Methods of collecting data
Survey, methods and type, response rate, variable language
Hands on: Graphical techniques II, SPSS
Questionnaire design
Tips on writing a research paper
Individual project: article critique
Seal of Good Local Governance (SGLG) 2024Final.pptx
3 survey, questionaire, graphic techniques
1. Quantitative Research Methods
Lecture 3
1. Methods of collecting data
2. Survey, methods and type, response rate, variable language
3. Hands on: Graphical techniques II, SPSS
4. Questionnaire design
5. Tips on writing a research paper
6. Individual project: article critique
2. Methods of Collecting Data…
• There are many methods used to collect or obtain
data for statistical analysis. The most popular
methods are:
• Direct observation (E.g. Number of customers
entering a bank per hour)
• Experiments (E.g. new ways to produce things to
minimize costs)
• Content analysis
• Surveys
3. Types of research
• Business
• Formative evaluation
• Summative evaluation
• Need analysis
• Consumption level
• Consumer satisfaction level
• Future trend prediction
• Identification of persuasive strategies
Survey
4. What is survey
• Survey is a research technique that uses a
standardized questionnaire to collect
information about attitudes, opinions,
behaviors, and background and lifestyle
characteristics from a sample of respondents.
5. Pre-research phase of survey
• Step 1. Decision maker monitor the environment
and literature
▫ How?
▫ Read/watch widely, attend conferences/seminars
• Step 2 Identify new trends, problems, issues,
opportunities, or theories.
6. • Step 3: Specify research questions and
hypothesis
• Step 4: Following ethical standards in
conducting survey
▫ Voluntary
▫ No harm to the participants
▫ Confidential
7. • Step 5: choose the survey methods regarding time,
budget/staff, response rate, security concern:
▫ Telephone
▫ Mail
▫ Face to face
▫ Online
• choose the survey type
▫ Cross-sectional
▫ Longitudinal
Trend: different people in each survey
Panel: same people
Cohort: people share same characteristics, e.g. baby
boom, Generation X (1960-80s)
8. • Step 6: population to survey, sampling
• Step 7: the budget and timetable
• Step 8: selecting a research expert:
9. Survey Type and Methods
• Choose survey type
▫ Cross-sectional
▫ Longitudinal
Trend: different people in each survey
Panel: same people
Cohort: people share same characteristics, e.g. baby
boom, Generation X (1960-80s)
• Choose survey methods regarding time,
budget/staff, response rate, security concern:
▫ Telephone
▫ Mail
▫ Face to face
▫ Online
10. Telephone interview
Advantages
• cheap to administer
• convenient for
interviewers and
respondents
Disadvantages
• interviews easily
terminated by
respondent
• biased sample through
households with phones
11. Mail Survey
Advantages
• cheaper to administer
• more private and confidential
• in most cases doesn't require
interviewers.
Disadvantages
• difficult to follow-up non
response
• respondent bias
• response rates are lower
• language must be simple
• no respondent rapport.
12. Face to face
Advantages
• special questioning techniques
can be used
• greater rapport with respondent
• allows complex issues
• higher response rates
• can explain things to
respondents
• can deal with language problems
Disadvantages
• expensive to conduct
• training for interviewers is
required
• more intrusive for the
respondent
• interviewer bias may become a
source of error
13. Online survey
Advantages
• cheap to administer
• private and confidential
• easy to use conditional questions
and to prompt if no response or
inappropriate response.
• can provide multiple language
versions
Disadvantages
• respondent bias may become a
source of error
• not everyone has access to the
internet
• language and interface must be
very simple
• cannot build up a rapport with
respondents.
14. Advantages of surveys
• 1) Standardization
▫ All respondents react to questions worded identically and
presented in the same order and format
Make data recording, analysis and comparison easier
• 2) Administration ease
▫ Easy to administer (e.g. mail survey, drop-off, online)
▫ Much simpler than in-depth interviews (that involves tape-
recording, taking notes and transcribing)
less time consuming for both researchers and subjects
easy to gather large sample size in a short period of time
cost-efficient
• 3) Tap the ‘unseen’
▫ Can measure unobservable things like attitudes/ values/
intention/ beliefs etc.
15. Disadvantages of Surveys
• 2 major problems in survey
• 1) Non response
▫ Questionnaires are not returned
• 2) Self-selection bias
▫ Those who respond are probably different those who
do not fill out the questionnaires and return it
E.g., more educated, more involved with the topic, more
opinionated
Non-representative of the general population
16. Response rate
• Response rate = completed and returned
surveys/ total number of surveys sent out x
100%
▫ Total number of surveys distributed= 1000
▫ Surveys returned= 250
▫ Response rate= 25%
• Normal response rate for mailing survey: Less
than 20%
▫ If you need 1000 subjects, at least how many
surveys you need to distribute? (At least 5000
surveys)
17. Methods to increase response rate
• Preliminary notification
▫ Advanced letter or postcard, telephone
prenotification
• Foot-in-the-door
▫ Small request large request
▫ Less sensitive issue first
• Personalization
▫ Personal signature, individually addressed cover
letter, individual invitation
• Anonymity
▫ Assurance of anonymity
18. Methods to increase response rate
• Response deadline
-Specifying due dates
• Appeals
-An act of helping the sponsor, the
community, or a needing group
• Sponsorship
▫ Company, trade association, university
researcher
19. Methods to increase response rate
• Incentives
▫ Nonmonetary : coupons, pens, summary of findings
▫ Monetary: given to respondent, or donated to charity
• Follow-ups
• Clear instructions
• Inquiry contact
20. Research Phase
1. Search for your research question.
2. Conduct literature review
3. Identify concepts, variables, and operational
definitions, i.e. measurement
4. Questionnaire design
21. The variable language
• So, what are you studying in variable
language?
• Regular: “I want to study whether men and
women differ in their feeling of happiness at
work.”
• Variable language: “I want to study the
relationship between gender and
perception of happiness at work.”
22. The variable language
• Not just a matter of language, it’s a relationship,
independent variable & dependent variables
Gendar
Perception of
happiness at work
?
23. Small exercise on variable language
• Do they think favorably towards UAEU DBA
program?
• Attitudes towards UAEU DBA program.
• How much they fear about statistics?
• Degree of fear in statistics
• How often they have conflict with their spouse in
a month?
• Frequency/ number of conflicts with spouse in a
month
24. Hands-on: Graphical Descriptive
Techniques I
• SPSS
Graphical
Techniques
Objective Data
type
Ex Tips
Histogram
(Stem-and-
leaf Display;
Ogive)
Describe a single
set of data
Interval P45 Xm03-01
Xm03-03
Xm03-02(chp4)
P49 Table 3.2/equation
Number of classes
Class width(not always)
Line chart Describe a single
set of time series
data
Interval P65 Xm03-06 Graphic Deception:
focus on numerical
values, scales on axes,
Scatter
Diagram
Describe the
relationship
between two
variables
Interval Ex P43/78
Xm03-00
Linearity
Strength & direction
Correlation is not
causation
25. Shapes of histogram
• Symmetric
• Skewness: positively skewed, negatively skewed
• Unimodal and bimodal
• Bell shape
27. The survey questionnaire
• The survey questionnaire can be divided into four
main sections:
▫ Record keeping information:
the respondents phone number, the interviewer’s id No.,
etc
▫ An introduction to the survey
Introduce the interviewer and the survey to respondents
▫ Questions:
Main part, must follow scientific principles
▫ A closing:
thanks respondents for participating
28. Ethical standards
• Do not intentionally mislead respondents
• Do not have false information in the question.
29. Guidelines on asking questions
Two options to choose when asking questions
Close-endedOpen-ended
30. Open-ended questions
• Respondents are asked to provide his/her own
answers
• “How do you feel about Emirates Airline?”
______________________________
______________________________
______________________________
______________________________
______________________________
______
31. Close-ended questions
• 1) The researcher provides a list of answers; and
• 2) The respondent is asked to select an answer
from the list
• “How much do you like Emirates Airlines?”
Not at
all
Very
much
1 2 3 4 5
32. Open-ended vs. close-ended questions
• Close-ended
• Greater uniformity
of responses
• More easily
processed: the
numbers chosen can
be input directly
into the SPSS
• No in-depth / ‘extra’
data
• Open-ended
• The textual responses
must be coded before
they can be processed
for statistical analysis
• Respondents may give
answers that are
irrelevant to your study
• Time-consuming
• In-depth/ ‘surprise’ data
33. Close-ended question
• Something to be careful about when asking
close-ended question:
• What’s wrong with the following close-ended
questions?
1) How many days a week do you watch Skynews
Arabia?
a) 0 days a week
b) 1 or 2 days a week
c) 2 or 3 days a week
34. Rule number 1 for close-ended questions
• Rule number 1:
• The response categories provided must be
exhaustive.
• i.e., they should include ALL the possible
responses that might be expected
35. Exhaustive list of responses
1) How many days a week do you watch Skynews Arabia?
a) 0 days a week
b) 1 or 2 days a week
c) 3 or 4 days a week
d) 5-7 days a week
36. Exhaustive list of responses
• For questions with various possible responses
such as:
• “What is the most important issue facing UAE?”
• Method 1: Make it an open-ended question and
code each response
• Method 2: Make a list of responses as exhaustive
as you can and add “Other” at the end of the list
37. Exhaustive list of responses
a) Drop of the oil price
b) Political instability
c) Competitive Gulf countries
d) Housing
e) Education
f) Increasing divorce rate
g)Environmental problems
h) Others (please specify:__________)
“What is the most important issue facing UAE?”
38. Rule number 2 for close-ended
questions
• What’s wrong in here?
• “How many members do you have in your
family?”
a) 2 or below
b) 2-3
c) 4-6
d) 6 or above
39. Rule number 2 for close-ended
questions
• What’s wrong in here?
• “How many members do you have in your
family?”
a) 2 or below
b) 2-3
c) 4-6
d) 6 or above
Overlap in choices
40. Rule number 2 for close-ended
questions
• Rule number 2:
• The answer categories must be mutually
exclusive
• A response is denoted by 1 category only
• “How many members do you have in your
family”
a) 2 or below
b) 3-4
c) 5-6
d) 7 or above
41. Guidelines in asking questions
• What’s wrong with these questions?
• “Your impression on the President is:”
• “Are you a native speaker?”
• An item found in a Society for the Prevention of Cruelty
to Animals survey:
• “When your puppy has an accident, do you discipline
it?”
42. Guidelines in asking questions
• Rule 1: Make items clear. Avoid any possibilities
of misunderstanding
• “Your impression on the incumbent President of the
United States is:”
• “Are you an Emirati native speaker?”
• “When your puppy urinate on the floor, do you spank
it?”
43. Guidelines in asking questions
• What’s wrong with these questions?
• “How much you agree or disagree with this
statement: The United States should abandon its
space program and spend the money on
domestic programs.”
• Here’s one of the items in a Toys R’ us survey
• “Did you know that Toys’R Us sells mainly
educational toys, and it is the only American toy
retailer selling toys in UAE?
44. Guidelines in asking questions
• Rule 2: Avoid double-barreled questions
• Double-barreled questions:
▫ Two or more questions posed in one question.
▫ Asking for a single answer to a combination of
questions
• Solutions: Simply break them into 2 questions
45. Guidelines in asking questions
• Rule 3: Respondents must be competent to
answer the questions
• Asking people on these:
▫ “What was the first word you spoke out?”
▫ “When did you first go to a hospital?”
• Asking people who are not the employees of ADNOC
on this:
▫ “The employee benefits of ADNOC are excellent.”
• Asking a high school student on this:
▫ “How would you comment on Abu Dhabi government’s new
housing policy?”
• What’s wrong with these?
46. Guidelines in asking questions
• Rule 4: Questions should be relevant to
the participants
• Asking university students about child-rearing
practices
• Asking primary school kids about their attitudes
towards salary cut
• Not relevant no opinions make
artificial responses only
47. Guidelines in asking questions
• Rule 5: Avoid negative questions
• “The United States should not recognize Taiwan.”
____ yes _____ no
• What’s wrong in here?
• Possibility 1: A sizable portion of respondents
often read over the word “not” and answer on
that basis
• Possibility 2: A ‘yes’ may mean: (a) respondents
agreeing to the statement; or (b) respondents
thinking the United States should recognize
Taiwan.
48. Guidelines in asking questions
• Rule 6: Avoid biased items and terms
• Bias: any property of questions that encourages
respondents to answer in a particular way
• Avoid leading questions
• Look at the following item in a Ray-Ban survey:
• “How willing are you to pay for a pair of sunglasses
that will protect your eyes from the sun’s harmful
ultraviolet rays, which are known to cause
blindness?”
• Present questions in a neutral tone rather than in a
positive or a negative tone
• How would you do it?
49. Guidelines in asking questions
• Rule 6: Avoid biased items and terms
• Which is better?
a) “Don’t you see some danger in the new policy?”
b) “Do you see some danger in the new policy?”
• B is better.
• In A, the respondent is led to expect that there
are dangers in the new policy
50. Guidelines in asking questions
• Rule 6: Avoid biased items and terms
• Avoid emotionally-charging words
• ‘absurd policy’ ‘cruel massacre’ ‘the inhumane
abortion policy’
• Avoid setting conditions
• ‘If (Some political leader) is found being
involved in bribery, should he step down?”
51. Guidelines in asking questions
• Rule 7: Check on the possibility of social
desirability of questions and answers
• Social desirability: A human tendency to
respond in a way that will make the person looks
good/ ethical
52. Guidelines in asking questions
• Studying Emirati males’ gender stereotype
• “I want my wife to stay at home and take care
of the family.”
• A possibility of social desirability that male
respondents may not dare to choose ‘strongly
agree’
• Tone down a bit:
• “I find it more comfortable if my wife puts more
energy in taking care the family than caring
about developing her career.”
53. Ways of asking
• Two ways of asking:
• 1) In question format
• 2) In statement format
54. Question format
•2 ways to use question format
•1) open-ended
•“How much do you like Starbucks?”
______________________________
______________________________
______________________________
_____________
55. Question format
• 2 ways to use question format
• 2) close-ended
• “How much do you like Starbucks?”
Not at
all
Very
much
1 2 3 4 5
56. Statement format
• In the format of a statement:
• Agree or disagree—and how much
• The most common statement format:
• Likert-type scale: a format in which respondents are
asked how much they agree / approve a statement.
• “I like Starbucks.”
Strongly
disagree
Strongly
agree
1 2 3 4 5
58. Original Likert-type scale
• 1) “I am very satisfied with my weight.”
• 2) “ I think I weigh heavier than I should have.”
Totally
disagree
Totally
agree
1 2 3 4 5
Totally
disagree
Totally
agree
1 2 3 4 5
59. Modified Likert-type scale
Strongly Strongly
disagree agree
• A) “I support tuition 1 2 3 4 5
fee reduction.”
B) “ I think private 1 2 3 4 5
university’s tuition fee
is too high.”
C) “We would be burdened 1 2 3 4 5
a lot more if tuition fee is
raised.” Matrix Question
Format
60. Alert!!
Never do this:
Strongly Strongly
disagree agree
• A) “I support tuition 1 2 3 4 5
fee reduction.”
• B) “ I think private 1 2 3 4 5
university’s tuition fee
is too high.”
61. Shortcoming of matrix format
questions
Strongly Strongly
disagree agree
A) “I support tuition 1 2 3 4 5
fee reduction.”
B) “ I think private 1 2 3 4 5
university’s tuition fee
is too high.”
C) “We would be burdened 1 2 3 4 5
a lot more if tuition fee is
raised.” Foster
response-
62. Various Formats for Respondents
• 1) Circling the appropriate the number
Strongly
disagree
Strongly
agree
1 2 3 4 5
63. Various Formats for Respondents
• 2) Checking a response
( ) Yes [ ] Yes
( ) No [ ] No
( ) Don’t know [ ] Don’t know
64. Various Formats for Respondents
• 3) Contingency questions
• “Are you a smoker?”
1 Yes
2 No
If yes: How often do you smoke:
1 Two or less cigarettes / day
2 3-5 cigarettes/ day
3 6-8 cigarettes/ day
4 9 or above / day
65. Various Formats for Respondents
• For Complex Contingency questions:
• For more than 3 contingency questions, use
‘skip’
• Have you ever tried to lose weight?
__ yes (Please answer questions 19-24)
__ no (Please skip questions 19-24. Go
directly to question 25 on page 3)
66. Things to Keep in Mind while Drafting
the Questionnaire
• Study purpose: what type of information is
required to test the hypothesis
• Respondents: who they are, how well educated
and operative they are, etc.
• Interview mode: different requirements for face-
to-face, phone, or mail/self-administered survey
• Data entry and analysis: what the data structure
will look like and what analysis will be involved
67. Key Elements of Questionnaire
• How to design questions and answers:
▫ Content of questions and answers
▫ Format of questions and answers
▫ Wording of questions and answers
▫ Order of questions
▫ Length of questionnaire
68. Content of Questions & Answers
• Is the question absolutely necessary to ask?
• Is it a double- or multi-barrel question (i.e., a
question with two or more objects)?
• Do respondents have required background
knowledge to answer the question?
• Will respondents tell the truth to the question?
69. Format of Questions & Answers
• Open-ended questions
• Close-ended questions
▫ Multiple choice questions
Only one choice permitted
Multiple choices permitted
Dichotomous questions (dummy variable)
Check list based on dichotomous/multiple choice questions
▫ Likert Scale for attitudinal questions
▫ Semantic Differential Scale for attitudinal questions
▫ Ranking questions
▫ Paired comparison questions
70.
71. Paired comparison questions
• This requires the respondents to choose between pairs of choices at a time,
e.g. there will be 4 brands of cars: BMW Z4, Chevrolet Corvette, Porsche
Boxster, Dodge Viper
• A respondent may be asked to do a paired comparison to choose which
brand is better, but only two brands at a time.
• This process has to be repeated for as many pairs as exit in the given set of
objects or brands.
72. Wording of Questions & Answers
• Correct and accurate
• Clear and easy to understand
• Precise
• In spoken language
• Personalized
• Complete and mutually exclusive
• Mid-ground position provided for attitudinal
questions
73. Order of Questions
• From simple to complicated questions
• From general to specific questions
• From easy to difficult questions
• From insensitive to sensitive questions
• Questions about personal information asked at
the end
• Avoid potential impacts of previous questions on
subsequent questions (e.g., hints, framing,
expectations)
75. Impact of Questionnaire Length on
Refusal Rate
0
2.5
5
7.5
10
0 5 10 15 20 25 30
Time
Refusal(%)
76. Criteria for Questionnaire Evaluation
Will it provide sufficient information to test the
hypothesis?
Will it be easy and friendly for the respondents?
Will it be convenient to administer the
interviews and enter the data?
Will it be appropriate for statistical analysis?
78. A Research Paper
Title page
Abstract
Introduction
Literature Review
RQ & Hypothesis
Method
Results
Discussion
References
Tables
Figures
Appendix
Subjects
Measures
Procedures
Main Data
Analysis &
Report of the
results
79. How to do literature review
• Step 1. Become familiar with the library and its
reference resources and tools for retrieving
information
• Step 2. Indentify relevant publications, database,
unpublished reports, and websites.
• Step 3. Find and skim articles to identify those
that are relevant. Review the list of references
and footnotes to indentify additional sources
that can be searched.
80. • Step 4. Read relevant articles carefully and
summarize.
• Step 5. Organize the major findings of articles.
• Step 6. Sort results according to consistent and
discrepant findings.
• Step 7. Identify areas that lack any research.
• Step 8. Draft a synthesis of the overall research
findings on the topic as well as the areas that
have yet to be researched.
81. How to use library resources
• UAEU library:
▫ http://www.library.uaeu.ac.ae/
82. Individual project
• Article critique:
▫ Academy of management review
▫ Academy of management journal
▫ Organizational Studies
▫ Organizations
▫ Management Science
▫ Journal of Management Studies
▫ Human Resource Management Journal
▫ MIS Quarterly