1. 11 Tips on How to
Collect Good
Quality Primary
Data During
Research
By Collins Owuor
2. Data quality is a measure of the condition of data
based on factors such as accuracy, completeness,
consistency, reliability and whether it's up to date.
(https://www.techtarget.com/searchdatamanage
ment/definition/data-quality)
Here are some of the tips to collect good quality
research data in the field
3. An ambiguous question is a question which is not
specific. For instance, asking a respondent, "Do you
watch the videos?". The question is not clearly
defining the subject (videos). Someone would ask,
"what videos?"
Avoid ambiguous questions in your survey
tools.
1.
4. Leading questions give the respondents a clue of the
answers the researcher expects.
Example: What improvements have you observed in
your hospital since we trained your staff on Triage? In
this question, the researcher already expects an
improvement, which might not be the case.
The questions could read, "what changes have you
observed...."
2. Avoid leading questions in your survey tools.
5. Bias in research is any deviation from the truth
which can cause distorted results and wrong
conclusions.
Sampling bias occurs when some sampling units
have higher chances of being selected and surveyed
than others.
Probability sampling methods can be used to avoid
sampling bias.
3. Avoid sampling bias
6. Remember you have to build a trust between you
and the study participant. They may give you
inaccurate information if they don't trust you with
their information. Therefore, it is important to create
a good rapport by creating friendly interview
environment and informing the participants about
your study.
4. Create a friendly interview environment
and give informed consent
7. a) Your name and cadre
b) Organization you are working for/ learning at
c) Purpose of the study
d) Motivation, risks and benefits of the study
e) Assurance that there will be no name identification
and that their information will be treated with utmost
confidentiality.
f) Making them understand that they have a right to
chose whether to participate.
g) The estimated length of interview
A good introduction and informed consent
should have the following information:
8. Ethically, there should be no name identification of
participants in a study to protect their identity.
5. Anonymize participants
9. You may be interested in collecting confidential
information from participants. If you interview a
respondent in a public environment, there is a high
chance he/ she will either decline to participate in
your study or give you inaccurate information.
6. Interview participants in private
environments
10. It is good to make it possible for participants who
only understand their local languages to understand
the questions. You may need qualified interpreters to
assist you interpret the questions to the respondents.
7. Develop standard procedures for
interpreting questions in local languages if
necessary.
11. It is a good practice to make data collectors
understand the standard procedures of data
collection. The data should be conducted in line
with study objectives , ethical and free from bias
8. Train data your collectors
12. It is a good practice to implement all the surveys in
a consistent manner. Ask the exact questions in a
similar manner throughout the survey period to get
consistent results.
9. Be consistent in how you ask the survey
questions (Reliability).
13. Data are valid when they accurately represent what
you intend to measure. Therefore the questions
should represent the objectives of the study.
10. Design and ask the survey questions in
line with the study objectives (Validity).
14. 11. Conduct a pilot/ prior study with few
respondents to test the accuracy, reliability
and validity of your tools