Data Collection

Types of data
Methods of data collection
Developing research instruments
 Introduction
• Research is a systematic process that seeks to generate
new knowledge through collection, analysis and
interpretation of data to answer a certain question or
solve a problem.
• Hence in research, the investigator has to collect data to
address the study objectives
– In quantitative research, variables should be derived from the
study objectives and data collected to measure them
• The measurement of the stated variables should be done using
validated tools
– The operational definition of every variable should be done;
that is a specification of how the variable will be measured in
a given study
 Types of Data
• Data refers to all the information a researcher for
his or her study
• Types of data
– Primary vs. secondary data
• Primary data: Information that a researcher obtains by taking
observations or measurements directly from study subjects
• Secondary data: Information a researcher obtains from
published documents e.g. research articles, books, reports,
databases etc.
– Quantitative vs. qualitative data
• Quantitative data- information in form of numbers ( numerical)
• Qualitative data- information in form of words, phrases,
narratives
 Methods of Data collection
• In health research, several methods and tools
may be applied to collect data, depending on
the study objectives and variables to be
measured:
– Physical measurements using validated machines
e.g. blood pressure machines for measuring blood
pressure, weighing machines for measuring
weight etc.
 Methods of Data collection
– Laboratory and molecular tests using standard
methods e.g. genetic analysis, biomarkers for
certain exposures and outcomes etc.
– Use of questionnaires to collect information on
socio behavioral factors related to health e.g.
lifestyles, behaviors, attitudes, practices etc.
• Questionnaires represent one of the most commonly
used data collection tool in population surveys, hence a
good understanding of how they are developed is
fundamental
 Questionnaires
• A questionnaire is a set or series of questions designed
to collect information from people on certain issues
– Studies where data is collected by asking questions to be
responded to by the study participants (respondents) are
also referred to as surveys
• Each item in a questionnaire is developed to address a
specific objective or research question
• At the design stage, the researcher should determine
the method of analysis to be applied for every item in
the questionnaire.
 Types of Questions
1. Structured/closed ended questions
2. Unstructured/open ended questions
3. Contingency questions
4. Matrix questions
Structured/closed ended questions
• These are questions accompanied by a list of
all possible alternatives from which
respondents select the answer that best
describes their situation
– For certain questions the category ‘other’ is usually
included to cater for other possible answers that
may not have been stated among the given
categories
• e.g. Which religion do you belong to? 1. None 2.
Catholic 3. Protestant. 4. Muslim. 5. Other, specify.
Structured/closed ended questions
• Advantages
– Closed ended questions are easier to analyze statistically since
the responses are in form of specified categories; the codes of
the selected categories can be enumerated and analyzed using
proportions
– Easier to administer since each question is followed by a list of
alternative answers to be chosen from
• Disadvantages
– They are more difficult to construct because categories must be
well though out
– Responses are limited, and the respondent is compelled to
answer questions according to the researcher’s choices
 Unstructured/open ended questions
• These are questions which allow the respondent to
respond in his/her own words
– They give the respondent complete freedom of
response
– The space provided for each question may serve as a
guide as to whether a brief or lengthy answer is
required; in some cases only a phrase or figure is
needed for the response
– e.g. What do you do for a living?..........................
– What do you do to protect yourself from being infected
with HIV?......................................
 Unstructured/open ended questions
• Advantages
– They permit greater depth of response; reasons
underlying the responses may be directly or
indirectly given
– Open-ended questions are simpler to formulate
since the researcher does not have to generate
response categories
– Respondents have the freedom to express
themselves hence the responses can give insight
into motivation, hidden concerns, interests and
feelings about the issue at hand
 Unstructured/open ended questions
• Disadvantages
– Open ended questions may sometimes lead to responses
that are not in line with the stipulated research objectives
– The responses may vary widely hence making it difficult
to categorize them for statistical analysis
– Responding to open ended questions is time consuming
since the respondents have to come up with the
responses themselves as opposed to structured questions
where they only choose from a set of categories;
• Some respondents may find the questions too demanding and
therefore decide to opt out
 Contingency Questions

• These are follow up questions that only apply

to certain groups as per responses given in an
initial/opening question
• The purpose of the questions is to probe for
more information
• They also simplify the task of the respondents
in that they can skip questions that are not
relevant to them
 Contingency Questions
• Contingency questions may be assume different formats:
– Format 1
• Please put a tick in the box next to the right response
• Do you smoke? 1. Yes □ 2. No □
• If yes, how many do you smoke in a day?
• 1. One □ 2. 2 -5 □ 3. More than 5 □
– Format 2
• Have you ever used hormonal contraceptives?
• □ Yes ( please answer questions 3-5)
• □ No (please skip questions 3-5 and go directly to question 6)
– Note: Contingency questions can either be closed ended
( as above) or open ended
 Matrix questions
• These are questions that have a range of possible response
categories and for which a scale is applied to evaluate each
of the given statements
• There are several scales are used to measure perception,
attitude, values and behavior
– The most commonly used scale is the Likert scale
• The items/statements used in Likert scales are usually in the declarative
form.
• In these scales, numbers are assigned to categories of possible
responses in order of agreement or disagreement with given statements
• Note: the numbers are labels, and do not have any numerical
quantitative significance
 Matrix questions
• Example of matrix questions and Likert scale
– …Please circle the number that best describes your
views about how you felt satisfied by the services
provided in hospital X…
“The services provided in the hospital met my
expectations”:
• 1 = Strongly disagree
• 2 =Disagree
• 3 = Neutral
• 4= Agree
• 5= Strongly agree
 Guidelines in designing questionnaire items

• 1. Before constructing a questionnaire, list the

objectives that you want the questionnaire to
achieve i.e.
– Define the information that is required from the
questions as per the study objectives
• 2. Determine how information obtained from
each questionnaire item will be analyzed
– For a quantitative study, formulate the variables to
be derived from each of the questionnaire items
and specify its operational definition and method
of analysis
 Guidelines in designing questionnaire items

• 3. Formulate questions clearly to bring out the

required information without ambiguity;
– In general, it is a good idea to search for, and to critically
review, how others have asked specific questions,
especially if these questions have been formally validated.
• 4. Avoid long questions, but vary the length of
questions to avoid administration of the questionnaire
becoming repetitive and boring for the interviewer or
interviewee
– short questions are preferred over long ones; short
questions are easier to understand
 Guidelines in designing questionnaire items

• 5. Avoid jargon and sophisticated language; assessing

understanding at the pre-test and pilot test stages is
essential.
– The wording of all questions must be appropriate to the
educational, social, and cultural background of the
respondents.
• 6. It may be necessary in some cases to define a term
or a concept before asking about it.
– If the definition is short, it can be included in the question,
but otherwise it is better given separately before the
question is asked.
 Guidelines in designing questionnaire items

• 7. Avoid double barreled items.

– These are items that have two separate ideas in one
question.
• Keep to a single subject for each question.
• For example, do not say ‘Do the cost and times of the clinic
prevent you going?’.
• 8. Avoid leading or biased questions.
– These are questions that suggest that one answer
may be better than the other; hints as to preferable
answers should be avoided
• For example do not say ‘Do you think the improved clinic
arrangements are better?’
 Guidelines in designing questionnaire items

• 9. Avoid negative questions (for example, ‘Do you not

think…’)
– Negatively stated questions are usually misread by many
respondents; in most cases the negative word is overlooked
• 10. Very personal, sensitive and psychologically
threatening questions should be avoided.
– The respondent may be naturally dishonest in replying to
them
– In some cases such questions may also make the
respondent uncomfortable and even decide to opt out of
the study
 Organization of a questionnaire
• Items in a questionnaire should be arranged in a logical
sequence
• The questionnaire should have an introductory statement
capturing the general purpose and scope of the items
included
• The following are important tips on how to organize a
questionnaire:
– 1. Begin with non threatening, non intrusive items
– 2. Important questions should be put in the initial sections of the
questionnaire
– 3. Have a logical sequence when putting items together;
• Related items can be put into thematic sections derived from the
objectives of the study
Organization of a questionnaire
– 4. If the questionnaire is arranged into content sub
sections, have a short introductory statement for
each section
– 5. Personal data ( socio-demographic information
e.g. age, education, etc.) is usually put in the initial
section of the questionnaire
– 6. Each questionnaire should be given an
identification number
 Presentation of the questionnaire
• 1. Make the questionnaire attractive by using good
quality paper to make it inviting to the respondents
• 2. Adopt a good layout and structure of the
questionnaire to make it readable without straining
– Leave enough space for responses; avoid overcrowding
questions in one area which may make it difficult to read
and write responses
• 3. All pages and questions in the questionnaire
should be numbered; this makes it easy to follow
 Presentation of the questionnaire
• 4. Provide brief but clear instruction on how responses are to be
given;
– General instructions can be given for items with a similar format
– Some questions may require specific instructions
• 5. Make the questionnaire short; questionnaires that are too
long may discourage respondents and lead to high non response
rate
– However the key information as required in the study objectives
should not be left out
– Also do not squeeze in questions in order to make the questionnaire
short or save on space
• 6. Depending on the educational level of the respondents, it may
be necessary to translate questionnaires into a language that is
understandable to them
 Pretesting of the questionnaire
• Once the questionnaire has been finalized, a
field trial should be carried out before actual
implementation of the study
– This activity is called pretesting of the
questionnaire
Pretesting of the questionnaire
– The pretesting is done on a selected sample with
similar characteristics to the actual sample to be
used in the study;
• The sample for pretesting should not be drawn from the
same site where the actual study participants will be
recruited from
• Procedures used in pretesting the questionnaire should
be identical to those to be used during the actual data
collection
• The number of subjects for pretesting should not be too
large- usually between 1-10% of the actual sample size
Pretesting of the questionnaire
• Larger sample sizes should have smaller percentage to
avoid having too many subjects for pretesting which
may lead to higher costs
• During pretesting, subjects should be encouraged to
make comments and suggestions regarding instructions,
clarity of questions and relevance
 Importance of pretesting questionnaires

• 1. Allows the researcher to identify questions

that are vague and rephrase them accordingly
• Once any ambiguity or vagueness is detected, the
questions are rephrased and fine tuned until they
convey the same meaning to all subjects
• Improving the formulation of questions enhances the
validity of the questionnaire
– The validity of a research instrument is the extent to which it
accurately measures the characteristic or variable under study
 Importance of pretesting questionnaires

• 2. Pretesting allows for identification of inconsistencies and

deficiencies in the questionnaire;
– For example, unclear instructions, insufficient space to write
responses, cluttered questions and wrong numbering can be
revealed and corrected thus improving the questionnaire
– Comments and suggestions made by respondents can also be
incorporate dto enhance the questionnaire
• 3. Pretesting of research instrument facilitates planning for
activities to be undertaken in actual implementation of the
study;
– e.g. Time taken for completing a questionnaire can be used in
estimating the number of data collection assistants and the time
required for data collection
 Methods of administering questionnaires
• Questionnaires in research may be administered
via various methods:
– 1) Self administered questionnaires
• Respondents are given questionnaires to complete by
themselves;
• The questionnaires may be administered via different
modalities:
– hand delivered in paper form, electronic copy sent via email or
through a web page, send through mobile phone technology e.g.
sms
• This method is applicable in cases where the respondents
comprise of literate people with ability to read and write
Methods of administering questionnaires

– 2) Researcher/ Interviewer administered

questionnaires
• The researcher uses the questionnaire to interview the
respondents
• The questionnaire can be in paper form or electronic
formats
• This is approach is adopted in cases where the
respondents have no capacity to read and write for
themselves (illiterate)
 Analysis of data from questionnaires
• At the proposal stage, all key variables for the study
should be formulated and the specific data collection
tools and data analysis methods also described
• If data is to be collected using a questionnaire, then each
of the items/questions should be translated into a variable
– Some variables may be formulated from a combination of items
depending on the operational definition of a given variable
• A data analysis template should be prepared with
variables and their operational definitions
– For categorical variables the coding to be adopted in data entry
should be specified per variable
Qualitative Research Methods
 Introduction
• Qualitative approaches to data collection usually
involve direct interaction with individuals on a one
to one basis or in a group setting.
– These methods of data collection are time consuming
and consequently data is collected from smaller
numbers of people than would usually be the case in
quantitative approaches such as the questionnaire
survey.
• The benefits of using these approaches include
richness of data and deeper insight into the
phenomena under study.
 Introduction

• Unlike quantitative data, raw qualitative data

cannot be analyzed statistically.
– The data from qualitative studies often derives from
face-to-face interviews, focus groups or observation
and so tends to be time consuming to collect.
• Samples are usually smaller than with
quantitative studies and are often locally based.
• Data analysis is also time consuming and
consequently expensive
 Introduction
• The main methods of collecting qualitative
data are:
– Focus group discussions
– Qualitative interviews
– Observation
 Focus Group Discussions
• Sometimes it is preferable to collect information from groups
of people rather than from a series of individuals
– A focus group discussion (FGD) is a group discussion of a selected
group of people ( usually 6-12 persons) guided by a facilitator,
during which group members talk freely and spontaneously about
a particular and pre-determined topic
• Focus groups can be useful to obtain certain types of
information or when certain circumstances would make it
difficult to collect information using other methods of data
collection.
– They can be a useful way of bringing to the surface issues that
might not otherwise have been discovered:
• the dynamics of a group can often make people bolder in advancing their
opinions.
 Focus Group Discussions
• The purpose of an FGD is to obtain in-depth
information on concepts, perceptions, and
ideas of the group.
– It aims to be more than a question-answer
interaction.
 Conducting a focus group discussion
• Recruitment of participants:
– Participants should be roughly of the same
socioeconomic group or have a similar background
in relation to the issue under consideration.
– The age and sex composition of the group should
facilitate free discussion.
 Conducting a focus group discussion
• Physical arrangements:
– Communication and interaction during the FGD
should be encouraged in every possible way.
– Arrange the chairs in a circle.
– Make sure the area will be quiet, adequately lighted,
etc., and that there will be no disturbances.
• Preparation of a discussion guide:
– There should be a written list of topics to be covered.
– It can be formulated as a series of open-ended
questions.
 Conducting a focus group discussion
• During the discussion:
– One of the members of the research team should act as a "facilitator"
for the focus group. One should serve as "recorder.“
• Functions of the facilitator:
– Introduce the session
– Encourage discussion
– Encourage involvement
– Listen carefully and move the discussion from topic to topic.
– Subtly control the time allocated to various topics so as to maintain
interest and ensure all areas/objectives are covered
– Take time at the end of the meeting to summarize, check for
agreement and thank the participants.
• In general, the facilitator should not act as an expert on the topic.
– His or her role is to stimulate and support discussion
 Report writing in focus group discussions

• Start with a description of the selection and

composition of the groups of participants and a
commentary on the group process,
– so the reader can assess the validity of the reported
findings.
• Present your findings, following your list of topics
and guided by the objective(s) of your FGD.
• Include questions whenever possible, particularly
for key statements
 Further reading
• Further reading required on other methods in
qualitative research
– Interviews
– Observations
– Others
 Choosing an appropriate method for data collection

• The method of data collection chosen for a study

should be appropriate for the type of information
required.
• Whether the required information is quantitative
or qualitative in nature is the major consideration
• In some cases, a study may be designed with a
combination of both quantitative and qualitative
methods
– This approach is referred to as mixed methods design