Research Methodology Course

Lecture # 2
Data Collection Tools
&
questionnaire design

For Third year medical students – 2023-2024

1
–Differentiate between primary and secondary data

–Identify different tools for data collection

–Design a well structured questionnaire

Extended Modular Program 2

 Types of Data

Data collection is essential for answering a research question.

Methods used to gather data,

either primary data, or secondary data.

3
 Primary Data
Data originated for the first time by the researcher specifically for
the purpose of his research problem.
Known as the first hand or raw data.
Under direct control and supervision of the investigator.
Could be collected through various methods like surveys,
observations, physical and laboratory examination, questionnaires,
focus groups, case studies …etc.

4
 Secondary Data
❑ Second-hand information
❑ Already collected and recorded by other person than the investigator
❑ Not related to the current research problem.
❑Collected from various sources like censuses, government publications,
internal records of the organization, reports, books, journal articles, websites
…etc.
❑ Advantages:
▪ easily available,
▪ saves time and cost on the researcher.
❑ Disadvantages:
▪ May or may not be specific to the researcher's need and objective.
▪ Relatively less accurate.
5
 Basis for
Primary data Secondary data
comparison
Primary data refers to the first-
Secondary data means data collected by
Meaning hand data gathered by the
someone else earlier.
researcher himself.
Process Time consuming Quick and easy
censuses, government publications,
Surveys, observations,
websites, books, journal articles, internal
Source experiments, questionnaire,
records, patient records, birth and death
personal interview, etc.
certificates etc.
Expensive: requires resources like
Cost
investment, manpower, training Economical
effectiveness
and time.

Always specific to the researcher's May or may not be specific to the

Specific
needs and objective. researcher's need and objective.
Available in Crude form Refined form
Accuracy and Under the control of the
Relatively less
Reliability researcher 6
 Data Collection Tools

The process by which the researcher collects

information needed to answer the research
question

7
 Examples for data collection tools

• Checklists
• Questionnaires
• Personal interviews
• Case Report Form (CRF) and
worksheets

8
 Examples for data collection tools

Checklists
–Used for documenting a person’s observation
or evaluation regarding a certain
performance.

–They can be simple lists of criteria that can

be marked as present or absent, or over a
certain scale and can provide space for
observer comments.

9
 OBSERVATION #1 #2 #3
Hand hygiene technique with soap and water
1. Wet hand with water using temperature that is comfortable.
2. Apply appropriate soap, and wash hands and wrists vigorously for
15 seconds, covering all surfaces of hands and fingers. (No bar soap.)

3. Rinse hands under running water, allowing water to drip from

fingertips.
4. Thoroughly dry hands and wrists with paper towel.
5. Turn faucet off using a dry paper towel to touch the handle,
protecting your clean hands from the contaminated handle.
Hand hygiene technique for waterless sanitizer
1. Apply product to palm of one hand.
2. Rub hands together covering all surfaces of hands and fingers until
hands are dry.
3. Fingernails (Direct patient care givers)
Fingernails (Direct patient care givers)
1. No artificial nails or enhancements.
2. Natural nails are short (1/4 inch) and well groomed.
3. Hands, including the area around nails, free of inflammation.
4. Nail polish is in good repair.
Describe corrective action for any item not performed properly.
Signature of validator
Signature of employee
10
 Examples for data collection tools

Case Report From (CRF) and worksheets:

Tools for organizing and reporting data in the medical research

as:
laboratory and radiological tests, clinical examination and others
like patient reported outcomes: physical functioning or pain
assessment.

11
Case Report From (CRF) and worksheets:

12
 Examples for data collection tools
Questionnaires:
–Instruments used for collecting data in survey research.

–Include a set of questions that explore a specific topic and

collect information about demographics, opinions,
attitudes, or behaviors.

–Could be self-administered or interview questionnaire.

13
14
15
Questionnaire Design

16
 Constructing a Questionnaire

1. Formulate objectives
2. Review necessary information
3. List the variables to be collected
4. Formulate questions (preliminary draft), translate
5. Test the questionnaire (pilot study)
6. Revise and redesign

17
 Self administered Interview
Definition Questionnaire given to the Questionnaire filled by an
respondent to fill by himself interviewer
Cost Cheap Expensive
Time Less More
Criteria of respondent Well-educated and Illiterate
motivated respondents
Response rate Relatively low High
Interviewer bias Absent Present
Others The language, editing and requires training of data
formatting of the collectors
questionnaire are important 18
 Types of questions

Open Closed
ended ended
questions questions
Respondents
Respondents
choose one or
answer in their
more of pre-
own words
selected answers

19
 Types of questions
Open-ended questions:
Are useful in exploring the motives, attitudes and beliefs.
They are difficult to code and need special methods of
analysis.

Example:
What habits do you believe increase a person’s
chance of having a heart attack?

20
 Types of questions
Advantages Disadvantages

It gives a variety of
responses and it truly Difficult to code and
reflect the opinions of the to analyze.
respondents especially in
exploratory phases of
question design.
21
 Types of questions
Closed-ended questions :
Range of possible responses is already known.
They are easier to code ad analyze,
What do you think about the standard of service you
received at the family health care center:
• Excellent
• Very good
• Good
• Bad

22
 Types of questions:
Closed-ended questions
Advantages Disadvantages

Quicker and easier to answer. They lead respondent in certain directions and
don’t allow him to express his own answers.
Answers are easier to tabulate and analyze.
Responses listed by researchers may not include
List of possible answers often helps to clarify the an answer that is most appropriate for a
meaning of the questions. particular respondent

23
 Types of questions

The Dichotomous answer :

➢ The simplest form,

➢ Easy to analyze
➢ (Yes-No, Agree-Disagree, Approve-Disapprove).

24
 Types of questions

Likert scale:
➢ Is a type of psychometric response scale widely used in survey
research for attitudes.
➢ Measures level of agreement to a statement.
Likert scaling usually includes the options:
totally agree, agree, neutral, disagree, and totally disagree.

25
26
A Visual Analogue Scale (VAS):
is a measurement instrument that tries to measure a
characteristic or attitude that is believed to range across a
continuum of values and cannot easily be directly
measured.

27
A Visual Analogue Scale (VAS):

28
 General rules for constructing the questionnaire

1. Health questionnaires start with some basic socio-

demographic questions, such as age, gender, education.

2. Start with easy questions. Questions dealing with sensitive

issues should be delayed to near the end of the questionnaire.

3. The questions should be ordered so that they flow logically.

29
 Errors in Questionnaire Design
There are 2 types of errors in designing a questionnaire:

Errors in Errors in
Question Construction of
Wording Response
Categories

30
1. Errors in Question
Wording

31
 1- Question wording: what to avoid

1. Leading questions
2. Emotionally loaded words or phases
3. Questions that assume knowledge of technical terminology
4. Double- barrelled questions
5. Use of negatives
6. Vague, imprecise or ambiguous questions

32
 1. Leading questions
Using leading words that suggest that there is a more desirable
answer

You don’t sleep well, Do you?

Yes ( ) No ( )

Correct
Do you sleep well?
Yes ( ) No ( )

33
 2. Emotionally loaded words or phases
The use of strong harsh or emotive words may distort responses.

e.g. : “Do you consider smoking a bad/dirty habit?

Yes ( ) No ( )

In this example the use of the word “dirty” is likely to increase an

affirmative response.

34
3. Questions that assume knowledge of
technical terminology
Example:
Instead of “jaundice”
Use “yellow coloured skin/sclera”

35
 4. Double-barrelled questions
“Have you had nausea, vomiting or abdominal colic since your
last visit?
Yes ( ) No ( )
It is not clear whether a “Yes” response refers to any or all of the
manifestations mentioned in the question head.
Since your last visit
Have you had nausea ? Yes ( ) No ( )
Have you had vomiting ? Yes ( ) No ( )
Have you had abdominal colic? Yes ( ) No ( )
36
 5. Use of negatives
Negative should be avoided, as it makes the question confusing

“Are you against not smoking in public places?

Yes ( ) No ( )

37
 6. Vague, imprecise questions
For example:
Words such as “many”, “often”, “occasionally”, “enough”, …

▪ Refer to imprecise quantity, meaning different things to different

people.
▪ If frequency is of interest, the number of times an event occurred
in a specified period should be incorporated into the question
wording.
38
 2. Errors in
Construction
of Response
Categories

39
 1. Providing an incomplete list
of response categories
What is your current marital status?

1. Single ( ) 1. Single ( )
Correction 2. Married ( )
2. Married ( )
3. Divorced ( ) 3. Divorced ( )
4. Widow ( )

In this example, the category “widowed” is omitted.

40
 2. Response categories are not mutually
exclusive
What is your age?

1. 15-25 ( ) 1. 15-24 ( )
2. 25-35 ( ) Correction 2. 25-34 ( )
3. 35-45 ( ) 3. 35-44 ( )
4. 45-55 ( ) 4. 45-54 ( )

41
 2. Keep numerical variables without
categorization
What is your age? ……………………….

20 years.

42
 3. Response categories are not mutually
exclusive
What are the elements of Diabetes management explained to
you by the physician?
1. Diet control ( ) 1. Diet control 1-Yes 2-No ( )

2. Physical exercise ( ) 2. Physical exercise 1-Yes 2-No ( )

Correction
3. Oral hypoglycemic ( ) 3. Oral hypoglycemic 1-Yes 2-No ( )
4. Insulin ( ) 4. Insulin 1-Yes 2-No ( )

The categories are not mutually exclusive, as the physician

may recommend more than one line of management.

43
 4. Response categories are not logically
ordered
How many diarrheal attacks do you have per day?

1. Less than 5 ( ) 1. None ( )

2. 10-14 ( ) Correction 2. Less than 5 ( )
3. 15 or more ( ) 3. 5-9 ( )
4. None ( ) 4. 10-14 ( )
5. 5-9 ( ) 5. 15 or more ( )

44
 5. Instructions are inappropriate

What do you understand by SARS? (Tick only one)

•Viral disease ( )
•Pandemic ( )
•Fatality rate 50% ( )
Correction (Tick only one)
•There is valid vaccination ( )

There is more than one correct answer, therefore "tick only one" is
an inappropriate instruction.
45
 6. Inconsistent layout of response
categories
oConfusing layout of the response
oBoth vertical and horizontal ordering at the same time
1. Yes/No
2. Yes ❑ No ❑
3. Yes ( ) No ( )
4. Yes ( ) No ( )
5. Don’t know ( )

46
 7. Use of abbreviations
They should be written in full or they may confuse the respondent.

Example:
o"Y/N" should be written as "Yes/No",
o"DK" as "Don’t know"
o"N/A" as "not applicable" ,
o"PA" as" per annum"
o"< >" less than or more than.

47
 8. Not providing the opportunity to respond
"Yes or No" before continuing with
subsequent parts of the question
❑ What type of contraception are you using?
Pills ( ) IUD ( ) Local ( ) Injections ( ) Others ( ) No ( )

This question should be reconstructed in 2 parts as:

Do you currently use any contraceptive method? Yes ( ) No ( )
If yes, what type of contraception are you using?
Pills ( ) IUD ( ) local ( ) injections ( ) others ( )

48
Validity and Reliability
of the collected data:

49
Reliability is the consistency of a measure (whether the
results can be reproduced under the same conditions).
Validity is the accuracy of a measure (whether the results
really do represent what they are supposed to measure).

50
Validity of a measure is the degree to which any
measurement technique or instrument succeeds in describing
or quantifying what is designed to measure.
The sphygmomanometer proved to be a valid instrument to measure the
pressure exerted by the blood on the vessel walls.
Types of validity:
• Face Validity:
• Content validity:
• Criterion validity
• Construct validity
51
Types of validity:
• Face Validity: Subjective validity judged by experts
in the field is known.
• Content validity: Ensuring that there is no element
or aspect neglected or missed in the construct.

In simple language, Face Validity looks at what is

there while Content Validity looks at what is missing.

52
Types of validity:

• Criterion Validity: is an objective validation method where

the score obtained from the questionnaire (construct)
can be correlated with “Golden Standard” test. A golden
standard is well known test to be specific and sensitive
than the one used.

53
Types of validity:

• Construct Validity: is resorted to when there is no golden

standard; the score can be correlated with the score of
another tool with proved validity.
For example, the construct validity of the Disabilities of the
Arm, Shoulder, and Hand (DASH) questionnaire is evaluated by
examining its correlation to the Medical Outcomes Study Short
Form-36 (SF-36).

54
Reliability: to what extent the measurement technique
provide consistent results when applied repeatedly.

A weight scale is reliable if it gives the same reading for the same
“standardized” weight repeatedly.

The reliability of a research tool is concerned with the extent to

which the instrument yields the same results on repeated trials.

55
Reliability:
Tests used to assess reliability of a tool:

• Inter-Rater or Inter-Observer Reliability: Used to assess

the degree to which different raters/observers give
consistent estimates of the same phenomenon.

• Test-Retest Reliability: Used to assess the consistency of

a measure from one time to another

56
57