Most instrument advice tells you how to choose a questionnaire or how to validate one. This page is about building one: the actual items, written out, for four different Nigerian departments, plus a complete semi-structured interview guide for a qualitative study. Adapt them to your own variables rather than copying them.
Two site guides sit on either side of this one. Designing and validating a questionnaire for a public health project is the ten-step procedure with the validation route. Which validated instrument to use for an education project covers the case where you should adopt a published instrument instead of writing one. Use a published instrument where one exists for your construct; write your own where it does not.
The four rules that decide whether your items survive validation
- One idea per item. “Assessment records are kept regularly and used to plan lessons” is two claims. A respondent who keeps records but does not use them cannot answer it, and a validator will delete it.
- Match the response format to what you are measuring. A practice that happens or does not is a checklist or a frequency item, not an agreement scale. Agreement scales are for opinions and perceptions.
- Every item serves a numbered objective. Write the objective number next to each item on your working draft. Items that serve none are the ones validators cut.
- No leading wording, no negatives inside a scale item. “Do you agree that the current system is inadequate?” tells the respondent the answer. “I do not find the records unhelpful” cannot be scored reliably.
How to read the specimens
They are shapes, not claims. Bracketed italics mark what you replace. Nothing here is drawn from a published instrument, so nothing needs permission — but equally, nothing here has been validated. Your own face and content validation, and your own pilot, still have to happen, and your Chapter Three still has to report what changed.
Section A: the part every questionnaire shares
Section A collects respondent characteristics. Keep it short, and include a demographic item only if an objective uses it. Every unused item is a line a panel can ask you to justify.
SECTION A: RESPONDENT CHARACTERISTICS
Instruction: Please tick (✓) the option that applies to you.
- Sex: ☐ Male ☐ Female
- Highest academic qualification: ☐ NCE ☐ B.Sc / B.A / B.Ed ☐ PGDE ☐ M.Sc / M.Ed ☐ Other (specify) __________
- Years of experience in [role]: ☐ Below 5 ☐ 5–10 ☐ 11–15 ☐ 16 and above
- [Department-specific item, e.g. subject taught, unit, ward, or crop grown]: __________
- Have you attended any training on [the topic of your study] in the last three years? ☐ Yes ☐ No
Three things to notice. Experience is banded rather than open, because bands are easier to analyse and respondents answer them more honestly. Qualification options are the ones that actually occur in Nigerian institutions, with an “other” escape. And item 5 exists only because a specific objective examines the relationship between training and practice — without that objective, it comes out.
Education: instructional use of assessment records
SECTION B: RECORDS KEPT
Instruction: Tick (✓) all the records you personally keep for the classes you teach.
- ☐ Continuous assessment score sheet
- ☐ Mark register for assignments
- ☐ Record of test items administered
- ☐ Notes on individual pupils needing remediation
- ☐ Class attendance register
- ☐ Other (specify) __________
SECTION C: INSTRUCTIONAL USE OF RECORDS
Instruction: Indicate how far you agree with each statement. SA = Strongly Agree (4), A = Agree (3), D = Disagree (2), SD = Strongly Disagree (1).
S/N Item SA A D SD 6 I look at the previous assessment scores before planning a new topic. 7 I change how I teach a topic when assessment scores on it are low. 8 I identify specific pupils for extra help using assessment records. 9 I discuss individual assessment results with the pupils concerned. 10 I keep assessment records mainly to satisfy school returns. SECTION D: CONSTRAINTS
Same scale as Section C.
S/N Item SA A D SD 11 My class size makes it difficult to act on individual assessment results. 12 There is no period in the timetable for remedial work. 13 The volume of administrative returns leaves little time for record analysis. 14 I have not received training on how to use assessment data instructionally.
Why this works. Section B is a checklist because keeping a record is a fact, not an opinion; Section C is a scale because use is a matter of degree. Item 10 is the diagnostic one: it separates the measurement conception from the feedback conception of assessment, which is exactly the distinction a well-built conceptual review would have adopted. And Section D’s items are each a named constraint rather than a general “I face challenges”, so the analysis produces a ranked list rather than a single number.
Business, accounting or marketing: a perception-plus-behaviour instrument
SECTION B: AWARENESS AND USE
- How often do you use [the service, channel or practice under study]? ☐ Daily ☐ Weekly ☐ Monthly ☐ Rarely ☐ Never
- How did you first learn about it? ☐ Staff of the organisation ☐ Friend or family ☐ Radio or television ☐ Social media ☐ Other __________
SECTION C: PERCEIVED [CONSTRUCT, e.g. SERVICE QUALITY]
SA (4), A (3), D (2), SD (1).
- Staff attend to my request within a reasonable time.
- Staff explain charges clearly before they are applied.
- The service works as described when I need it.
- I can reach someone who resolves a problem without being transferred repeatedly.
- I would recommend [the organisation] to someone in my position.
SECTION D: [OUTCOME, e.g. INTENTION TO CONTINUE]
- I intend to continue using [the service] over the next twelve months.
- I have considered moving to a competitor in the last six months.
- [Charge, price or cost] is the main factor in that decision.
Why this works. Section B captures behaviour with a frequency scale, Section C captures perception with an agreement scale, and Section D captures the outcome variable separately — which is what lets you test the relationship your objectives promised instead of correlating a construct with itself. Item 12 is deliberately the last perception item, because a general recommendation item placed early contaminates the specific ones after it.
Agriculture: a practice-adoption instrument
SECTION B: FARM AND PRACTICE PROFILE
- Farm size under [named crop] in the last season: ☐ Under 1 ha ☐ 1–2 ha ☐ 2.1–5 ha ☐ Above 5 ha
- Source of planting material last season: ☐ Own saved ☐ Local market ☐ Agro-dealer ☐ Extension or government programme ☐ Other __________
SECTION C: COMPONENTS OF THE RECOMMENDED PACKAGE
Instruction: For each component, tick one column.
S/N Component Used fully Used partly Not used 8 Recommended improved variety 9 Recommended spacing 10 Recommended fertiliser type and rate 11 Recommended weeding regime 12 Recommended harvesting and storage method SECTION D: REASONS FOR NON-USE
Instruction: For each component you marked “Used partly” or “Not used”, tick the main reason.☐ Cost ☐ Not available when needed ☐ Did not work when I tried it ☐ Labour requirement ☐ Not aware of it ☐ Other __________
Why this works. The three-column format in Section C measures partial adoption, which a yes/no item destroys — and partial adoption is the whole point of an adoption study worth doing. Section D is conditional on Section C, which keeps the instrument short. Note the “Did not work when I tried it” option: omitting it forces a farmer with direct negative experience into “Not aware”, and produces a finding that is simply false.
Computer science and engineering: a system evaluation instrument
SECTION B: USER PROFILE
- Role in the current process: ☐ Course lecturer ☐ Examinations officer ☐ Head of department ☐ Clerical staff ☐ Student
- How often do you perform [the task the system supports]? ☐ Weekly ☐ Monthly ☐ Each semester ☐ Rarely
SECTION C: SYSTEM EVALUATION
SA (4), A (3), D (2), SD (1). Completed after using the system.
- I completed the task without needing help from anyone.
- The steps followed the order I use in practice.
- Error messages told me what to do next.
- The system was faster than the current method for the same task.
- I could see who had made each change and when.
- I would use this system in place of the current method.
SECTION D: TASK OUTCOMES (recorded by the researcher, not the respondent)
Task Completed without assistance (Y/N) Time taken Errors observed [Task 1] [Task 2]
Why this works. Section C is perception; Section D is measurement, and it is filled in by you while watching. A system evaluation built only on agreement items measures whether people were polite to you. Pairing it with observed completion, time and error counts gives Chapter Four something to report other than screenshots. Fix the evaluation criteria before you build, not after.
A semi-structured interview guide
For a qualitative study the instrument is a guide, not a questionnaire. It has six to ten main questions, each with prompts you use only if the answer stalls, and it is written so that no question can be answered with yes or no.
INTERVIEW GUIDE: [STUDY TITLE]
Opening (read aloud): Thank you for agreeing to speak with me. I am a final year student in the Department of [department], [university], and I am studying [topic]. There are no right or wrong answers; I want to understand your own experience. The conversation should take about [number] minutes. Nothing you say will be reported with your name or the name of your [school, unit or organisation]. May I record the conversation so that I do not have to write while you speak? You may ask me to stop at any point, and you may decline any question.
- Tell me about your role here, and how long you have been doing it.
Prompts: What does a typical week look like? What has changed since you started?- Walk me through what happens when [the process under study] takes place.
Prompts: Who else is involved? Where does it usually slow down?- Tell me about a time when [the problem under study] happened.
Prompts: What did you do? What happened next? How common is that?- What makes [the desired practice] easier or harder to do here?
Prompts: Which of those matters most? Has anything been tried?- Some people say [a contrasting view from your literature]. What is your view on that?
Prompts: Can you give me an example?- If you could change one thing about [the process], what would it be, and why that one?
- Is there anything I should have asked you about that I did not?
Closing: Thank you. May I contact you again if I need to check something you said?
Why this works. Question 1 is a warm-up that also yields context. Questions 2 and 3 ask for a process and then a specific incident, which produces concrete detail rather than generalities. Question 5 tests your literature against the respondent instead of leading them towards it. Question 7 is the one that regularly produces the best material in the whole study, and it costs one line. Consent is spoken before recording, and the member-checking offer at the close supports your credibility claim later.
Guidance on how many interviews you need, and how to justify the number, is in how many respondents a qualitative project needs, and the tools for handling the recordings are in transcription and qualitative analysis tools.
Before you print
| Check | Why |
|---|---|
| Every item maps to a numbered objective | Unmapped items are what validators delete and panels query |
| No item contains “and” joining two claims | Double-barrelled items cannot be answered or scored |
| Response format matches the construct | Facts take checklists or frequencies; opinions take scales |
| Instructions appear at the head of every section | A section with no instruction produces blank columns |
| Items are numbered continuously across sections | Restarting numbering per section breaks your data entry |
| It fits on the pages you can afford to photocopy | Length is a real constraint and long instruments lower response rates |
| A blank copy is in your appendix | Required in most Nigerian departments, and the panel will look |
Then validate it — typically three experts — pilot it on a comparable group outside your sample, compute reliability per section, and report all of it in Chapter Three. Our guide to what Cronbach’s alpha your department will accept covers the figure and what to do if it is low, and Google Forms versus KoboToolbox covers administering it digitally.
Building an instrument, validating it and writing it up consistently across Chapters Three and Four is where the weeks go. Tesify keeps your objectives, instrument and analysis aligned as you draft, so the instrument you build is the one your chapters describe — 100% written by you, alongside 9,000+ students and 15,000+ chapters.
Build your project with Tesify
Frequently asked questions
How many items should a final year project questionnaire have?
Enough to cover every objective and no more, which for an undergraduate project is commonly 20 to 40 items including Section A. Long instruments lower response rates and raise photocopying costs, and neither helps your findings.
Can I use a questionnaire from another student’s project?
Not without knowing where it came from. If it was copied from elsewhere, you inherit its errors and its permission problem. Where a published, validated instrument exists for your construct, seek permission and adapt it properly; otherwise build your own and validate it.
Four-point or five-point Likert scale?
A four-point scale removes the neutral midpoint and forces a direction, which suits practices and behaviours. A five-point scale allows genuine neutrality, which suits attitudes. State your reason in Chapter Three either way.
Should knowledge questions use a Likert scale?
No. Knowledge has a correct answer, so it takes a multiple-choice or true/false format scored right or wrong. Asking a respondent to agree that they know something measures confidence, not knowledge.
Do I need a separate interview guide as well as a questionnaire?
Only if your design is mixed methods and an objective requires qualitative data. Adding interviews to a purely quantitative study because it sounds thorough gives you transcripts you have no analysis plan for.
How many questions should an interview guide have?
Six to ten main questions with prompts underneath. More than that and the interview becomes an oral questionnaire, which loses the depth that justified the method in the first place.
Does the questionnaire go in the appendix?
Yes, a blank copy, in nearly all Nigerian departments. The appendix copy is also what an examiner reads alongside your Chapter Four tables to check that your items and your findings match.
Do I need ethical clearance to administer a questionnaire?
It depends on your discipline and respondents. Health-related studies and any study involving patients or clinical records generally do. Check with your department before collecting anything, because clearance cannot be obtained retrospectively.
How do I number items across sections?
Continuously, from item 1 in Section A through to the last item of the last section. Restarting at 1 in each section creates duplicate item numbers, which breaks data entry and makes your Chapter Four tables ambiguous.
Can I administer the questionnaire online?
Yes, where your respondents reliably have connectivity and devices. Online administration changes your sampling in ways you must report honestly, because the people who respond online are not a random subset of your population.
What if my validators want major changes?
Make them, then return the corrected instrument to the validators for approval before piloting, and report both the changes and the re-approval in Chapter Three. That record is what turns validation from a claim into evidence.