Chapter 4:

Research Design I: Sampling and Methods of

Data Collection
 Topics to be discussed
4.1 Research design
4.2 Primary and secondary data
4.3 Sources of data
4.4 Population Vs. sample survey
4.5 Sampling techniques
4.6 Methods of data collection
4.7.Editing, coding and data entry
 THE RESEARCH PROCESS

1. Selection of topic

2. Reviewing the literature

3. Development of theoretical and conceptual frameworks

4. Clarification of research question/hypothesis

5. Research design

6. Data collection

7. Data analysis

8. Drawing conclusions
 4.1.Research Design
The research design is the master plan/blueprint
specifying the methods and procedures for collecting and
analyzing the needed information.
•A detailed outline of how an investigation will take
place.
•A research design will typically include:
1. How data is to be collected,
2. What instruments will be employed,
3. How the instruments will be used and
4.The intended means for analyzing data collected.
 What type of research is going to be used and the source(s) of the data.

 For example, deciding between exploratory, experimental, or

descriptive research
 A Classification of Business Research Designs

Research Design

Exploratory Conclusive
Research Design Research Design

Descriptive Causal
Research Research

Cross-Sectional Longitudinal
Design Design

Single Cross- Multiple Cross-

Sectional Design Sectional Design
 Exploratory & Conclusive Research Differences

Exploratory Conclusive
Objective: To provide insights and To test specific hypotheses and
understanding. examine relationships.

Character- Information needed is defined Information needed is clearly

istics: only loosely. Research process defined. Research process is
is flexible and unstructured. formal and structured. Sample is
Sample is small and non- large and representative. Data
representative. Analysis of analysis is quantitative.
primary data is qualitative.

Findings Tentative. Conclusive.

/Results:

Outcome: Generally followed by further Findings used as input into

exploratory or conclusive decision making.
research.
 A Comparison of Basic Research Designs
Exploratory Descriptive Causal
Objective: Discovery of ideas Describe market Determine cause
and insights characteristics or and effect
functions relationships

Characteristics: Flexible, versatile Marked by the prior Manipulation of

formulation of one or more
specific hypotheses independent
variables
Often the front Preplanned and
end of total structured design Control of other
research design mediating
variables
Expert surveys Secondary data
Methods: Pilot surveys Surveys Experiments
Secondary data Panels
Qualitative Observation and
research other data
 Exploratory Research
What are its characteristics?

Research is flexible and unstructured

Findings tentative
Used as building block for more research
e.g. pilot surveys, secondary data, focus groups
Done on a small non-representative sample

Why is it used?

When looking for insights into the problem

To help define hypotheses and key variables
To identify alternative courses of action
 When Information needs vague
 For Establishing priorities for further research
 Exploratory Research Cont’d
What Methods Are Used?
 secondary data
 qualitative research
 focus groups
 case studies

When is it done?

 Generally initial research conducted to clarify and define the

nature of a problem

What is it’s main limitation?

Does not provide conclusive evidence Subsequent research expected

 Conclusive
Research
Design

Descriptive Causal
Research Research
 Conclusive Research
Characteristics

 inflexible, versatile
 Results conclusive
 Research formal and structured
 Used when information needs clearly defined

Why is it Used?

 to provide decision maker with the information needed to make sound decision
 Testing hypotheses and insights and examining relationships
 Descriptive Research
What is the Objective?

To describe something, e.g. Target market population characteristics

What are its characteristics?

 Problem well understood

 Tests specific hypotheses
 Formal and structured
Large representative samples
 Provides a snapshot of the market environment
 Descriptive Research
What Methods are Used?
• Surveys (primary data)
• panels
• scanner data (secondary data)

When Used?
• Often a follow-up to exploratory research
 Descriptive Research Designs (contd.)
Cross-sectional Designs
• Involve the collection of information from any given sample of
population elements only once.
• In single cross-sectional designs, there is only one sample of
respondents and information is obtained from this sample
only once.
• In multiple cross-sectional designs, there are two or more
samples of respondents, and information from each sample is
obtained only once. Often, information from different
samples is obtained at different times.
 Descriptive Research Designs (contd.)
• Longitudinal Designs
– A fixed sample of the population is measured
repeatedly on the same variables, i.e., same
respondents studied over time
– Large amounts of data can be collected but can
be expensive
– Useful for tracking changes in consumer attitudes
and behavior over time
 Causal Research
What is the objective of causal research?

 To obtain information regarding cause and effect relationships

Characteristics?

Independent variable manipulated in a relatively controlled environment

Main method is experiment
Used to understand which variables are causes (independent variables), and
which variables are the effects (dependent variables)
 4.2. Where do data come from?

• •Take a step back – if we’re starting from

scratch, how do we collect / find data?
–Secondary data
•Secondary data – data someone else has
collected
–Primary data
•Primary data – data you collect
 Secondary Data
Data gathered and recorded by someone else prior to and
for a purpose other than the current project.
Is often:
• Historical
• Already assembled
• Needs no access to subjects or respondents
 Secondary Data – Limitations
•What did you find on the frustrating side as you looked
for data on the government offices’ official websites?
•When was it collected? For how long?
–May be out of date for what you want to analyze.
–May not have been collected long enough for detecting
trends.
•Is the data set complete?
–There may be missing information on some observations
–Unless such missing information is caught and corrected
for, analysis will be biased.
 Secondary Data – Limitations

•Are there confounding problems?

–Sample selection bias?
–Source choice bias?
–In time series, did some observations drop out over time?
•Are the data consistent/reliable?
–Did variables drop out over time?
–Did variables change in definition over time?
•Is the information exactly what you need?
–In some cases, may have to use “proxy variables” – variables that
may approximate something you really wanted to measure. Are they
reliable? Is there correlation to what you actually want to measure?
Secondary Data – Advantages
•No need to reinvent the wheel.
–If someone has already found the data, take advantage of it.
•It will save you money.
–Even if you have to pay for access, often it is cheaper in
terms of money than collecting your own data.
•It will save you time.
–Primary data collection is very time consuming.
•It has great exploratory value
–Exploring research questions and formulating hypothesis to
test.
 Primary Data - Examples
1. Focus groups discussion
2. Questionnaires
3. Personal interviews
4. Experiments and observational study
 Primary Data - Limitations

•Do you have the time and money for:

–Designing your collection instrument?
–Selecting your population or sample?
–Pretesting/piloting the instrument to work out sources of bias?
–Administration of the instrument?
–Entry of data?
•Uniqueness
–May not be able to compare to other populations
•Researcher error
–Sample bias
–Other confounding factors
 Data collection choice

•in regarding data collection the first question should be

–Will the data answer my research question/address
objectives?
•To answer that
–You must first decide what your research question/objective
is.
–Then you need to decide what data/variables are needed to
scientifically answer the question
•If that data exist in secondary form, then use them to the
extent you can, keeping in mind limitations.
•But if it does not, and you are able to fund primary
collection, then it is the method of choice.
 4.3.Basics of sampling

• A sample is a “part of a whole i.e. Population to represent it”.

•Sampling helps to determine the corresponding value of the
population and plays a vital role in business research.
• Why sample ? Samples offer many benefits:
•Save costs: Less expensive to study the sample than the
population.
•Save time: Less time needed to study the sample than the
population .
•Accuracy(??): Since sampling is done with care and studies
are conducted by skilled and qualified interviewers, the results
are expected to be accurate.
 Sampling Process

1. Defining the population

2. Developing a sampling Frame
3. Determining Sample Size
4. Specifying sample Method
5. Selecting the sample
 Sampling: Step 1
• Defining the Universe
• What is your population of interest?
• To whom do you want to generalize your results?
• Can you sample the entire population?
•Universe or population is the whole mass under study.
•How to define a universe:
–What constitutes the population?
–What are the sampling units?
Sampling: Step 2
• Establishing the Sampling Frame
•A sample frame is the list of all elements in the
population (such as telephone directories,
electoral registers, club membership etc.) from
which the samples are drawn.
•A sample frame which does not fully represent an
intended population will result in frame error and
affect the degree of reliability of sample result.
Step – 3: Determination of Sample Size
•Sample size may be determined by using:
A. Subjective methods (less sophisticated methods)
1. The rule of thumb approach: eg. 5% of population
2. Conventional approach: eg. Average of sample sizes of
similar other studies;
3. Cost basis approach: The number that can be studied with
the available funds;
B. Statistical formulae (more sophisticated methods)
•Confidence interval approach.
Determining Sample Size
•What data do you need to consider
–Variance or heterogeneity of population
–The degree of acceptable error (confidence interval)
–Confidence level
–Generally, we need to make judgments on
all these variables
Determining Sample Size
•Variance or heterogeneity of population
–Previous studies? Industry expectations?
Pilot study?
 Determining Sample Size
Formulas:
Means n = (ZS/E) 2
Proportions n = Z2 pq/ E2
Percentiles n = pc (100 – pc) Z2/ E2

Z at 95% confidence = 1.96

Z at 99% confidence = 2.58
• Where n is the sample size,

• Z2 is the abscissa of the normal curve that cuts off an area α at the
tails;
• (1 – α) equals the desired confidence level, e.g., 95%);
• e is the desired level of precision,
• p = proportion of the population that is expected to have the
attribute under study, and
• q = (1- p), the proportion of the population that is expected NOT to
have that attribute
• The value for Z is found in statistical tables which contain the area
under the normal curve. e.g Z = 1.96 for 95 % level of confidence
 Sample Size (Mean) Exercise 1
•We are about to go on a recruitment drive to hire some auditors
at the entry level. We need to decide on a competitive salary
offer for these new auditors. From talking to some HR
professionals, I’ve made a rough estimate that most new hires
are getting starting salaries in the $38-42,000 range and the
average (mean) is around $39,000. The standard deviation seems
to be around $3000.
• •I want to be 95% confident about the average salary and I’m
willing to tolerate an estimate that is within $500 (plus or minus)
of the true estimate. If we’re off, we can always adjust salaries at
the end of the probation period.
• • What sample size should we use?
• Sample Size (Mean)
•Solution:
n = (ZS/E) 2
n = ((1.96*3,000)/500)^2
n= 139
 Sample Size (Proportion)

Exercise 2
•We’ve just started a new educational TV program that
teaches viewers all about research methods!
•We know from past educational TV programs that such
a program would likely capture 2 out of 10 viewers on a
typical night.
•Let’s say we want to be 99% confident that our
obtained sample proportion of viewers will differ from
the true population proportions by not more than 5%.
•What sample size do we need?
• Sample Size (Proportion)
•Solution
n = Z2 pq/ E2
n = ((2.58^2)*0.2*0.8)/(0.05)^2
n= 426
 Sample size (Percentage)
Exercise 3
• •We wish to determine the required sample
size with 95% confidence and 5% error
tolerance that the percentage of Americans
preferring the Liberal party.
• •A recent poll showed that 52% of Americans
questioned preferred the Liberals.
• •What is the required sample size?
• Sample size (Percentage)
•Solution:
n = pc (100 – pc) Z2/ E2
n = 0.52 (100 – 0.52) (1.96)2/ 0.052
n = 384
A Simplified Formula For Proportions
• Yamane (1967:886) provides a simplified
formula to calculate sample sizes.
• ASSUMPTION:
– 95% confidence level
– P = .5 ;
Where n is the sample size,
N is the population size,
e
is the level of precision.
 Sampling Vs. Non-sampling error

•Sampling error refers to differences between the sample and

the population that exist only because of the observations that
happened to be selected for the sample.
–Increasing the sample size will reduce this type of error.
•Non-sampling errors are more serious and are due to mistakes
made in the acquisition of data or due to the sample
observations being selected improperly.
Three types of non-sampling errors:
1. Errors in data acquisition,
2. Non-response errors, and
3. Selection bias.
Note: increasing the sample size will not reduce this type of error.
 Step 4: Sampling Method
Sampling Methods
1. Probability Samples
1.1. Simple Random
1.2. Cluster
1.3. Systematic
1.4. Stratified
2. Non- probability
2.1. Quota
2.2.Judgment
2.3. Convenience
2.4. Snowball
 Classification of Sampling
Methods
Sampling
Methods

Probability Non-
Samples probability

Systematic Stratified Convenience Snowball

Simple
Cluster Judgment Quota
Random
 Probability sampling

•Probability sampling – representativeness is most important

•Does the sample represent the population as a whole?
•Techniques of probability sampling get at different ways of ensuring
representativeness
•Simple random sampling – randomly pick individuals to include in
the sample
–All individuals must have an equal chance of being selected.
–As sample size increases, sample becomes more and more
representative of population.
–Sampling is generally without replacement
–Problem: can be very costly if population is large.
Choices come from a list; who makes the list?
 Probability sampling
2. Systematic random sampling – samples according to a
rule
–E.g., every fifth person is chosen
–Problems: same as simple random.
Rule must not lead to bias.
3. Stratified sampling – break the sample into various
subgroups or strata and sample from them.
–Must have good knowledge of strata
- the subjects are selected in groups or clusters rather than
randomly
 4. Cluster(area) sampling
• The problem with random sampling methods when we have
to sample a population that is disbursed across a wide
geographic region is that you will have to cover a lot of
ground geographically in order to get to each of the units you
sampled. This problem can be minimized by the cluster
random sampling.
• In cluster sampling, we follow these steps:
 divide population into clusters (usually along geographic
boundaries)
 randomly sample clusters
 measure all units within sampled clusters
Advantages
• Significant cost gain.
• Easier and more practical method which facilitates the field
work.
Disadvantages
• Probability and the representativeness of the sample are
sometimes affected, if the number of the cluster is very large.
• The results obtained are likely to be less accurate if the
number of sampling units in each cluster is not approximately
the same.
 Probability sampling
One-stage cluster sampling –
• Select clusters randomly
• sample all members of the cluster
•Two-stage cluster sampling –
• Select clusters randomly
• random sampling within the clusters
•Effectiveness of cluster sampling
–Much more efficient; less costly
–Not quite as effective as random sampling
Multi-stage sampling – random sampling in
stages
•E.g., voting in Addis Ababa
–Stage one – randomly choose 10 Kebelles in
Addis Ababa
–Stage two – randomly choose households within
those Kebelles.
–Stage three – randomly choose individuals within
those households.
•Can have many stages.
•Especially useful if the population is very large.
 Non-probability sampling

•Qualitative researchers are not as concerned

about representativeness
–Relevance to the research topic
–Importance of context
•Sample size does not have to be determined in
advance.
–Selection of cases gradually over time
•Important: many statistics assume random
sampling
 Non-probability sampling
Types of non-probability sampling
1. Convenience sampling (haphazard, accidental)
– sample whoever is available.
•Used by both quantitative and qualitative
researchers
•Problems
–no representativeness
–It is haphazard, can be very biased
–Not random
2. Quota sampling - quotas for certain types of people or
organizations are selected as sample
–Interviewers are required to find cases with particular characteristics
–E.g., certain number of samples from each regions, age groups,
etc.
–Like non random version of stratified
•Pros
–Better than convenience; introduce some diversity
•Cons
–Theoretical quotas must be accurate to be useful.
–It is nonrandom sampling
3. Purposive sampling - Use judgment and deliberate
effort to pick individuals who meet a specific criteria.
•Especially good for exploratory or field research.
•Appropriate for at least 3 situations.
1. Select cases that are especially informative.
2. Desired population for the study is rare or very
difficult to locate.
3. Case studies analysis – find important individuals and
study them in depth.
4. Snowball sampling – an individual or
group of individuals are sampled. They
provide other sources to be sampled.
–Sampling snowballs into a large selection.
–Chain sampling
•Useful for hard to identify groups.
–E.g., study of criminal organizations
•May lead to biased sample
 Choosing probability vs. non-probability sampling
Probability Evaluation Criteria Non-probability
sampling Sampling
Conclusive Nature of Research Exploratory

Larger sampling errors Relative Magnitude

Sampling Vs Non sampling Larger non-sampling error
errors
High Population variability Low
[Heterogeneous ] [Homogeneous]
Favorable Statistical Considerations Unfavorable

Sophistication Needed Low

High
Time Relatively shorter
Relatively Longer
High Budget Needed Low
 Designing a Questionnaire

Is a tool for collecting information to describe,

compare, or explain an event or situation, as
well as, knowledge, attitudes, behaviors, and/or
sociodemographic characteristics on a
particular target group
• Questionnaire design is a systematic process in
which the researcher contemplates various
question formats, considers a number of factors
characterizing the survey at hand, ultimately
words the various questions very carefully, and
organizes the questionnaire’s layout.
•A survey is only as good as the questions it asks
•The underlying research problem and objectives must be
clearly understood by the researchers
•Using common sense, good grammar and imagination
alone cannot guarantee producing a high quality
questionnaire
•Good questionnaires require experience and a knowledge
of the respondents’ level of awareness, education and
understanding in order to maximize feedback and
minimize respondent and other potential sources of error.
Typical Issues in Questionnaire Design
•What should the respondents be asked?
•How should each question be phrased?
•In what sequence should the questions be arranged?
•What is the best questionnaire layout for the research
problem in question?
•What communication medium should be utilized
(personal interview, telephone interview etc.)
•Should the questionnaire be pretested?
•How should the questionnaire be pretested?
•Does the questionnaire require a revision?
 Questionnaire Relevancy & Accuracy

•Questionnaires must fulfil the two fundamental

criteria of relevance and accuracy
•Relevancy means that no unnecessary
information is obtained from the questionnaire
and that all the information that is needed for the
purpose of the research is collected, and that no
important information is omitted
•Accuracy means that the questions are worded in
a manner which ensures the collection of
correct information from respondents
 Phrasing Questionnaires

•Open-Ended Questions – These require the respondent to

answer the question in his or her own words
•Open-ended questions allow the respondent to ask or probe
the respondent further if he/she feels that clarification or
additional information, is needed.
•Good for exploratory research.
•Disadvantage:
–Difficulties of analyzing the data and in categorizing and
summarizing answers because of the unique responses.
–Also, there is the possibility of bias and bias caused by
the different education levels of the respondents
 Phrasing Questionnaires….

•Fixed-Alternative Questions –require the respondent

to choose, among a limited number of alternative given.
– Advantages include requiring less time, less skill and
ease of answer for respondents.
–Disadvantages are that the researcher may be basing
the response structure on assumptions which may be
incorrect, and that there may be additional responses
which are not indicated.
–Also, care must be taken to ensure that response
categories are mutually exclusive and do not overlap
 Phrasing Questionnaires…

•Some important guidelines for phrasing questionnaires

include:
–Avoidance of Complexity / Use Simple Language –
Respondents may have quite different educational
backgrounds and conversational levels (from quiet and shy
to articulate) which have to be taken into consideration
when designing a questionnaire which is intended for
general circulation.
Technical jargon should only be considered when the
group of respondents is homogenous and familiar with the
jargon
 Phrasing Questionnaires…
•Avoidance of “Leading” and “Loaded” Questions
•Leading means implying certain answers which respondents
are pushed towards;
•Loading pushes the respondent towards a socially desirable or
politically correct answer or puts a question which is
emotionally charged
•Counter-biasing Statements
•Respondents are presented with an introductory statement
or preface to a question or section of questions that may help
reduce the respondents reluctance to answer the question(s).
 Phrasing Questionnaires…

•Avoidance of Ambiguity – Respondents

must not be presented with terms and words
that are vague or general or which cause
confusion and misunderstanding (example:
How “frequent” do you go to the internet cafe?
• •Avoidance of Double-Barreled Questions –
Questions which address two or more issues
simultaneously .
 Phrasing Questionnaires…

•Avoidance of Assumptions – Respondents

are presented with assumptions which cause
them to give biased responses
• •Avoidance of Burdensome (and Memory
Taxing) Questions – Respondents are asked to
give answers to questions which they may not
be in a position to give due to forgetfulness
Sequencing Questions in Questionnaires
•Asking questions requiring respondents to give
personal details at the beginning of the questionnaire
is generally not recommended unless it is relevant to
the objectives
•Usually, researchers prefer to ask general questions
from respondents before moving on to specific
questions (funnel technique)
•Often, it is advisable to produce multiple versions of
a questionnaire in which fixed-alternative responses
are shuffled about.
 Layout of Questionnaires
•Don‘t overcrowd the questionnaires
•Use margins of adequate size
•Use white space if needed to separate sections of the
questionnaire
•Keep questionnaires as brief as possible
•Use good quality paper
•Ensure that the title and subtitles of the questionnaire are
carefully phrased and captures the respondents attention
•Include a privacy and confidentiality clause
 Pretesting Questionnaires

•Pretesting is a very useful method for determining

whether respondents have any difficulty/ambiguity
in understanding the questions
•The respondents involved in a pretest should be
similar in essence to the target respondents
•Personal interviewers are often used for pretesting
in order to ascertain why questions appear
ambiguous or confusing etc. to respondents, and to
solicit their comments
 Pretesting Questionnaires …

•Pretesting provide answers to important questions

for the business researcher, such as:
–Can the questionnaire format be followed by the
respondents?
–Does the questionnaire flow naturally and
conversationally?
–Can respondents answer the questions easily?
–Which alternative forms of questions work best?
 Administering a written questionnaires

•Overall purpose
–When need to quickly and/or easily get lots of information from
people in a non threatening way
•Advantages
–Can be completed anonymously
–Inexpensive to administer
–Easy to compare and analyze
–Administer to many people
–Many sample questionnaires already exist
•Challenges
–Might not get careful feedback
–Wording can bias client's responses
–Doesn't get full story
 Points to consider in phrasing questions

Clarity – asking for frequency:

•Rather than saying: You may ask

–Very Often •Every Day or More
–Often •2-6 Times a Week
–Sometimes •About Once a Week
–Rarely •About Once a Month
–Never Never
 Questionnaires....

•Avoid Double Negatives

–It is not the case that I have never cheated on
my tax returns.
–Never should one not help others
•Avoid Objectionable or Irrelevant Questions
unless directly related to the topic
–How old are you?
–Have you answered each question truthfully?
 Questionnaires….

•Reverse Score to Reduce Response Bias

–I am often sad.
•Strongly Agree
•Agree
•Undecided
•Disagree
•Strongly Disagree

–I often feel happy.

•Strongly Agree
•Agree
•Undecided
•Disagree
•Strongly Disagree
 Questionnaires….

•Exhaustive & mutually exclusive categories

–What is your age?
•under 10
•10-20
•20-30
•30-40
•40-50
–What is your marital status?
•single
•Married
•divorced
 Observation

•The researcher makes himself/herself available at the

premise of the study subject and observes events
•There is high degree of reliability on the data collected
•Costly and time taking
•Impractical for large sample size study
•To gather accurate information about how a program
actually operates, particularly about processes
•View operations as they are actually occurring
•Can adapt to events as they occur
 Observation, cont’ed
•Disadvantage:
–can be difficult to interpret seen behaviors
–can be complex to categorize observations
–can influence behaviors of program participants
–can be expensive
• Note: to make the best use of observation, it is better
to have a pre-set categorization of possible observable
phenomena or criteria of acceptance or rejection. To
do so, preset observation checklist is indispensable.
 Interview

•Overall Purpose
–When a researcher wants to fully understand someone's impressions
or experiences, or learn more about their answers to questionnaires
•Advantages
–Get full range and depth of information
–Can be flexible with interviewees
•Disadvantages
–Can take much time
–Can be hard to analyze and compare
–Can be costly
–Interviewer can bias client's responses
 Preparation for Interview

1.Choose a setting with little distraction

2.Explain the purpose of the interview
3.Address terms of confidentiality
4.Explain the format of the interview
5.Indicate how long the interview usually takes
6.Tell them how to get in touch with you later if they want to
7.Ask them if they have any questions
8.Don't count on your memory to recall their answers record
or note down points
 Types of Interviews

1. Informal,
conversational interview
–The interviewer “goes with the flow”

2. General interview guide approach

–Ensure that the same general area of information are collected from all
interviewees

3. Standardized, open-ended interview

–Same open-ended questions are asked to all interviewees
–This approach facilitates faster interviews that can be more easily analyzed and
compared

4. Closed, fixed-response interview

–all interviewees are asked the same questions and asked to choose answers from
among the same set of alternatives
 Sequence of Questions

1.Get the respondents involved in the interview as soon as

possible.
2.Before asking about controversial matters (such as feelings
and conclusions), first ask about some facts
3. intermingle fact-based questions throughout the interview
4.Ask questions about the present before questions about the
past or future
5.The last questions might be to allow respondents to provide
any other information they prefer to add and their impressions
of the interview.
 Carrying Out Interview

•Ask one question at a time

•Attempt to remain as neutral as possible
•Encourage responses with occasional nods of the
head
•Be careful about the appearance when note taking
•Provide transition between major topics
•Don't lose control of the interview
•Write down any observations made during the
interview right after the interview
End of Chapter 4