How to Make a Thesis Questionnaire for a BEEd or BSEd Study, Item by Item (2026)

Education theses in the Philippines are questionnaire-heavy for a simple reason: the things BEEd and BSEd students study — teaching strategies, classroom management practices, learners’ engagement, the implementation of a curriculum guide — mostly live inside people’s heads and have to be measured by asking. That makes your instrument the single most examined part of your paper. Your adviser will read it before your Chapter 2, and the panel will open your appendices before they open your abstract.

Instruments come back for revision for three recurring reasons. Items do not trace back to a specific sub-problem in your statement of the problem. The scale has numbers but no verbal interpretation, so nobody can tell what a computed mean of 3.41 means. And there is no evidence anyone qualified checked the items before you distributed them. Every step below exists to close one of those three gaps.

One standing caveat: your college’s research manual outranks this article. Some SUCs prescribe a fixed profile section, some private universities require a specific validation form with three named signatories, and a few programs mandate a four-point scale to remove the neutral midpoint. Open your manual first and let it overrule anything here.

Step 1 — Break your statement of the problem into item clusters

Do not start by writing questions. Start by copying your numbered sub-problems into a blank page, one per line. Every item you eventually write must sit under one of them.

Suppose your specific problems read:

1. What is the profile of the respondents in terms of age, sex, length of teaching experience, and highest educational attainment?
2. What is the extent of use of differentiated instruction by Grade 3 teachers in terms of content, process, and product?
3. What challenges do the teachers encounter in implementing differentiated instruction?
4. Is there a significant difference in the extent of use when respondents are grouped according to profile?

Expected output of this step: a page with four headings and nothing under them yet. Sub-problem 1 becomes Part I of your instrument. Sub-problem 2 becomes Part II with three sub-clusters. Sub-problem 3 becomes Part III. Sub-problem 4 needs no items at all — it is answered by comparing Part I against Part II statistically, which is a job for your statistical treatment section rather than your instrument.

That last point catches out a lot of groups. Questions about relationships and differences are answered by tests, not by asking respondents whether they think there is a difference.

Step 2 — Decide what kind of data each cluster needs

Go back through the clusters and label each with the data it will produce, because that decides both the item format and the statistic.

1. Part I (profile) — nominal and ordinal data. Checkboxes and blanks. Analysed with frequency and percentage.
2. Part II (extent of use) — interval data by convention. Likert items. Analysed with weighted mean and standard deviation.
3. Part III (challenges) — either a second Likert block on severity, or two or three open-ended items. Decide now, not later; an open-ended item turns your study into a mixed-methods design and your Chapter 3 has to say so.

Expected output: each cluster tagged with a format and a statistic. If you cannot name the statistic, the cluster is not ready.

Step 3 — Write Part I: the respondent profile

Only ask for a profile variable if you will actually use it. If sub-problem 4 groups respondents by length of service and educational attainment, those two must be there. Asking for civil status when no sub-problem touches it gives the panel an easy question you cannot answer.

Keep categories mutually exclusive and exhaustive. The most common error in education theses is overlapping service brackets — 1–5, 5–10, 10–15 — which leaves a teacher with exactly five years unable to answer. Write them as 1–5, 6–10, 11–15, 16 and above.

Do not ask for a name. Write (Optional) beside the school name field, and put a one-sentence confidentiality note at the top of the page. Under the Data Privacy Act, personal information you never collect is personal information you never have to protect.

Step 4 — Write the Part II Likert items

Rules that survive validation:

1. One behaviour per item. “I use varied materials and group learners by readiness” is two items pretending to be one. A respondent who does the first but not the second cannot answer honestly.
2. Write statements, not questions. Part II items are declarative: “I adjust the difficulty of tasks based on learners’ readiness levels.”
3. Start with a verb the respondent actually performs. Observable classroom behaviour beats abstraction. “I provide tiered activities” is measurable; “I believe in differentiation” is not the same construct.
4. Avoid absolutes. “Always,” “never” and “all learners” push responses to the extremes.
5. Six to ten items per sub-cluster. Fewer than five and your Cronbach’s alpha will be unstable. More than twelve and teachers stop reading by the third page.

Filled-in template you can adapt

Part II. Extent of Use of Differentiated Instruction

Direction: Below are statements describing practices in differentiated instruction. Please indicate the extent to which you use each practice in your Grade 3 class by putting a check mark (✓) under the number that corresponds to your answer. There are no right or wrong answers.

A. Content

1. I adjust the difficulty of reading materials based on learners’ reading levels.
2. I present the same lesson objective using more than one type of material.
3. I prepare alternative examples for learners who need additional support.
4. I use pre-assessment results to decide which content each group receives.
5. I provide advanced materials for learners who master the objective early.
6. I relate the content to the learners’ community context.

B. Process

1. I group learners according to their readiness for the day’s objective.
2. I vary the amount of time given to different groups to finish a task.
3. I use tiered activities within a single lesson.
4. I give differentiated guide questions to different groups.
5. I conduct small-group instruction while other groups work independently.

Part C on product follows the same pattern. Notice that every item is a single observable act, phrased so a teacher can picture doing it this morning.

Step 5 — Set the scale, the verbal interpretation and the mean ranges

A Likert block without a verbal interpretation table is an incomplete instrument. Print the scale on the questionnaire itself, and print the interpretation ranges in Chapter 3.

For a four-point extent scale:

• 4 — Always Used
• 3 — Often Used
• 2 — Sometimes Used
• 1 — Never Used

The corresponding ranges, computed as (4 − 1) ÷ 4 = 0.75 per interval:

• 3.26 – 4.00 — Always Used — Very High Extent
• 2.51 – 3.25 — Often Used — High Extent
• 1.76 – 2.50 — Sometimes Used — Low Extent
• 1.00 – 1.75 — Never Used — Very Low Extent

Two things panels check here. First, that your ranges have no gaps and no overlaps — 3.25 and 3.26 must be adjacent, not 3.25 and 3.30. Second, that your adjectives match your construct. A scale measuring extent of use takes usage adjectives, not agreement adjectives. Mixing “Strongly Agree” into a usage scale is a guaranteed revision.

Whether you use four points or five is a real decision, not a default. Four points forces a direction and is common in Philippine education research; five allows a genuine neutral. Your manual or your adviser usually settles it. When you get to writing up the numbers, the same ranges carry straight into your tables — the mechanics of that are covered in the walkthrough on presenting results table by table.

Step 6 — Adapt rather than invent, and cite the adaptation

An instrument you built from scratch has no established validity. An instrument you adapted from a published study inherits some, and your Chapter 3 gets a much stronger paragraph. Search for existing scales in your area before writing your own items, and when you find one, adapt it and say so.

The Chapter 3 sentence looks like this:

The instrument was adapted from the Differentiated Instruction Implementation Scale developed by Santos and Reyes (2021), with permission from the authors. Of the original 28 items, 18 were retained and reworded to fit the Grade 3 setting of the present study.

And the reference entry, in APA 7th:

Santos, M. L., & Reyes, A. B. (2021). Development and validation of a differentiated instruction implementation scale for elementary teachers. Philippine Journal of Educational Research, 14(2), 45–63. https://doi.org/10.xxxx/pjer.2021.14.2.45

Note what the citation does not do: it does not excuse you from validating the adapted version. Once you reword items or change the setting, the original validity statistics no longer strictly apply, which is exactly why Steps 7 and 8 exist. Where to look for those source instruments is its own problem — the roundup of databases that actually hold Philippine studies is the faster route than Google Scholar alone.

Step 7 — Have three experts validate the content

Standard practice across Philippine education programs is three validators: typically a content expert in your area, a language or measurement expert, and a practitioner such as a master teacher or department head.

1. Prepare a validation packet: your statement of the problem, the draft instrument, and a rating sheet.
2. The rating sheet asks each validator to score every item, or every section, on criteria such as clarity, relevance to the objectives, and appropriateness of language — usually on a 1-to-4 or 1-to-5 scale, with a comments column.
3. Collect the sheets, compute the mean rating, and revise every item a validator flagged.
4. Keep the signed sheets. They go in your appendices, and the panel does look for the signatures.

Expected output: a revised instrument plus three signed validation sheets and a sentence in Chapter 3 reporting the overall validity rating and its descriptive interpretation.

Step 8 — Pilot test and compute reliability

Administer the revised instrument to 20 to 30 respondents who resemble your actual respondents but who will be excluded from the final sample — a neighbouring school, a different section, another district.

Encode the pilot data and compute Cronbach’s alpha per sub-cluster, not for the whole instrument at once. Convention treats 0.70 as the working floor, with 0.80 and above considered good. If a sub-cluster falls below 0.70, look at the item-total correlations, find the item pulling the scale apart, and either rewrite or drop it.

Report it plainly: The instrument yielded Cronbach’s alpha coefficients of .87 for Content, .84 for Process and .81 for Product, indicating good internal consistency. If you have never run this before, the step-by-step on analysing thesis data in SPSS covers the reliability procedure alongside the descriptive runs you will need later.

Step 9 — Secure the permissions education research needs

Education theses have an approval chain other programs do not. If your respondents are public school teachers or learners, you will normally need, in order: your adviser’s endorsement, a letter from your dean or research office, approval from the Schools Division Superintendent, and then the school head’s permission. If learners are respondents, add parental consent forms and learner assent.

Start this chain the week your instrument passes validation, not the week you plan to distribute. Division-level approval routinely takes two to three weeks, and it is the single most common cause of a data-gathering timeline collapsing.

Step 10 — Format the final instrument for the appendix

The version that goes in your appendices should carry a letterhead or title block, the confidentiality and consent statement, clear directions per part, the scale legend printed above the item table, and numbered items with response boxes. Label it Appendix B — Research Instrument or whatever your manual specifies, and make sure the item numbering in the appendix matches the item numbering in your Chapter 4 tables exactly.

Decide your distribution mode at this point too, because it changes the layout. A printed form and a Google Form are not the same document, and the trade-offs — including response caps and the Data Privacy Act implications — are laid out in the comparison of Google Forms versus paper for Philippine thesis data collection. Plan too for what happens when forms come back half-answered; the guide on handling incomplete questionnaires is worth reading before you distribute, not after.

Drafting the instrument section without starting from a blank page

The items are yours — they have to be, because only you know your sub-problems and your respondents. What slows most groups down is the surrounding prose: the instrument description, the validation paragraph, the reliability write-up, the scale interpretation table, and the APA 7th adaptation citation, all in the register a Philippine panel expects.

Tesify drafts those sections from your own statement of the problem and your own validation results, keeps your references formatted in APA 7th as you go, and leaves the research decisions where they belong. Start your thesis on Tesify and get the instrument section written while the validation sheets are still coming back.

Frequently asked questions