ACFN 3111-

Research Methods in Accounting & Finance

Chapter V
The Sources and Collection of Data

Highly recommended text :

Zikmund. Business Research Methods
Content of the Lecture

1. Introduction
2. Data Collection methods
 Types and sources of Data
 Methods of primary Data collection
 Methods of secondary Data collection
3. Data management issues
4. Sources and Techniques of Data Collection
5. Questionnaire Design
Introduction
 Data are the foundations of research.

 The quality of any investigation heavily depends on the

quality of information or data used.
 So, proper data collection, retention, and sharing are
vital to the research enterprise.

 What is Data?
 Data refers to any group of facts, measurements, or
observations used to make inference about the problem of
investigation.
Introduction
 It can range from material created in a laboratory, to
information obtained in economic research, such as a filled-
out questionnaire, video and audio recordings, or
photographs, etc.

 We need to design strategies that would enable us to

create meaningful and unbiased data,
 that will not waste resources, and
 that will appropriately protect human and animal subjects.
Introduction
 When selecting data we need to be clear on:
 The data type (e.g., nominal, ordinal, interval or ratio
measures).
 Samples ("frames") and sample size, instruments.
 Methodologies for collecting data, etc.

 And, to ensure high quality output, data gathering

should be well planned.
Introduction
 Hence, we need to plan and define properly:
 The objective of the data collection exercise
 The kind of data (information) to be collected
 The source of the data
 The methods to be used to collect data

 This helps us to determine the statistical techniques to be

employed.
 It also helps us to avoid gathering of irrelevant
information.
Data Collection Methods
 Data collection is the process of gathering and measuring
information on the variables of interest.

 The data source could be:

 an area from where we can collect the data (i.e.
geographic location),
 persons to be interviewed,
 an entity or organization,
 discussions to be observed, etc.).
Data Collection Methods
 Although data collection methods may vary by discipline
and data types, the emphasis on ensuring accurate
collection remains the same.

 Some of the consequences from improperly collected data

may include:
 Inability to repeat and validate the study.
 Distorted and inaccurate findings.
 Wasted resources.
 Misleading other researchers.
 Harm to human participants and animal subjects .
Data Collection Methods
 More importantly, poor data collection may cause harm
to society when the results of the flawed research are
used to support public policy recommendations.
Data management issues
 Storage and Protection
 Data must be stored securely both during a research
project and after it ends.

 Risks like fire, water or other environmental damage,

or common technical failures like hard disk crashes,
must be considered.

 So, make backup copies of the data periodically and

store the copies in a secure location.
Data management issues…
 Confidentiality
 refers to limiting information access and disclosure to
authorized users- preventing access by or disclosures
to unauthorized persons.

 So you need to decide

 who can handle which portion of data,
 at what point during the project,
 for what purpose, and so on
Data management issues…
 Integrity
 Integrity refers to the trustworthiness of the
information.

 Researchers need to have sufficient skills to ensure the

integrity of their data collection efforts.
 For instance, data collection requires a rigorous and
detailed recruitment and training plan for data
collectors.

 So, data should not be modified inappropriately,

whether accidentally or deliberately.
Sources and Techniques of
Data collection
Sources and Techniques of Data collection
 Sources of data can be grouped into
 primary and
 secondary sources of data

Secondary Sources of data

 Secondary sources are those, which have been collected by
other individuals or agencies.

 i.e. refers to already existing information which have

previously been collected and reported by some
individual or organization for their own purposes.
Sources and Techniques of Data collection…

 As much as possible secondary data should always be

considered first, if available.

 WHY REINVENT THE WHEEL IF THE DATA ALREADY

EXISTS!
Sources and Techniques of Data Collection…
 But, when dealing with secondary data you should ask:

 Is the owner of the data making them available to

you?
 Is it free of charge? If not, how will you pay?
 Are the data in a format that you can work with? etc.

 A description of the sampling technique, i.e., how the

sample was collected is also necessary, etc.
Sources and Techniques of Data Collection…
Sources of Secondary Data
 Secondary data may be acquired from various sources:

 Documents (reports of various kinds, books,

periodicals, reference books (encyclopedia),
university publications (thesis, dissertations, etc.),
policy documents, statistical compilations,
proceedings, personal documents (historical
documents, Data archives, etc.
 The Internet
Sources and Techniques of Data Collection…
Advantages of Secondary data
 Can be found more quickly and cheaply.
 Most researches on past events or distant places have
to rely on secondary data sources.

Limitations
 Authenticity:
 genuine?
 credible?
 representative?
Sources and techniques of Data Collection…
 Completeness???
 The information often does not meet one’s specific
needs.
 Definitions might differ, units of measurements may
be different and different time periods may be
involved.

 Data could also be out of date.

Sources and Techniques of Data Collection…
Primary Sources of Data
 These are data collected by the investigator (immediate
user) for the 1st time.

 Two approaches to primary data collection:

 the qualitative approach and

 the quantitative approach
Primary Data Collection

• Qualitative vs. Quantitative

I. Qualitative Data Collection
Qualitative data collection approaches
 Qualitative data can be acquired from:
 case studies,
 Participant Observation
 Rapid appraisal methods,
 focus group discussions and
 key informant interviews, etc.

i) Case studies: involves a detailed investigation of a

particular case.
Qualitative Data Collection…
 (ii) Participant Observation: this is when a researcher
attempts to observe in some way.

 It can be a good way of getting below the surface information

and help to reveal complex social processes.
 The researcher can play an overt or covert role.
 Example: the observation of consumer behaviour in
supermarkets.
Qualitative Data Collection…
iii) Participatory Rapid Appraisal (PRA)
 PRA is a systematic expert observation usually by a
multidisciplinary team.

 The PRA method:

 takes only a short time to complete,
 tends to be relatively cheap, and
 make use of more 'informal' data collection
procedures.
Qualitative Data Collection…
iv) Focus group discussions
 A FGD is a group discussion guided by a facilitator, during
which group members talk freely.

 The researcher creates a relaxed atmosphere and records in

some way what is being said (e.g. by use of a tape-recorder,
video, note-taker, etc.).
 Qualitative Data Collection…
 Mostly used to gather opinions, from a selected group of
people on a particular and pre-determined topic, e.g.
consumer topic; political topic etc.

 About 10 people would be engaged in a discussion on the

topic in an informal setting.

 The researcher encourages free discussion, but is ready to

intervene if necessary to resolve group problems.
 The group of individuals are expected to have experience
or opinion on the topic.
Qualitative Data Collection…
 Its purpose is to obtain in-depth information on concepts,
perceptions and ideas of a group.

 Focus groups can be a useful way of finding out what the

main issues and concerns of any group are.

 It is more than a question-answer interaction.

 group members discuss the topic and interact among
themselves with guidance from the facilitator .
Qualitative Data Collection…
Why use focus groups?
 The main purpose of a focus group research is to draw upon
group’s attitudes, feelings, beliefs, experiences and reactions.

 attitudes, feelings and beliefs may likely be revealed via

interaction in social gatherings.

 Compared to individual interviews, which aim to obtain

individual attitudes, beliefs and feelings, focus groups elicit a
multiplicity of views.
Qualitative Data Collection…
Strengths and weakness of FGDs

 It provides valuable information in a short period of time

and at relatively low cost if the groups have been well
chosen, in terms of composition and number.

 But, their use as a research tool is limited when it comes

to measuring the results objectively.
Qualitative Data Collection…
 FGD should not be used for quantitative purposes, such as
the testing of hypotheses or the generalization of findings
for larger areas

 Which require more elaborate surveys.

 In addition, the logistical complexity of focus group

research is frequently cited as a deterrent .
 Qualitative Data Collection…
 In addition, it may be risky to use FGDs as a single tool
because in:

 group discussions, people tend to center their opinions on

the most common ones.
 In case of very sensitive topics group members may
hesitate to express their feelings and experiences freely.

 Therefore, it is advisable to combine FGDs with other

methods (in-depth interviews).
Qualitative Data Collection…
v) Key Informant Interview
 An interviewing process with opinion leaders such as
elected officials, government officials, and business leaders,
etc.

 This technique is particularly useful for:

 Raising community awareness about socio-economic
issues
 Learning minority viewpoints
 Gaining a deeper understanding of opinions and
perceptions, etc.
 Qualitative Data Collection…
v) Triangulation
 refers to the use of more than one approach to the
investigation of a research question in order to enhance
confidence in the findings.

 Why use triangulation

 The purpose of triangulation is to obtain confirmation of
findings through convergence of different perspectives.

 By combining multiple methods, and empirical

materials, researchers can overcome the weakness or
biases and problems that are associated with a single
method.
Qualitative Data Collection…
Taxonomy of triangulation
1. Data triangulation: Involves gathering data at different
times and situations, from different subjects using different
sampling techniques.
 Example: Using time series data.

2. Investigator triangulation: involves using more than one

researcher to collect and analyze the data.
 Example: organizing scientific experiments to replicate
each other’s work.
Qualitative Data Collection…
3. Theoretical triangulation: making explicit references to more than
one theoretical tradition to analyze data.
4. Methodological triangulation: combination of different research
methods or different varieties of the same method.

 Within method triangulation involves making use of

different varieties of the same method.
 Example: making use of alternative econometric
estimators
 Between method triangulation involves making use of
different methods.
 Example: Using ‘quantitative’ and ‘qualitative’ methods.
II. Quantitative Data Collection
Quantitative Primary Data Collection Methods

 This method involves the collection of quantified data that

can be subjected to statistical treatment.

 Primary data may be collected through:

 Direct personal observation method, or
 Survey or questioning other persons,
 From a literature search, or
 by combining them.
Quantitative Data Collection…
The Observation Method
 Observation includes all kinds of monitoring behavioral and
non-behavioral activities.

 The investigator will make the observation personally by

being guided by some outlined questions.

 i.e. the investigator does not make inquiries but notes down
the observation himself/herself
 (e.g. evaluation of the class room teaching-learning
process by investigators).
Quantitative Data Collection…
 Advantages
 It is less demanding and has less bias.

 One can collect data at the time it occurs and need not
depend on reports by others.

 with this method one can capture the whole event as it

occurs.
Quantitative Data Collection…
Weakness of the Method
 The observer normally must be at the scene of the event
when it takes place.
 But it is often difficult or impossible to predict when and
where an event might occur.

 It is also a slow and expensive process.

 Its most reliable results are restricted to data that can be
determined by an open or surface indicator.
 Difficult to learn about past events and to gather
information on intensions, attitudes, opinions and
preferences.
Quantitative Data Collection…
The Survey Method: the most commonly used method in
economics.
 To survey is to ask people questions in a questionnaire.

 In a survey, a trained interviewer asks the questions and

records responses on a specially designed form.

 The form contains all the questions which would extract

information from a respondent.
Quantitative Data Collection…
Strength of the Survey Method
 It is a versatile or flexible method - capable of many
different uses.
 More efficient and economical than observations -surveying
using telephone or mail is less expensive.

 If planned correctly, a survey makes it easy for people to

participate.

 Same questions can be asked in several ways to double-

check for accuracy and consistency.
Quantitative Data Collection…
Weakness of the Method
 The quality of information depends heavily on the ability
and willingness of the respondents.

 A respondent may interpret questions or concept

differently from what was intended by the researcher.

 A respondent may deliberately mislead the researcher

by giving false information.
Quantitative Data Collection…
 People surveyed may not answer all questions.

 Low response rate is common.

 People can and do lie.

 Can’t test knowledge with mailed surveys.

Quantitative … Survey admin
 Main modes of administration of a survey
Quantitative … Survey admin
 Main modes of administration of a survey
Quantitative … Survey admin
 Surveys could be carried out through:
 face to face personal interview
 telephone interview
 mail or e-mail, or by a combination.

 a) Personal Face to face Interview

 It is a two-way conversion where one person
interviews another person.

 interviewers ask the questions and mark the answers,

either on paper or by typing them onto a computer.
 Quantitative … Survey admin
 Advantages of face to face Interview:
 enhances respondent participation,
 guiding the questioning process,
 answering the respondent’s questions,
 clarifying the meaning of questions and responses,
 probing, clarifying and motivating the respondent to
complete the questionnaire,
 making sure that the questionnaire is answered in the pre-
defined sequence and by the respondent himself or herself.
 In addition, visual aids are possible in face-to-face
interviews but not in telephone surveys.
Quantitative … Survey admin
 The depth and detail of the information that can be secured
far exceeds the information secured from telephone or mail
surveys.

 Interviewers can probe additional questions, gather

supplemental information.

 Interviewers can make adjustments to the language of the

interview because they can observe the problems and effects
with which the interviewer is faced.
Quantitative … Survey admin
Limitations of the Method
 The method is an expensive enterprise – (e.g., locating
respondents) – US$50-80.
 Hence, personal interviews are generally used only
when subjects are not likely to respond to other
methods.

 susceptible to interviewers’ bias/mistakes

 Interviewer may also be reluctant to visit unfamiliar

places.
Quantitative … Survey admin
b) Telephone Interview
 Telephone can be a helpful medium of communication in
setting up interviews.
 telephone numbers are picked, using some sampling
technique, from a telephone directory.
 Then the survey will be administered by calling and
interviewing those who are selected in the sample via
telephone.
 Telephone surveys are the fastest method of gathering
information from a relatively large sample - generally last
less than ten minutes.
Quantitative … Survey admin
Strengths of this method
 Moderate travel and administrative costs
 Faster completion of study
 Responses can be directly entered on to the computer

 Speed of administration; no geographic limitations; etc.

Quantitative … Survey admin
Limitations of this method
 Respondents must be available by phone.

 The length of the interview period is short.

 limited in length and scope.

 those interviewed by phone find the experience to be

less rewarding than a personal interview.
Quantitative … Survey admin
C) Interviewing by mail (Solicited responses)
 Self-administrated questionnaires may be used in surveys.

 Questionnaires will be sent to respondents through their

email or postal address.

 Respondents are expected to fill out the questionnaires and

send them back the same way.

 Self-administered surveys do not need interviewers.

 Respondents mark, write or type the answers themselves.
 Quantitative … Survey admin
 They are ideal for large sample sizes, or when the sample
comes from a wide geographic area.
 Questionnaires must be easily understandable since there is
no possibility for respondents to ask how to mark answers or
what a question might mean.
Advantages
 Lower cost than personal interview
 Persons who might otherwise be inaccessible can be
contacted (major corporate executives)
 Less interviewer bias
 better protects privacy/anonymity
Quantitative … Survey admin
Disadvantages
 Non response error is high
 Large amount of information may not be acquired
 Literacy of the respondents is necessary, and the language
of the respondents must be known in advance.
 We have no control who responds or whom the respondent
consults.
 We have no control for the bias originating from the self-
selection process (respondents choose freely whether they
respond or not).
Quantitative … Survey admin
d) Online Surveys (E-mail / Internet)
 Internet surveys allow researchers to reach a large sample
more quickly.
 Sample size requirements can be met easily.

 Internet surveys are relatively new and little is known about

the effect of sampling bias in internet surveys.

Advantages:
 Very inexpensive -saves inputting costs as well
 Respondents feel privacy
Quantitative … Survey admin
Disadvantages:
 A sample of Internet users is representative only of Internet
users, who tend to be younger, better educated, and more
affluent than the general population.
 Very biased toward wealthy and the young.

 The demographic profile of the internet user does not always

represent the general population.
 Therefore, before doing an e-mail or internet survey,
carefully consider the effect that this bias might have on
the results.
Survey - Questionnaire
Design
Questionnaire Design
 The instrument design begins by drafting specific
measurement questions in the form of a questionnaire.

 A questionnaire is a research instrument consisting of a

set of questions intended to capture responses from
respondents.

 Questionnaires consist of questions.

 A list of questions each with a range of answers.

 A well formed questionnaire is key to good research

Questionnaire Design…

 The construction of a questionnaire requires a marriage of

art and science to achieve two primary goals:

 address the survey objectives,

 smooth conversational flow.

 But, why use questionnaires?

Questionnaire Design…
 Questionnaires are easy to analyze.
 Data entry and tabulation can be easily done with
many computer software packages.

 Questionnaires are familiar to most people.

 Nearly everyone has had some experience completing
questionnaires and they generally do not make people
apprehensive.

 Questionnaires reduce bias.

 There is uniform question presentation.
Questionnaire Design…
 Questions may be unstructured (respondents provide a
response in their own words) or structured (respondents
select an answer from a given set of choices).
Questionnaire Design…

 Open-ended response questions

 Questions that pose some problem and ask respondents
to answer in their own words.

 Fixed-alternative questions
 Questions in which respondents are given specific,
limited-alternative responses and asked to choose the one
closest to their own viewpoint.
Questionnaire Design…

 Structured questionnaires can be designed as:

 Dichotomous: having only two alternative responses
E.g. are you married? Yes No
 Multiple choice: having numerous alternative responses
 E.g. what is your marital status?
 Married Single
 Widowed
Divorced
Questionnaire Design…
The main Components of a questionnaire
 Identification data: respondent’s name, address, time and
date of interview, code of interviewer, etc.

 Covering letter: brief purpose of the survey, who is doing

it, time involved, ethical statements, etc.

 Instruction: clear and concise instructions on how to

complete the questionnaire.

 Information sought: the actual information needed -

major portion of the questionnaire
Questionnaire Design…
 The following can have important implications on the survey
response.
 Question content

 Question type

 Question wording

 Response form

 Question sequence

 Questionnaire layout
Questionnaire… 1. Question Content

 Questions should be
 Relevant (about issues respondents have knowledge of)

 Reliable – same response by same individual and

different people should understand the question the
same way
 discriminating – should capture sufficient variation

 increasing response rates – sensitive questions and

poor survey administration can reduce response rates
 Simple and short
Questionnaire… 1. Question Content

 Responses obtained in survey research are sensitive to the

types of questions asked.

 For each question in your survey, you should ask yourself

how well it addresses the content you are trying to get at.

 So you need to examine each question to see if you

need to ask it at all and,
 if you need to ask it at the level of detail you currently
have.
Questionnaire… 1. Question Content

 Question content usually depends on the respondent’s

 ability, and
 willingness to answer the question accurately.

a) Respondents’ ability: Do respondents have the needed

information?

 Check whether the respondent is likely to have the

necessary information to be able to answer the question.
 Questions that overtax the respondent’s recall
ability may not be appropriate.
Questionnaire… 1. Question Content

b) Willingness of respondent to answer

 Whether the respondent will have any difficulty answering
the question truthfully.

 Respondents may be unwilling to share it because:

 The situation is not appropriate for disclosing the
information – embarrassing or sensitive
 Disclosure of information is a potential threat to the
respondent
 They consider the topic to be irrelevant and
uninteresting
Questionnaire… 1. Question Content

 If there is some reason why they may not, consider

rewording the question to secure more complete and
truthful information.

 For instance: some people are sensitive about answering

questions about their exact age or income.

 In this case, you might give them response brackets to

choose from (e.g., between 30 and 40 years old, between
$50,000 and $100,000 annual income).
Questionnaire… 1. Question Content

 Use indirect statements i.e., “other people”

 You might get reasonable estimates if you ask the
respondent how much money "people with similar
qualification" typically get.

 Apply appropriate questioning sequences that will lead a

respondent from „safe“ question gradually to those that
are more sensitive.

 Begin with non-threatening and interesting questions.

Questionnaire… 1. Question Content

 Provide incentives as a motivation.

 What does the respondent get for completing your

questionnaire?
 Altruism may not be an effective motivator.

 Attaching an incentive to the questionnaire works well.

(of course, you have to consider the ethical implications
of such a practice!).
Questionnaire… 2. Question type

 Types of questions depend on the research question.

 Attributes – characteristics of respondents (e.g., age,

sex, etc.)
 Behaviour – what people do
 Beliefs – what people believe
 Knowledge – what people know
 Attitudes/opinions – what is desirable
 How much – measurements
Questionnaire… 3. Question wording

Question Wording/Language: Using Shared Vocabulary

 Getting the wording right is also a major difficulty in
writing good survey questions.

 Even slight wording differences can confuse the

respondent or lead to incorrect interpretations of the
question.
 In a survey the two parties must understand each other
and this is possible only if the vocabulary used is common
to both parties.

 So, don’t use uncommon words or abbreviations and

make items as brief as possible.
Questionnaire… 3. Question wording

 Don’t use emotionally loaded or vaguely defined words.

 For instance, if you ask a question about the "mass media," what
do you mean? The newspapers? Radio? Television?
 Avoid technical and slang terms understood only by a small subset of
the population.
 Be specific in the use of terms and concepts (e.g. government: national,
regional, local)
 loaded question - A question that suggests a socially desirable
answer or is emotionally charged. For example, a truthful answer to
the following classification question might be painful: Where did you
rank academically in your high school graduating class? □ Top
quarter □ Second quarter □ Third quarter □ Fourth quarter
Questionnaire… 3. Question wording

 Be specific when using a time period (“during last week”

instead of “recently”).
 Specify the time frame precisely.

 Make sure that facts within the question are accurate.

 Use correct grammar.

Questionnaire… 3. Question wording

 Questions should not be

 Double-barrelled – don’t ask two questions in one.

 You can often spot these kinds of problems by

looking for the conjunction "and" in your question.

 Example: are you satisfied with the university

and its computer science department?
Questionnaire… 3. Question wording

 Avoid leading or biased questions– pushing people to

answer in a certain way

 For instance: If you ask a question like: What do you see

as the benefits of a tax cut?
 you're only asking about one side of the issue.
 You might get a very different picture if you also
asked about the disadvantages of tax cuts.

 Words like usually, often, sometimes, occasionally,

seldom, etc., are ambiguous.
Questionnaire… 3. Question wording

 leading question

 A question that suggests or implies certain answers.

 E.g. Many people are using dry cleaning less because of improved
wash-and-wear clothes. How do you feel wash-and-wear clothes
have affected your use of dry cleaning facilities in the past 4 years?

□ Use less □ No change □ Use more

Questionnaire… 3. Question wording

 Ambiguity
 Introduce errors in the data
 Example – “Why did you fly to Dire Dawa on Ethiopian Airlines?”
 Most might interpret this question as was intended, but it
contains three possible questions, so the response might concern
any of these:
 I flew (rather than another mode of travel) because…
 I went to Dire Dawa because…
 I selected EAL because…
Questionnaire… 3. Question wording

 Avoid Making Assumptions

 Consider the following question: Should General Electric continue to

pay its outstanding quarterly dividends? Yes No
 This question has a built-in assumption: that people believe the
dividends paid by GE are outstanding. By answering “yes,” the
respondent implies that the program is, in fact, outstanding and that
things are fine just as they are. When a respondent answers “no,” he or
she implies that GE should discontinue the dividends. The researchers
should not place the respondent in that sort of bind by including an
implicit assumption in the question.
Questionnaire… 3. Question wording

 Unbalanced questions
 “Are you in favor of fair value?” is an unbalanced question
because it provides only one alternative. It can be reworded to “
Do you favor or not favor fair value?”, to give respondents more
than one alternative.
 Similarly the use of persuasive tone can affect the respondent’s
answers. Wording should be chosen carefully to avoid a tone that
may produce bias in responses.
Questionnaire… 3. Question wording

 Recall/memory error (Avoid Burdensome Questions That May Tax the

Respondent’s Memory)

 Respondents tend to remember what should have been done rather

than what was done.
 Telescoping is a specific type of memory error. This occurs if the
respondent reports events as occurring either earlier or later than
they actually occur. Error occurs when respondents included details
of an event which actually occurred outside the specified reference
period.
 Minimizing the recall period helps to reduce memory bias.
Questionnaire… 4. Response format

Response structure or format -

 the degree of the structure imposed on the responses.
 Structured (closed) versus unstructured (open)
formats

 Structured Response Formats:

 The respondent chooses one of several given responses.

 help the respondent to respond more easily and help the

researcher to accumulate and summarize responses
more efficiently. Preferable in large surveys.
Questionnaire… 4. Response format

 Structured Response Formats…

 They are more difficult to write than open questions

because the response choices must be known in
advance.
 They produce standardized data that can easily be
analyzed statistically.
Questionnaire… 4. Response format

 Structured Response Formats…

 The response choices of categorical questions should be

both inclusive (or exhaustive):
 Inclusive: all possible answers should be listed in the
questionnaire;
 Exclusive: no respondent belongs to more than one
category. (except in the case of multiple response
categorical questions),
Questionnaire… 4. Response format

Structured Response Formats…

Advantages
 Easier and quicker for respondents to answer
 Easier to compare the answers of different respondents
 Easier to code and statistically analyze
 reduce the variability of responses
 make fewer demands on interviewer skill, etc.

 don’t discriminate against the less talkative

Questionnaire… 4. Response format

Structured Response Formats…

Limitations

 Can suggest ideas that the respondents would not

otherwise have

 too many choices can confuse responden ts

Questionnaire… 4. Response format

 Unstructured Response Formats- These are generally a

written texts.

 If the respondent (or interviewer) writes down text as

the response, you've got an unstructured response
format.

 These can vary from short comment boxes to the

transcript of an interview.
Questionnaire… 4. Response format

 Respondents can give any answer and there are no given

alternative responses.

 They may express themselves extensively in their

own everyday language

 Responses to open questions are often difficult to compare

and interpret.
Questionnaire… 4. Response format

Advantage: Permit an unlimited number of answers

 Respondents can qualify and clarify responses
 Permit creativity, self expression, etc.

Limitations: responses may not be consistent.

 Some responses may be irrelevant
 Comparison and statistical analysis difficult.
 Articulate and highly literature respondents have an
advantage
 Requires greater amount of respondent time, thought
and effort.
Questionnaire… 5. Question sequence

Question Sequence – the order of the questions

 This can also affect the response as well as the overall data
collection activity.

 Which topics should be introduced early in the survey,

and which later?
 If you leave your most important questions until the end,
your respondents may be too tired.

 If you introduce them too early, your respondents may not

yet be ready to address the topic, especially if it is a
difficult or disturbing one.
Questionnaire… 5. Question sequence

 There are no easy answers to these problems - you have

to use your judgment.

The Opening Questions: Just as in other aspects of life, first

impressions are important in survey work.

 The first few questions you ask will determine the tone
for the survey, and can help put your respondent at ease.
 So, the opening few questions should, in general, be
easy to answer
Questionnaire… 5. Question sequence

 The first questions in a questionnaire must maintain

respondent interest and make responding easy.
 They should be easy to understand and nonthreatening .

 You should never begin your survey with sensitive or

threatening questions.

 Once respondents have been drawn into the interview,

complex or difficult-to-answer questions may follow —
 Of course, before respondent fatigue becomes an issue.
Questionnaire… 5. Question sequence

 Sensitive Questions: In much of our social research, we

have to ask respondents about difficult or uncomfortable
subjects.
 Before asking such questions, you should attempt to
develop some trust with the respondent.

 A Checklist of Considerations: There are lots of

conventions or rules-of-thumb in the survey design
business.
Questionnaire… 5. Question sequence

 Here's a checklist of some of the most important items you

can use to review your instrument:

 Start with easy, nonthreatening questions put more

difficult, threatening questions near end

 Grouping questions that are similar will make the

questionnaire easier to complete, and the respondent
will feel more comfortable.

 Ask about one topic at a time

 When switching topics, use a transition
Questionnaire… 5. Question sequence

 Questions that jump from one unrelated topic to another

are not likely to produce high response rates.

 Transitions between questions should be smooth.

 Each question should follow comfortably from the
previous question.

 present general questions before specific ones in order to

avoid response contamination.

 Reduce response set

Questionnaire… 5. Question sequence

 Finally
 Thank the respondent at the beginning for allowing you
to conduct your study

 Be sensitive to the needs of the respondent

 Be alert for any sign that the respondent is uncomfortable

 Thank the respondent at the end for participating

Questionnaire … 6. Questionnaire layout

Physical Characteristics of a Questionnaire

 Reduce respondent fatigue!
 An improperly laid out questionnaire can lead respondents to
miss questions, can confuse them.

 So, take time to design a good layout

 ease to navigate within and between sections

 ease to use the questionnaire in the field; e.g.,

questions on recto and codes on verso sides of the
questionnaire
 leave sufficient space for open-ended questions
Questionnaire … 6. Questionnaire layout

 If you put more than one question on a line some

respondents might skip the second question.
 Abbreviating questions will result in misinterpretation of
the question.

Formats for Responses

 A variety of methods are available for presenting a
series of response categories.
 Boxes
 Blank spaces
Questionnaire … 6. Questionnaire layout

 Yes/no
Questionnaire … 6. Questionnaire layout

 Which category? How many? How much?

Questionnaire … 6. Questionnaire layout

 Choose from a list

Questionnaire … 6. Questionnaire layout

 Agree/disagree with a statement

Questionnaire … 6. Questionnaire layout

 Rating scale
Questionnaire … 6. Questionnaire layout

 Open question
 To explain an answer to a previous question.
 What do you think about . . .?
 Anything else you want to say . . . ?
 Filter questions
Questionnaire … 6. Questionnaire layout

Providing Instructions
 Every questionnaire whether self administered or
administered by an interviewer should contain clear
instructions.

 General instructions: basic instructions to be followed in

completing it.

 Introduction: If a questionnaire is arranged into subsections

it is useful to introduce each section with a short statement
concerning its content and purpose.
Questionnaire … 6. Questionnaire layout

 Specific Instructions: Some questions may require special

instructions.

 Interviewers instruction: It is important to provide clear

complementary instruction where appropriate to the
interviewer.
Questionnaire … 6. Questionnaire layout

Reproducing the questionnaire - Nowadays less important

 A neatly reproduced instrument will encourage a higher
response rate, thereby providing better data.

 Pilot Survey: The final test of a questionnaire is to try it

on representatives of the target audience.
Questionnaire … 6. Questionnaire layout

 Type size: a font size between 10 and 12 is considered the best in most
circumstances.
 Use of all upper case text – it is best to avoid upper case text.
 Consistency of layout
 If consistency and logical patterns are introduced into the form
design, it eases the form filler’s task. Patterns that can be useful
are:
 White spaces for responses
 Using the same question type throughout the form using the
same layout throughout the form using a different style,
consistently, for instructions or directions.
 Line length – keep short.
 Character and line spacing – 7-8mm between lines and 4-5mm
between characters
Questionnaire Design… Conclusion

 The survey-design business has lots of conventions or rules of

thumb. You can use the following suggestions when reviewing
your instrument.
 The Golden Rule: Remember that you are imposing in the life of
your respondent.
 You are asking for their time, their attention, their trust, and
often, for personal information.
 Therefore, you should always keep in mind the "golden rule"
of survey research:
 Do unto your respondents as you would have them do unto you!
Questionnaire Design… Conclusion

 Do unto your respondents as you would have them do unto

you! To put this in more practical terms, you should keep the
following in mind:
 Thank the respondent at the beginning for allowing you to
conduct your study.
 Keep your survey as short as possible—only include what is
absolutely necessary.
 Be sensitive to the needs of the respondent.
 Be alert for any sign that the respondent is uncomfortable.
 Thank the respondent at the end for participating.
 Assure the respondent that you will send a copy of the final
results—and make sure you do.
Questionnaire Design… Conclusion

 Start with easy, nonthreatening questions.

 Put more difficult, threatening questions near the end.
 Never start a mail survey with an open-ended question.
 For historical demographics, follow chronological order.
 Ask about one topic at a time.
 When switching topics, use a transition.
 Reduce response set (the tendency of respondent to just
keep checking the same response).
 For filter or contingency questions, make a flowchart.
Principles of Good Interviewing
 Have integrity, and be honest.
 Have patience and tact.
 Pay attention to accuracy and detail.
 Exhibit a real interest in the inquiry at hand, but keep
your own opinions to yourself.
 Be a good listener.
 Keep the inquiry and respondents’ responses
confidential.
 Respect others’ rights.
Assignment – Questionnaire design

Do it in group/individually
Solution will be supplied upon
submission
Measurement Scales
 Measurement Scales

Population - Parameter

Sample - Statistic

Variable - Variate

Attributes

Values

Nominal scale Ordinal scale - Interval scale – Ratio scale –

- name Rank no absolute zero absolute zero

120
Measurement Scales

 An attribute (variate) is a specific value on a variable. For

instance, the variable sex or gender has two attributes: male and
female, or, the variable agreement might be
defined in a particular study as having five attributes:
 1 = strongly disagree
 2 = disagree
 3 = neutral
 4 = agree
 5 = strongly agree
Measurement Scales

 Variables aren’t always quantitative or numerical and

it’s not necessary to assign numbers for something to be
a variable.
 The variable gender consists of two values expressed in words:
male and female. These values of the variable gender can be
called “text values” to differentiate them from numeric values.
However, if it is useful, quantitative values like “1” for female
and “2” for male can be assigned instead of (or in place of) the
words.
Measurement Scales
Measurement Scales

 Levels of measurement
 The relationship between numerical values on a measure.
 There is a hierarchy implied in the level of measurement idea.
 At lower levels of measurement, assumptions tend to be less
restrictive and data analysis tend to be less sensitive.
 At each level up the hierarchy, the current level includes all of
the qualities of the one below it and adds something new.
 In general, it is desirable to have a higher level of measurement
(such as interval or ratio) rather than a lower one (such as
nominal or ordinal).
Measurement Scales

 Rationale - knowing the level of measurement helps (or

important):
 Decide how to interpret the data from the variable.

 Decide what statistical analysis is appropriate on the

values that were assigned (e.g. For instance, it makes
sense to compute an average of an interval or ratio
variable but does not for a nominal or ordinal one)
Measurement Scales
 Measurement levels (in decreasing order of precision)
1. A ratio variable is measured on a mathematical scale with
equal intervals and a fixed zero point
 Eg Journey time from London to Brussels was 2 hours in 2006,
but 1½ hours in 2009 (so 25% faster)
2. An interval variable is measured on a mathematical scale
with equal intervals and an arbitrary zero point
 Eg It was 5oC yesterday, but 10oC today (so it was warmer by
5oC, but not twice as warm because 0oC does not mean there is
no temperature)
 As they are measured on a mathematical scale, ratio and
interval variables are quantitative variables
126
Measurement Scales
3. An ordinal variable uses numerical codes to identify the order or
rank of each category
 Eg Order of preference (1st , 2nd, 3rd)
 Rating scales (eg where 5 = strongly agree, 4 = agree, 3 =
neutral, 2 = disagree and 1 = strongly disagree) can be
treated as ordinal or interval variables
4. A nominal variable uses numerical codes to identify named
categories
 Eg Geographical location where 1 = England, 2 = Wales, 3 =
Scotland, 4 = Northern Ireland
 As they are not measured on a mathematical scale, ordinal and
nominal variables are categorical variables
127
Measurement Scales
 We can summarize our description of the four scales this way: If
you can say that:
 One object is different from another, you have a nominal scale;

 One object is bigger or better or more of anything than another, you have an
ordinal scale;
 One object is so many units (degrees, inches) more than another, you have
an interval scale;
 One object is so many times as big or bright or tall or heavy as another, you
have a ratio scale.
Measurement Scales
 Other distinctions
 A quantitative variable (i.e. all ratio and interval variables) can
be
 A continuous variable where the data can take any value
within a given range (eg time = 7 or 7½ hours)
 Or a discrete variable where the data can take only one of a
range of distinct values (eg Employees = 7 but not 7½)
 A dichotomous variable has two groups and can be
 A categorical dichotomous variable with two categories (eg
gender might be coded 1 if female and 0 if not)
 Or a quantitative dichotomous variable known as a dummy
variable (coded 1 if characteristic is present and 0 if not)
129
Measurement Scales
 The measurement level of the variable determines which
descriptive statistics are appropriates
 Your choice depends on your research questions, which may
also require the use of inferential statistics
Exploratory Descriptive Measurement
analysis statistics level
Frequency Percentage frequency Ratio, interval, ordinal, nominal
distribution
Measures of central Mean Ratio, interval
tendency Median Ratio, interval, ordinal
Mode Ratio, interval, ordinal, nominal
Measures of Range Ratio, interval
dispersion Standard deviation Ratio, interval
Measures of Skewness Ratio, interval
normality Kurtosis Ratio, interval
130
Measurement Scales
Measurement Scales
Measurement Scales
Measurement Scales
Measurement Scales
Exercise 1
Classifying variables (more than one answer may apply)

 A ratio variable can be

 Categorical / Continuous / Discrete / Interval / Quantitative
 A categorical variable can be
 Continuous / Dichotomous / Discrete / Interval / Nominal
 A dummy variable can be
 Categorical / Continuous / Dichotomous / Quantitative
 A quantitative variable can be
 Categorical / Continuous / Discrete / Interval / Ratio
 A dichotomous variable can be
 Categorical / Continuous / Dummy / Interval / Quantitative

136
Solution 1
Classifying variables

1. A ratio variable can be

 Categorical/ Continuous/ Discrete/ Interval/ Quantitative
2. A categorical variable can be
 Continuous/ Dichotomous/ Discrete/ Interval/ Nominal
3. A dummy variable can be
 Categorical/ Continuous/ Dichotomous/ Quantitative
4. A quantitative variable can be
 Categorical/ Continuous/ Discrete/ Interval/ Ratio
5. A dichotomous variable can be
 Categorical/ Continuous/ Dummy/ Interval/ Quantitative

137
Exercise 2 (Source: Practical Research (Leedy)

1. An environmental scientist collects water samples from streams

and rivers near large industrial plants and saves exactly 1 liter of
water from each sample. Then, back at the lab, the researcher
determines the amounts of certain health-jeopardizing bacteria in
each sample. What measurement scale does the measurement of
bacteria content reflect?

2. A market researcher is studying the relationship between (a) a

country’s average annual relative humidity levels and (b) the
number of raincoats sold in the country during the year. What
scales underlie the two variables in this study?
Exercise 2
3. A sports journalist in Spain wants to determine whether a football
club’s fan club membership correlates to the frequency with which
the club has won in the Spanish football league during the past five
years. The researcher can easily obtain information about fan club
membership and wins from the club’s records. To simplify data
collection, the researcher uses the following coding scheme for the
fan club membership: 1 = FC Barcelona, 2 = Real Madrid, and 3 =
Atletico de Madrid. What measurement scale(s) underlie (a) fan
club membership and (b) frequency of wins?
Exercise 2
4. A new audio system was installed in a theatre that has a seating
capacity of 500 people. The seating area has been divided into three
distinct areas: platinum, gold, and silver. At the end of every show
in one week, the audience was asked to fill out a questionnaire to
determine the effectiveness of the new audio system. What
measurement scale does the researcher’s coding scheme for the
seating area represent?
Exercise 2
5. An advertiser is studying the impact of a chocolate commercial on
different age groups. He puts the age groups in four categories: Group 1
includes 3- to 12-year-olds, Group 2 includes teenagers aged 13 to 19,
Group 3 includes adults aged 20 to 49, and Group 4 includes people aged
50 and above. What kind of scale is the classification of age groups in this
study?

6. A car manufacturer is studying sales patterns over five years of four

different types of car that vary in size and available utilities: hatchbacks
(small cars with no trunk space), midsized sedans (cars with limited
trunk space and more utilities compared to hatchbacks), sedans (proper
trunk space and all utilities), and luxury sedans (big trunk space and
luxurious interiors). Based on size and available utilities, which
measurement scale does the type of car reflect?
Exercise 2
7. A child psychologist is developing an instrument designed to
measure the social etiquette of small children. The instrument
includes 25 statements; for example, “wishes elders when they meet
them,” “courteous to their sibling,” and “says please when asking for
something or thank you when someone has helped.” Parents must
rate each of these statements on a 5-point scale as follows:
0 never; 1 rarely; 2 sometimes; 3 often; 4 always
Parents who answer “never” to each of the 25 questions get the
lowest possible score of 0 on the instrument. Parents who answer
“always” to each of the 25 questions get the highest possible score of
100 on the instrument. Thus, scores on the instrument range
from 0 to 100. What kind of scale do the scores represent?
End of Ch05 – Next Ch06 (Self study)

End of the Course.

Thank you.

Compiled by Andualem 143

(2015/2023)