed over the counter without prescription.

b. A study involving a sample of 325 students in a university where 2,000 students are enrolled.
c. An investigation of the career salience of professionals in the fields of medicine, engineering,
business, and law.
d. The generalizability of the attitudes of blue collar workers from a sample of 184, to the total
population of 350 blue collar workers in the entire factory of a particular company.
Answers
a. Purposive Judgment Sampling.
b. A Systematic sampling design (using university listing of students).
c. A Stratified Random sampling with stratification along profession, gender, age, etc.
d. Simple random sampling because of the importance attached to high generalizability.
Estimation of Population Characteristics from Sample Characteristics

The book works out a problem on the estimations of the population characteristics
from the sample characteristics for different levels of precision and confidence. Another
example that can be used in class is provided below.
Problem
1(a).You want to estimate the production days that would be lost during the next
three months by sampling the vacation intentions of a few employees. You randomly
select 36 employees in the organization and find that the average number of days they
intend taking off is 16 during the coming three Summer months, with a standard
deviation of seven (7) days. Based on these sample statistics, you want to estimate at a
99 percent confidence level, the days that will be lost due to the entire population of
workers taking vacation time during the next three months, so that the plant manager
knows how much temporary help he should plan on hiring during the summer months in
order for work to proceed smoothly.
Calculation
μ X zSx

Sx  S/ n  7/6 1.167 μ
16 2.576 1.167 
163.01
12.99to19.01
If there are 100 employees in the organization expected to take vacation during
Summer, then the most optimistic estimation of the days lost through vacation time
during the summer would be (13  100 =) 1,300 days and the most pessimistic would be
(19  100 =) 1,900 days. This would mean that temporary help would be needed
anywhere between 1,300 and 1,900 days worth of labor, for production to proceed
smoothly.
1(b). Now you might want to narrow the gap in the estimation since 1,300 and
1,900 days is a wide range, and you would like more precision using the same
information that is available to you. Well, it can be done only at the cost of taking more
chances that your estimation, though more precise, may not be stated as confidently as
before. You may, however, decide to lower your confidence level to gain more precision
in your estimation. At the 95 percent confidence level, the most pessimistic and the
most optimistic estimations would range from 1372 to 1828, a narrower range – we are
now gaining more precision, i.e., narrowing confidence interval – but only at the risk of
being wrong five percent (as opposed to one percent) of the time. We could seek a
further narrower interval range at the 90 percent confidence level (1408 to 1792), if
taking a higher risk in being wrong in the estimation is not going to be disastrous. You as
a manager will have to decide on the amount of risk you want to take by increasing or
decreasing the confidence level.
Estimating Sample Size

Estimating the sample size for given levels of required accuracy and precision is
discussed in the book with examples. Further problems can be given to students in class.
A sample problem given to them could be as follows.
Problem
How large a sample do you need if you want to estimate the mean age of a
population of 100 residents and be within 3 years of the true mean with a 95 percent
level of confidence, when you know that the sample mean is 28 years and the standard
deviation is 17 years?

Solution
Sx  3/1.96 1.53
Sx s/ n
1.53 17/ n
1.53 n 17 n
17/1.53
n 123
1. Identify the relevant population for the following research foci, and suggest the
appropriate sampling design to investigate the issues, explaining why they are
appropriate. Wherever necessary, identify the population frame as well.
a). A company wants to investigate the initial reactions of heavy soft‐
drink users to a new “all natural” soft‐ drink.
“Initial reactions” non‐probability sampling.

Heavy soft‐drink users: purposive sampling.
b). A gun manufacturing firm would like to know the types of guns possessed by various age
groups in Washington, D. C.
When guns are sold, the dealers are expected to maintain records of the names and
ages of the buyers, and the types of guns sold. Thus, the population frame for gun
owners can be had from the records of gun dealers in the Washington, D.C. area. Once a
complete list is compiled, a stratified random sampling of appropriate age groups can be
done, and the relevant information obtained.
c). A hospital administrator wants to examine if single parents working in the hospital have
a higher rate of absenteeism than parents who are not single.
The marital status and parental status of the employees may be available from the
records of the hospital. If so, the population frame would be the hospital records from
which the information can be directly extracted. A stratified random sampling design can
be used and a t‐test done to examine if there are differences in absenteeism in the two
groups. If the marital or parental status of the employees is not available, then a
convenience sample of those who are known to be single parents and those who are
known to be not single parents can be used to check their records for absenteeism.

d). A researcher would like to assess the extent of pilferage in materials from storage warehouses
of manufacturing firms in the East coast.
First, the researcher will have to randomly select some big, medium‐size, and small
cities in the East coast – say, two of each. Thereafter, three warehouses from each of the
large, medium, and small size cities will be randomly selected. This then will form the
sample for study. The pilferage rates can then be assessed by checking the records of
original supplies to the warehouses, the amount of stock sold, and the balance at hand.
e). The Director of Human Resources wants to investigate the relationship between drug abuse
and dysfunctional behaviors of blue collar workers in a particular plant.
Identifying drug abusers is very difficult, unless there is a mandatory drug test
administered. Also, what is meant by ‘dysfunctional behaviors’ has to be clearly defined.
If a list of drug abusers is available through drug tests then, depending on the numbers,
either all of them (if the number is small), or a random sample among them (if the
number is large) can be observed for any dysfunctional behaviors and compared with
the behaviors of a random sample of those who are not drug abusers.
f). A marketer wants to generate some ideas on how
women differ from men in acquiring product knowledge
about cars.
Some ideas = non‐probability sampling
Women versus men = quota sampling.
2a). Explain why cluster sampling is a probability sampling design.

b). What are the advantages and disadvantages of cluster sampling?
c). Describe a situation where you would consider the use of cluster sampling.
a. Cluster sampling is a probability sampling design since each cluster has a known chance
of being selected as the sample.
b. Cluster samples offer ease of data collection within a relatively short time frame.
However, being homogenous, they do not offer the advantage of efficiency in sampling.
c. Cluster sampling could be used to assess the types of films (comedies, fighting, violence,
social dramas, adventure, etc.,) that are most popularly rented by theater owners.
Different theater complexes, each of which might have anywhere from four to eight
different films being shown at any given time, could be treated as clusters. More
homogeneity within clusters, and heterogeneity between clusters can also be expected
since different complexes might be catering to different types of clientele. As such,
cluster sampling would be ideally suited for this purpose.
3a). Explain what precision and confidence are and how they would influence sample size.
Precision indicates how close the sample statistics are to the population

parameters. Usually, a bigger sample size offers greater precision than a smaller one,
especially when the population is heterogeneous in the characteristics of interest to the
study. Confidence indicates the percentage of times when our projected statistics will be
truly reflecting the population parameters. This, again, is a function of the sample size – a
bigger sample size will allow for a projection to be made at a higher level of confidence.
But, for any given sample size, we cannot increase both precision and confidence
simultaneously.
3b). Discuss what is meant by the statement: There is a trade‐ off between precision and
confidence under certain conditions.
There is a trade‐off between precision and confidence as a function of sample size. For
any given sample size, if the precision has to be increased, then the level of confidence will
go down. On the other hand, if one needs more confidence, then the precision will have to
go down. In other words, for any given sample size, either the precision or the confidence,
but not both, can be increased. As a matter of fact, either can be increased only at the
expense of the other.
4. The use of a convenience sample in organizational research is correct because all
members share the same organizational stimuli and go through almost the same kinds
of experiences in their organizational life. Comment.
It is not correct to say that all organizational members share the same kinds of
experiences because different supervisors, different departments, and different work
environments offer different kinds of stimuli to organizational members. Hence,
selecting a few members based on convenience sampling and generalizing the results to
the total population would not be advisable. Even among the same group of people
exposed to the same kinds of stimuli, perceptions will vary due to individual differences
in selective perception and interpretation. Hence, a probability sampling design, rather
than a convenience sample would be more appropriate to project the properties of the
sample on to the population.
5. Use of a sample of 5,000 is not necessarily better than having a sample of 500. How
would you react to this statement?
Not only is a sample of 5,000 not necessarily better, but it could also be detrimental,
since the probability of rejecting the null hypothesis becomes too large – i.e., every
trivial difference becomes significant. We could compare this to using an electron
microscope to examine the differences between two atoms. The differences will be
enormously magnified if we use the “power” of the electron microscope when, actually,
the difference between two atoms is negligible. In fact, statistical tables are available
which specify the appropriate sample size for desired levels of confidence. Too big a
sample size is likely to lead to erroneous conclusions, since with larger sample sizes, even
small correlations reach statistical significance, and thus, the probability of our
committing Type 1 error is high.

6. Non‐probability sampling designs ought to be preferred to probability sampling designs
in some cases. Explain with an example.
The above statement is true in cases where the information needed for research can
be obtained only from a select category of people who are not plentiful in numbers and
for whom a population frame does not exist. In such cases, the nonprobability purposive
judgement sampling design is the most useful one. Here, the most conveniently
available members of the limited population of individuals who have the necessary
information are contacted to obtain information. This is the only meaningful way to get
information since a simple random sampling or any other probability sampling design is
impossible without the availability of a relevant population frame. As an example,
information on the effectiveness of AZT (even after taking into consideration its side‐
effects), for cancer patients, can be best obtained only from those who are suffering
from cancer and using AZT.
7. Because there seems to be a trade‐off between accuracy and confidence for any given
sample size, accuracy should always be considered more important than precision.
Explain with reasons, why you would or would not agree with this statement.
Whether confidence or precision is more important is a decision that each
researcher has to make, based on the goals of the research. In some cases, confidence
may be more important to the research, and in others, precision. For instance, a
salesperson might want to place greater emphasis on the confidence in the projected
number of sales he is likely to make, than wanting to know the exact number of sales he
will make – i.e. he desires greater confidence in the projection than precision. In other
words, he may be willing to accept an error margin of seven percent (rather than a
three percent) so long as he is sure that the chance of being off the mean is no more
than .01 percent! On the other hand, a marksman might insist on more precision (how
close he will hit to the bull’s eye, thus testing his potential abilities to shoot the target)
and not so concerned about the percentage of times he is likely to hit the bulls eye (the
confidence level). He might prefer precision, because if he knows he has the abilities, he
can practice more and improve the probability of his chances for success. These
examples illustrate the trade‐off between accuracy and precision. The decision whether
accuracy or confidence is more important has to be made by the researcher for good
reasons, and one may not always necessarily be preferred over the other.
8. Over‐generalizations give rise to a lot of confusion and other problems for researchers
who try to replicate the findings. Explain what is meant by this.
A careful description of the sample characteristics is necessary to enable the reader
to judge the extent to which the conclusions drawn from the research findings are
generalizable. If the sample is not adequately described and sweeping generalizations
are made, then the researchers who try to replicate the findings will run into difficulties,
since the findings of the previous study may not be replicated, simply because what is
relevant to one population may not be so for another. The conflicting findings would
cause confusion and researchers will not know whether the discrepancies are due to

faulty theories or methodological problems. It is for this reason that research reports
should carefully describe the population and the sample characteristics.
9. Double sampling is probably the least used of all sampling designs in organizational
research. Do you agree? Provide reasons for your answer.
Double sampling is probably used more in the exploratory stages of research where
data are collected several times from subsets of a sample – e.g., a subset of those who
were interviewed with unstructured questions are again interviewed with structured
questions in order to obtain more specific and focused information on phenomena that
may not be well‐understood. For instance, double sampling is not uncommon in cross‐
cultural research, in research on dual career families, and the like. While quantitative
data through questionnaires are essential, for more in‐depth probing of certain
phenomena, structured interviews are sometimes conducted with a subset of the
sample from which questionnaire responses have been obtained.
10. Why do you think the sampling design should feature in a research proposal?
Details of the sampling design, giving details of how many and what cadres of
personnel will be included in the survey, helps the sponsor to see if such people would be
available and could respond to the survey. It also helps determine the extent to which the
findings will be generalizable to the entire organization.
In published studies, it also helps the reader become aware of the settings to which
the findings can be generalized.
Exercises
For the situations below, what would be the relevant population and the most
appropriate sampling design? Make sure you discuss the reasons for your answers.
Exercise 10.1
The medical inspector desires to estimate the overall average monthly
occupancy rates of the cancer wards in 80 different hospitals which are evenly located
in the Northwestern, Southeastern, Central, and Southern suburbs of New York City.
All the 80 hospitals in the four parts of New York City form the population from
which the overall occupancy rates are to be established. A simple random sampling design
would be best suited because hospitals of different sizes, and those serving clientele from
different socio‐economic status, would then be included in the survey, and be most
representative of the population.

Exercise 10.2
A magazine article suggested that “Consumers aged 35 to 44 will soon be the
nation’s biggest spenders, so advertisers must learn how to appeal to this “over‐thethrill
crowd”. If this suggestion appeals to an apparel manufacturer what should be the
sampling design to assess the tastes of this group?
The population would, of course, be all those between the ages of 35 and 44.
However, targeting this population would be difficult since a population frame would not
be available (excepting with the Bureau of Census). A good way to obtain access to a
portion of this population could be through retail store outlets like J.C. Penney, Ross, or
Mervyns. If their cooperation can be obtained to generate a list of this age group through
their credit card information (which might have customers’ date of birth on them), then a
random sample or a systematic sample of this population can be used to administer a
mail survey. The responses obtained from consumers from a variety of cities can then be
used to formulate an idea of what the tastes of the target group are.
Exercise 10.3
The McArthur Co. produces special vacuum cleaners for conveniently cleaning
the inside of cars. About a thousand of these are produced every month with stamped
serial numbers and stored serially in a stock room. Once a month, an inspector does a
quality control check on fifty of these. When he certifies them as to quality, the units
are released from the stock room for sale. The production and sales managers,
however, are not satisfied with the quality control check, since quite often, many of
the units sold are returned by customers because of various types of defects. What
would be the most useful sampling plan to test the fifty units?
A systematic sampling design would be most useful since a probability sampling
design is called for, and also because it would be easy to locate the pieces to be sampled
since they have serial numbers and are stored according to the serial numbers.
Exercise 10.4
A consultant had administered a questionnaire to some 285 employees using a
simple random sampling procedure. As she looked at the responses, she suspected
that two questions might not have been clear to the respondents. She would like to
know if her suspicion is well founded.
A double sampling design will be used in this case. That is, a sub‐sample of the
original sample can be approached to see whether or not the two questions were clear
to them.

Exercise 10.5
The Executive board of a relatively small university located in Europe wants to
determine the attitude of their students toward various aspects of the university.
The university, founded in 1928, is a fully accredited governmentfinanced university
with 11,000 students. The university specializes in the social sciences and humanities
and has 5 faculties, 6 service departments, 8 research centers, and 2 graduate schools.
The Executive board has asked you to come up with a sampling plan. Develop a
sampling plan and pay attention to the following aspects: target population, the
sampling frame, the sample technique and the sample size.
The population will comprise all the students of the University. A sampling frame
is available; hence probability sampling is a viable option. There are identifiable
subgroups (students of the five different faculties and graduate schools) which may be
expected to have different parameters to the variable of primary interest (attitude). For
that reason you may want to use a stratified sampling design. A minimum sample size of
30 for each stratum is necessary.
Exercise 10.6
T‐Mobile is a mobile network operator headquartered in Bonn, Germany. The
company has enlisted your help as a consultant to develop and test a model on the
determinants of subscriber churn in the German mobile telephone market. Develop a
sampling plan and pay specific attention to the following aspects.
Define the target population. Discuss, in as much detail as possible, the sampling frame
and the sampling design that you would use. Give reasons for your choice.
Target population: German, T‐mobile, mobile phone subscribers.

Sampling Frame: A database of subscribers on a recent date (for instance 6 months ago).
This database will include both customers who have remained loyal to T‐mobile and
customers who have switched mobile phone providers.
Sampling design: proportionate stratified sampling with loyal customers and customers
who have switched mobile phone providers in proportion to their original numbers in
the population.
CHAPTER 11

Quantitative Data Analysis
Instructional Goals
1. To make statistics come alive and become meaningful to students through examples that
they can relate to.
2. To get students to analyze data using any of the available PC software programs.
3. To impress on students that the data collected should first be edited, the blank responses
given an appropriate value, and then entered for analysis.
4. To stress the importance of getting a feel for the goodness of data after recoding.
5. To explain how the frequency distribution becomes a tool for explaining sample
characteristics.
6. To connect the application of descriptive statistics such as the means, standard deviations,
and variance to the dependent and independent variables of interest to any study in order to
get an idea of the central tendencies and to obtain a feel for the data.
7. To make sense out of Pearson Correlations, by interpreting significant correlations, and to
sensitize students to extremely high correlations (over .8) which might cause concerns
regarding the discriminant validity of the measures.
8. To help them interpret the results they obtain from data analyses.
Exercises on Data Interpretation

Practice in the interpretation of the results of data analyses can be encouraged
through the exercises below. These can be done either as in‐class exercises or given as
take‐home assignments.
Exercise #1 On Frequency Distributions

Below is a tabulation of the demographic data from the Frequency distribution of
a survey done by Ms. Sandra Jones. Her sample consisted of 148 of a total of 3,700
clerical employees in three service organizations. Based on the tabulation provided
below, describe the sample characteristics.
Table 1: Frequency Distributions of Sample (n = 148)

RACE EDUCATION GENDER
Non‐whites = 48 (32%) High School = 38 (26%) Males = 11(75%)

Whites = 100 (68%) College Degree = 74 (50%) Females = 37 (25%)
Masters Degree = 36 (24%)
AGE # OF YEARS IN ORG. MARITAL STATUS

< 20 = 10(7%) < 1 year = 5 (3%) Single 20 (14%)
20‐30 = 20(14%) 1‐3 = 25(17%) Married 108 (73%)
31‐40 = 30(20%) 4‐10 = 98(66%) Divorced 13 (9%)
>40 = 88(59%) >10 = 20(14%) Alternative7 (4%)
Lifestyle
Answer
The tabulation indicates the percentages on all the demographic data (as shown
in parentheses above). A profile of the respondents will be as follows.
Population and Sample

The population consisted of 3,700 clerical employees in three service
organizations. Of these a sample of 148 clerical employees was drawn. Based on the
demographic characteristics provided in the table, we can say that the majority of the
respondents are white (67.6 percent of the sample), have at least a college degree (50
percent with college degrees and 24 percent with Masters degrees), are married (73
percent), are above 40 years of age (59 percent), and have served in their organizations
for four to ten years (66 percent).
Further breakdowns of educational levels can be given, depending on the nature
of the study, and the necessity for detailed descriptions.
Exercise #2 on Means, Standard Deviations, and Other

Statistics
Here is another tabulation of the Means, Standard Deviations, etc., for Ms. Jones’
data. How would you interpret these data?
Table 2: Means, Standard Deviations and Other Statistics
VARIABLE MEAN STD. DEV MODE MIN MAX

Age 37.5 18 38 20 64

# of Years Married 12.1 24 15 0 32
Stress 3.7 1.79 3 1 5
Job Involvement 3.9 1.63 4 2 5
Performance 3.6 0.86 3 3 5
Answer
The mean age of the respondents is 37.5 and since the mode is also 38, it can be
presumed that many of the respondents are about 38 years old. The average number of
years the respondents have been married is about 12, with some not married at all
(minimum = 0).
The amount of stress experienced by the subjects in this sample seems to be a little
over the average, but not high (3.7 on a 5‐point scale); the same can be said for their
performance level (3.6 on a 5‐point scale). The extent of their job involvement is close to a
4 (four) on a 5‐point scale and can be considered to be moderately high.
While there is more variance in stress and job involvement (1.79 and 1.63), the
variance in performance is rather low (.86). This would indicate that there is greater
dispersion in the experienced stress and extent of job involvement, but most of the people
have been performing at the same “just above the average” level.
We would thus describe the profile of the average worker in these three
organizations, assuming that the sample is representative of the population, as being about
38 years old, married for about 12 years, performing at just above the average level, fairly
involved in the job and experiencing a moderate amount of stress.
Exercise #3 on Correlations
From the same research done by Ms. Jones, the following inter‐correlation matrix
is shown. Interpret these results.
Table 3: Pearson Correlations
VARIABLE AGE # OF YRS. MARRIED STRESS JOB INVOLVEMENT PERFORMANCE
Age 1.0
# of yrs. married .86 1.0
Stress .43 .61 1.0
Job Involvement .53 .32 .58 1.0
Performance .09 .06 .49 .36 1.0
a. All correlations above .30 are significant, at least at the .05 level.

b. All correlations above .50 are significant, at least at the .01 level.
Answer
Eight correlations in the above matrix are significant at least at the .05 level, of
which four are significant at least at the .01 level. Only two of the ten correlations in the
matrix are not significant.
Age and the number of years married are highly correlated (r = .86) and this is not
surprising because older people are likely to have been married for a longer period of
time than younger persons.
Older persons also experience less stress (r = .43) possibly because they have either
learned the ropes well and things do not stress them, or they are better able to handle stress
and hence hardly experience them as stress, or they are given easy jobs.
Older people are also more job involved (r =.53), but age is not related to
performance. That is, though the older employees may be more involved in their jobs than
the younger ones, they do not necessarily perform better than the latter.
Those experiencing more stress are more involved in their jobs (r = .58) and also
perform better (r = .49). The more the job involvement, the better the performance (r
= .36). The above relationships do not indicate causality, merely correlations.
Those married for a longer period experience less stress, possibly because the
family demands are less on them since they would have settled down to a routine family
life which does not induce stress. The persons who are married for longer periods of time
are also more involved in their jobs. But the length of marriage is not related to
performance (r = .06).
If performance is the main variable of interest to this study we would say that
neither age nor the length of marriage has any relationship to performance, but that the
amount of stress and the level of job involvement do have significant bivariate
relationships with the performance of employees. The more job involved people are, the
better they perform; the more stress they experience, the better they perform. Note:
Based on the correlations among stress, job involvement, and performance, it would be
inappropriate to conclude that the high performer gets job involved and hence
experiences stress. Causal inferences cannot be drawn from correlational analyses.
1. What activities are involved in getting the data ready for analysis?

After data are obtained through questionnaires, they need to be coded, keyed in, and
edited. That is, a categorization scheme has to be set up before the data can be typed
in. Then, outliers, inconsistencies, and blank responses, if any, have to be handled in
some way.
2. What does coding the data involve?
Data coding involves assigning a number to the participants’ responses so they can be entered
into a database.
3. Data editing deals with detecting and correcting illogical, inconsistent, or illegal data in
the information returned by the participants of the study. Explain the difference between
illogical, inconsistent, and illegal data.
An example of an illogical response is an outlier response. An outlier is an observation

that is substantially different from the other observations. An outlier is not always an
error even though data errors (entry errors) are a likely source of outliers. Because
outliers have a large impact on the research results they should be investigated carefully
to make sure that they are correct.
Inconsistent responses are responses that are not in harmony with other information.
Illegal codes are values that are not specified in the coding instructions.
4. How would you deal with missing data?
One way to handle a blank response is to ignore it when the analyses are done. An
alternative solution would be to look at the participant’s pattern of responses to other
questions and from these answers deduce a logical answer to the question for the
missing response. A second alternative solution would be to assign to the item the mean
value of the responses of all those who have responded to that particular item. In fact,
there are many ways of handling blank responses, each of them having its own
particular advantages and disadvantages.
5. What is reverse scoring and when is reverse scoring necessary?
Take for instance the perceived inequity measure of the Excelsior Enterprises case.
Perceived inequity is measured by five survey items: (1) ‘I invest more in my work than I
get out of it’; (2) ‘I exert myself too much considering what I get back in return’ ; (3) ‘For
the efforts I put into the organization, I get much in return’ (reversed); (4) ‘If I take into
account my dedication, the organization ought to give me a better practical training’;

and (5) ‘In general, the benefits I receive from the organization outweigh the effort I put
in it’ (reversed). For the first, second and fourth item, a score indicating high agreement
would be negative, but for the third and fifth question a score indicating high agreement
would be positive. To maintain consistency in the meaning of a response, the first,
second, and fourth item have to be reverse scored (note that we are measuring equity
and not inequity). In this case a 5 (I completely agree) would be transformed to a 1 (I
completely disagree), a 4 to a 2, and so forth.
6. There are three measures of central tendencies: the mean, the median, and the mode.
Measures of dispersion include the range, the standard deviation, the variance (where the
measure of central tendency is the mean), and the interquartile range (where the
measure of central tendency is the median). Describe these measures and explain which
of these measures you would use to provide an overview of a) nominal, b) ordinal and c)
interval data?
Scale Measures of Central Measures of Visual Summary Measure of Relation Visual Summary of
Tendency …for a Dispersion ‐‐‐‐‐ …between Variables Relation
Single Variable …for a Single ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐…for …between
Variable a Single Variable Variables
Nominal Mode ‐ Bar chart, pie chart Contingency Table Stacked bars,
(Cross‐Tab) Clustered bars
Ordinal Median Semi‐interquartile Bar chart, pie chart Contingency Table Stacked bars,
range (Cross‐Tab) Clustered bars
Interval Arithmetic mean Minimum, maximum, Histogram, scatter Correlations Scatterplots
standard deviation, plots, box‐ and
variance, coefficient whisker plot
of variation
Ratio Arithmetic or Minimum, maximum, Histogram, scatter Correlations Scatterplots
geometric mean standard deviation, plots, box‐ and
variance, coefficient whisker plot
of variation
Measures of Central Tendency
The Mean. The mean or the average is a measure of central tendency that offers a
general picture of the data without unnecessarily inundating one with each of the
observations in a data set. The mean or average of a set of say, 10 observations, is the
sum of the 10 individual observations divided by 10 (the total number of
observations).
The Median. The median is the central item in a group of observations when they are arrayed
in either an ascending or a descending order.

The Mode. In some cases, a set of observations would not lend itself to a meaningful
representation through either the mean or the median, but can be signified by the most
frequently occurring phenomenon.
Measures of Dispersion
Range. Range refers to the extreme values in a set of observations.
Variance. The variance is calculated by subtracting the mean from each of the
observations in the data set, taking the square of this difference, and dividing the total
of these by the number of observations.
Standard Deviation. The standard deviation, which is another measure of dispersion for
interval and ratio scaled data, offers an index of the spread of a distribution or the
variability in the data.
Other Measures of Dispersion. When the median is the measure of central tendency,
percentiles, deciles, and quartiles become meaningful. Just as the median divides the
total realm of observations into two equal halves, the quartile divides it into four equal
parts, the decile into 10, and the percentile to 100 equal parts. The percentile is useful
when huge masses of data, such as the GRE or GMAT scores, are handled. When the
area of observations is divided into 100 equal parts, there are 99 percentile points. Any
given score has a probability of .01 that it will fall in any one of those points. If John’s
score is in the 16th percentile, it indicates that 84% of those who took the exam scored
better than he did, while 15% did worse.
Oftentimes we are interested in knowing where we stand in comparison to others — are

we in the middle, in the upper 10 or 25%, or in the lower 20 or 25%, or where? For
instance, if in a company‐administered test, Mr. Chou scores 78 out of a total of 100
points, he would be unhappy if he were in the bottom 10% among his colleagues (the
test‐takers), but would be reasonably pleased if he were in the top 10%, despite the fact
that his score remains the same. His standing in relation to the others can be
determined by the central tendency median and the percentile he falls in.
The measure of dispersion for the median, the interquartile range, consists of the
middle 50% of the observations (i.e., observations excluding the bottom and top 25%
quartiles). The interquartile range could be very useful when comparisons are to be
made among several groups. For instance, telephone companies can compare
longdistance charges of customers in several areas by taking samples of customer bills
from each of the cities to be compared.
7. A researcher wants to provide an overview of the gender of the respondents in his

sample. The gender is measured like this:

What is your gender?
0 Male
0 Female
What is the best way to provide an overview of the gender of the respondents?
Provide a frequency table.
8. Consider the following reliability analysis for the variable customer differentiation.
What could you conclude from it?
R E L I A B I L I T Y A N A L Y S I S ‐ S C A L E (A L P H A)
Item‐total Statistics
Scale Scale Corrected

Mean Variance Item‐ Alpha
if Item if Item Total if Item
Deleted Deleted Correlation Deleted
CUSDIF1 10.0405 5.4733 .2437 .7454
CUSDIF2 9.7432 5.0176 .5047 .3293
CUSDIF3 9.6486 5.3754 .4849 .3722
Reliability Coefficients
N of Cases = 111.0 N of Items = 3
Alpha = .5878
The scale is not reliable. We can delete item 1 to increase the Cronbach’s alpha, which will
increase to .7454.

CHAPTER 12
Quantitative Data Analysis Part 2:
Hypothesis Testing
Instructional Goals
1. To make students understand and apply the different types of analyses and tests to get
at the inferential information they need.
2. To help them interpret the results they obtain from data analyses.
1. When we collect data on the effects of treatment in experimental designs, which
statistical test would be most appropriate to test the treatment effects?
We would use a t‐test with paired samples to determine the treatment effects.
2. A tax consultant wonders whether he should be more selective about the class of
clients he serves so as to maximize his income. He usually deals with four categories of
clients: the very rich, rich, upper middle class, and middle class. He has information on
each and every client served, the taxes paid by them, and how much he has charged
them. Since many factors relating to the clients vary (number of dependents, business
deductibles, etc.), irrespective of the category they belong to, he would like an
appropriate analysis to be done to see which among the four categories of clientele he
should choose to serve in the future. What kind of analysis should be done in the
above case and why?
Undergraduate Students can be exposed to a simple ANOVA design.
An ANOVA will help the tax consultant to make the decision, since the differences
among the four groups on the dependent variable, income, can be obtained through
this analysis. A further test such as the Duncan Multiple Range test, will determine
which among the four groups provides the most income.

www.wileyeurope.com/college/sekaran
106
More advanced students can take this approach.

Since there are four categories of clients – very rich, rich, upper middle class, and
middle class – and the tax consultant wants to see how the clients can be differentiated
on various independent variables which are on a continuous scale, in order to increase
his income, he could use a MANOVA. MANOVA deals with multiple independent
variables simultaneously and differentiates the groups on the critical factors.
A discriminant analysis will indicate which of the entire set of variables are
maximally discriminated for the four groups. Action can then be taken based on the
results.

107
CHAPTER 13
QUALITATIVE DATA ANALYSIS
Instructional Goals
1. To make qualitative data analysis come alive and become meaningful to students
through examples that they can relate to.
2. To explain that qualitative data can be very rich and to stress the importance of
reducing qualitative data to key ideas and concepts.
3. To get students to analyze data using a structured approach.
4. To stress the importance of reliability and validity in qualitative data analysis.
1. What is qualitative data? How do qualitative data differ from quantitative data?
Qualitative data is data in the form of words. Quantitative data is data in the form of
numbers.
2. What are the main steps in qualitative data analysis?
According to Miles and Huberman, there are generally three steps in qualitative data
analysis: data reduction, data display, and the drawing of conclusions.
3. Define reliability and validity in the context of qualitative research.
Reliability and validity have a slightly different meaning in qualitative research in

comparison to quantitative research. Reliability in qualitative data analysis includes
category and interjudge reliability. Category reliability “depends on the analyst’s ability
to formulate categories and present to competent judges definitions of the categories
so they will agree on which items of a certain population belong in a category and which

do not.” (Kassarjian, 1977, p. 14). Interjudge reliability can be defined as the degree of
consistency between coders processing the same data
109
(Kassarjian 1977).
Validity refers to the extent to which the research results 1) accurately represent the
collected data (internal validity) and 2) can be generalized or transferred to other
contexts or settings (external validity).
4. How can you assess the reliability and validity of qualitative research?
A commonly used measure of interjudge reliability is the percentage of coding

agreements out of the total number of coding decisions. As a general guideline,
agreement rates at or above 80% are considered to be satisfactory.
Two methods that have been developed to achieve validity in qualitative research are
discussed next.
‐ Supporting generalizations by counts of events. This can address common concerns
about the reporting of qualitative data; that anecdotes supporting the researcher’s
theory have been selected, or that too much attention has been paid to a small number
of events, at the expense of more common ones.
‐ Ensuring representativeness of cases and the inclusion of deviant cases (cases that
may contradict your theory). The selection of deviant cases provides a strong test of
your theory.
5. What is grounded theory?
Categorization is the process of organizing, arranging, and classifying coding units.

Codes and categories can be developed both inductively and deductively. In situations
where there is no theory available, you must generate codes and categories inductively
from the data. In its extreme form, this is what has been called grounded theory.
6. How does narrative analysis differ from content analysis?
Narrative analysis is an approach that aims to elicit and scrutinize the stories we tell
about ourselves and their implications for our lives. Narrative data are often collected
via interviews. These interviews are designed to encourage the participant to describe a
certain incident in the context of his or her life history. Like this, narrative analysis
differs from other content analysis; it is focused on a process or temporal order, for

instance by eliciting information about the antecedents and consequences of a certain
incident in order to relate this incident to other incidents.
110


C 5 T R P: E O R D: Instructional Goals

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

C 5 T R P: E O R D: Instructional Goals

Uploaded by

Copyright:

Available Formats

CHAPTER 5

THE RESEARCH PROCESS: ELEMENTS OF

© 2009 John Wiley & Sons Ltd.

© 2009 John Wiley & Sons Ltd.

© 2009 John Wiley & Sons Ltd.

Because it is difficult (and probably unethical) to induce specific emotions in an

© 2009 John Wiley & Sons Ltd.

© 2009 John Wiley & Sons Ltd.

6. A production manager wants to see if, by providing enhanced automation technology,

© 2009 John Wiley & Sons Ltd.

© 2009 John Wiley & Sons Ltd.

© 2009 John Wiley & Sons Ltd.

© 2009 John Wiley & Sons Ltd.

© 2009 John Wiley & Sons Ltd.

© 2009 John Wiley & Sons Ltd.

© 2009 John Wiley & Sons Ltd.

© 2009 John Wiley & Sons Ltd.

JOB STATUS NUMBER OF YEARS WORKED IN THE SHIFT CURRENTLY

2. Your Opinions regarding the following.

STRONGLY AGREE 1 AGREE 2 NEITHER DISAGREE 4 STRONGLY

1. I will need additional training to 1 2 3 4 5 work in the

© 2009 John Wiley & Sons Ltd.

Items 4 and 7 measure opinion about their future jobs

© 2009 John Wiley & Sons Ltd.

Additional Exercises in Questionnaire Design

© 2009 John Wiley & Sons Ltd.

© 2009 John Wiley & Sons Ltd.

In‐Class Exercises on Sampling Designs

© 2009 John Wiley & Sons Ltd.

Estimation of Population Characteristics from Sample Characteristics

© 2009 John Wiley & Sons Ltd.

16 2.576 1.167 

Estimating Sample Size

© 2009 John Wiley & Sons Ltd.

“Initial reactions” non‐probability sampling.

© 2009 John Wiley & Sons Ltd.

2a). Explain why cluster sampling is a probability sampling design.

© 2009 John Wiley & Sons Ltd.

© 2009 John Wiley & Sons Ltd.

© 2009 John Wiley & Sons Ltd.

© 2009 John Wiley & Sons Ltd.

© 2009 John Wiley & Sons Ltd.

Target population: German, T‐mobile, mobile phone subscribers.

© 2009 John Wiley & Sons Ltd.

Exercises on Data Interpretation

Exercise #1 On Frequency Distributions

Table 1: Frequency Distributions of Sample (n = 148)

© 2009 John Wiley & Sons Ltd.

AGE # OF YEARS IN ORG. MARITAL STATUS

Population and Sample

Exercise #2 on Means, Standard Deviations, and Other

Table 2: Means, Standard Deviations and Other Statistics

VARIABLE MEAN STD. DEV MODE MIN MAX

© 2009 John Wiley & Sons Ltd.

VARIABLE AGE # OF YRS. MARRIED STRESS JOB INVOLVEMENT PERFORMANCE

© 2009 John Wiley & Sons Ltd.

© 2009 John Wiley & Sons Ltd.

2. What does coding the data involve?

An example of an illogical response is an outlier response. An outlier is an observation

4. How would you deal with missing data?

5. What is reverse scoring and when is reverse scoring necessary?

© 2009 John Wiley & Sons Ltd.

Measures of Central Tendency