How to Build a Data-Friendly Questionnaire That Supervisors Would Approve
Did you know that wrong questions = wrong data?
Introduction
When I designed the instrument for “Robot Journalism Revolution and Adoption: Dynamics and Dilemmas in Nigeria”, it took a pilot test, two revisions, and close review to get a tool that both collected the right data and satisfied academic scrutiny. Here's how you can do the same — without the back and forth.
1. Let Your Objectives Drive Your Questions
Every question must link back to one of your research objectives. Otherwise, you’re just filling pages.
🧠 Example: If your objective is “To assess the perception of journalists on AI tools,” then don’t ask them about social media memes or cryptocurrency.
2. Avoid Compound or Confusing Questions
🚫 “Do you think robot journalism is ethical and will it affect employment?”
✅ No.. instead, split it into two separate, clear questions.
3. Localise Language Without Losing Academic Tone
In my Kano-based study, imagine asking market women:
🚫“How frequently do you interface with social media platforms?”
✅ Instead, say:“How often do you use WhatsApp or Facebook on your phone?”
4. Use Tested Scales
Likert scales remain the gold standard. Try:
-
Strongly Agree
-
Agree
-
Neutral
-
Disagree
-
Strongly Disagree
✅ Don’t forget to reverse some items to catch patterned responses.
5. Always Pilot Before Full Rollout
Before administering your tool in Jos, Kano, or Lagos, test it with 5–10 participants. You’ll be shocked at the clarity issues you didn’t see.
Conclusion
A smart questionnaire saves you hours of data cleaning and Chapter Four editing. As someone who has tested questionnaires for multiple peer-reviewed studies, I can tell you — clarity is better than complexity.
📩 DM “QUESTIONNAIRE FIX” to submit your draft for a review.
Leave us a comment and visit our YouTube Channel to watch more interesting tutorials on thesisprojectsng
REMEMBER TO SHARE THIS POST IF YOU FOUND IT HELPFUL
.png)
Comments
Post a Comment