Building a Research Questionnaire

TL;DR: A research questionnaire is not a list of interesting questions. It is the instrument that answers your specific sub-problems, one item at a time, and a panel will check that mapping directly. The standard Philippine layout is three parts: profile, independent variable, dependent variable, with consent and instructions above them. Validity is normally established by expert review before data collection; reliability by a pilot test and Cronbach's alpha, with 0.70 the common floor and 0.80 to 0.95 preferred. Every item should be defensible with one sentence: "this item answers sub-problem number 3."

If you are drafting a questionnaire, the deadline is probably close and the temptation is to write forty plausible-sounding questions and sort it out later. That is the fastest route to a revision. This guide works through the layout Philippine advisers expect, a complete illustrative instrument built from a statement of the problem, how Likert responses get interpreted, and what "valid and reliable" actually requires you to do.

The instrument is also a graded deliverable in its own right. DepEd's Practical Research 2 curriculum guide lists the competency "constructs an instrument and establishes its validity and reliability" (CS_RS12-IIa-c-3), alongside describing the sampling procedure and planning the data analysis. Those three are assessed together because they are one design decision split into three sections.

Which subject is this for? Practical Research 1 and 2 are applied subjects of the 2016 K to 12 Senior High School curriculum, and that structure is being replaced. DepEd Order No. 017, s. 2026 institutionalises the Strengthened Senior High School Curriculum for Grade 11 from SY 2026-2027 and Grade 12 from SY 2027-2028, and its Annex A lists Research 1 and Research 2 as 80-hour Academic Track electives in the Field Experience cluster rather than applied subjects everyone takes. Grade 12 learners not enrolled in pilot schools continue under the 2016 curriculum for SY 2026-2027. The methodology below is the same either way; what changed is the subject label and who takes it. See our guide to the Strengthened SHS Curriculum.

What a Questionnaire Actually Has to Do

A questionnaire has exactly one job: produce the data that answers your statement of the problem. That gives you a hard test for every draft item.

Take each specific sub-problem, write it at the top of a page, and list the items that will answer it. If a sub-problem has no items, your instrument is incomplete. If an item belongs to no sub-problem, delete it, however interesting it is. Panels ask this question directly, usually in the form "which item answers sub-problem 2?", and the answer needs to be a number, not an explanation.

This mapping is also what protects you in Chapter 4. Every table you present in your results should be traceable back to a numbered item, and every item should end up somewhere in the results. If you collected data you never analyse, you collected personal information you did not need, which is a Data Privacy Act problem as well as a marking one.

The Standard Philippine Three-Part Layout

Part I is where most instruments go wrong. Age, sex, strand, monthly family income, number of siblings and parents' occupation get collected by habit. If your statement of the problem does not ask whether anything differs by number of siblings, that item should not exist.

Worked Example: A Questionnaire Built From Sub-Problems

The study below is an illustrative example constructed for this guide. It is not a real study, and the items have not been validated on real respondents. Use it as a pattern, not as an instrument to copy.

Illustrative study: the relationship between daily smartphone screen time and self-reported sleep quality among Grade 11 students in one public senior high school in Cebu City.

Illustrative sub-problems:

1. What is the profile of the respondents in terms of sex and strand?
2. What is the respondents' average daily smartphone screen time?
3. What is the respondents' level of self-reported sleep quality?
4. Is there a significant relationship between daily screen time and self-reported sleep quality?

Now the instrument, section by section, with the reason each part exists.

Part I. Profile. Two items only, because only two profile variables appear in sub-problem 1.

1. Sex:  ( ) Male  ( ) Female  ( ) Prefer not to say
2. Strand: ______________________

Nothing about family income, because no sub-problem asks about it.

Part II. Daily smartphone screen time. This is the independent variable, and it is measured in hours, not opinions. Asking "do you use your phone a lot?" would produce a number you cannot defend.

3. On a typical school day, what does your phone's built-in screen
   time report show as your total daily average for the past week?
   ______ hours ______ minutes

4. Which figure did you use?
   ( ) iPhone Screen Time weekly average
   ( ) Android Digital Wellbeing weekly average
   ( ) Estimated, no built-in report available

Item 4 exists so that Chapter 3 can honestly describe how the measure was obtained, and so Chapter 5 can list self-estimation as a limitation for the respondents who had no report. That is the kind of detail a panel notices is missing.

Part III. Self-reported sleep quality. This is the dependent variable, measured with a Likert scale because it is a perception, not a count.

Instruction: For the past two weeks, indicate how often each statement
was true. 4 = Always, 3 = Often, 2 = Sometimes, 1 = Never.

5. I fall asleep within 30 minutes of lying down.
6. I sleep through the night without waking.
7. I feel rested when I wake up on a school day.
8. I stay awake and alert during morning classes.
9. I need to nap during the day to keep going.

Item 9 is worded in the opposite direction from the rest on purpose. Reverse-worded items catch respondents who are ticking straight down one column without reading, and they must be reverse-scored before any total is computed. Say so in Chapter 3, or your reliability figure will be wrong.

How Likert Responses Get Interpreted

A mean of 3.42 means nothing to a reader on its own. Philippine research manuals almost always require a stated interpretation range, fixed in Chapter 3 before data collection, not chosen afterwards to make the results look better.

A widely used convention for a 4-point scale:

These boundaries come from dividing the 1-to-4 range into four equal bands of 0.75. They are a convention, not a DepEd rule, and departments do vary. Check your own research manual and use its table if it has one. What matters to a panel is that the table exists, that it was set in advance, and that every mean in Chapter 4 is interpreted against it consistently.

Establishing Validity

Validity asks whether the instrument measures what it claims to measure. For student research the practical route is content validity by expert review.

The usual Philippine practice is three validators: commonly the adviser or a subject expert, a research or statistics teacher, and a language expert who checks that items are unambiguous. Each rates the items for relevance and clarity on a short validation sheet, and you revise, reword or drop what they flag. The signed sheets go into your appendices, and Chapter 3 states who validated the instrument and what changed as a result.

Note the gap between local practice and the methodological literature. Boateng and colleagues (2018), writing in Frontiers in Public Health, recommend five to seven expert judges and quantifying their ratings with a content validity index or similar measure. If your department only requires three, three is what you submit, but knowing the wider standard is useful when a panel member asks why three.

Establishing Reliability

Reliability asks whether the instrument measures consistently. The standard student route is a pilot test: administer the draft instrument to roughly 20 to 30 respondents who resemble your sample but who will be excluded from the actual data collection, then compute Cronbach's alpha for each multi-item scale.

Boateng and colleagues (2018) summarise the benchmarks plainly: "an alpha coefficient of 0.70 has often been regarded as acceptable threshold for reliability; however, 0.80 and 0.95 is preferred for the psychometric quality of scales."

Three practical points that catch students out:

• Report alpha per scale, not per booklet. In the illustrative instrument above, alpha applies to the five sleep-quality items in Part III. Part I is demographic and Part II is a single measured quantity; neither has an internal consistency to compute.
• Reverse-score before computing. Item 9 must be flipped first, or alpha will be artificially low and you will "fix" a problem that is not there.
• Pilot respondents cannot be counted twice. Exclude them from the main sample, and say so in your description of the respondents of the study.

Consent, Minors and Permission to Collect Data

Two separate permissions are involved, and students routinely secure one and forget the other.

Data privacy. Personal information you collect is covered by the Data Privacy Act of 2012 (Republic Act No. 10173). In practice this means a consent statement naming the purpose, confirming that participation is voluntary and may be withdrawn, explaining how responses will be stored and for how long, and confirming that results will be reported in aggregate. If your respondents are minors, parental or guardian consent is normally required in addition to the student's own assent.

Institutional permission. Collecting data inside a school requires a written request to the school head. For public schools, divisions commonly require the request to be endorsed or approved through the Schools Division Office, and DepEd's own research governance is set out in DepEd Order No. 016, s. 2017, the Research Management Guidelines. The exact route, forms and lead time vary by division and by school, so ask your adviser and the target school early. Two weeks of waiting for a signature is a common and entirely avoidable cause of a missed defense date.

Common Mistakes That Cost Marks

• Double-barrelled items. "I sleep well and wake up early" measures two things, and a respondent who does one but not the other cannot answer it.
• Leading items. "Do you agree that excessive phone use harms sleep?" tells the respondent what you want to hear.
• Vague quantifiers. "Often," "regularly" and "a lot" mean different things to different people. Anchor them to the scale, or ask for a number.
• Profile items nobody analyses. Every unused personal question is dead weight and a privacy liability.
• Changing the scale mid-instrument. Four points in Part II and five in Part III makes the results incomparable and looks careless.
• Adopting an instrument without saying so. Cite the original, state your modifications, and re-establish reliability on your own respondents.
• Writing the questionnaire before the sub-problems are final. The instrument is downstream of the statement of the problem, always.

Before You Print It

1. Do the mapping test. Write each sub-problem and list its item numbers. Fix any blanks in either direction.
2. Read every item aloud. Anything you stumble over, a respondent will misread.
3. Check the variables are properly defined. Each concept in the instrument should match the operational definition in your independent and dependent variables section.
4. Confirm the statistics you plan to run are possible with the data you are collecting. Ordinal responses will not support every test; see our guide to analysing data in a student research paper.
5. Confirm the sampling plan first. How many copies you print follows from the sampling technique you chose, not the other way around.
6. Pilot it, then revise it, then validate the revised version. Validating a draft you then change is wasted effort.

Ready to find the right school? Browse schools or compare programs and reviews across hundreds of Philippine schools on SchoolFinderPH.

Disclaimer: departmental research manuals differ on the number of validators, the interpretation ranges, and the format of the instrument. Confirm your own school's requirements before finalising anything in this guide.

Sources

• DepEd — K to 12 Senior High School Applied Track, Practical Research 2 Curriculum Guide — competency CS_RS12-IIa-c-3, constructing an instrument and establishing validity and reliability
• DepEd — K to 12 Senior High School Applied Track, Practical Research 1 Curriculum Guide
• DepEd — DO 016, s. 2017, Research Management Guidelines
• Republic Act No. 10173, Data Privacy Act of 2012 (LawPhil)
• Boateng, Neilands, Frongillo, Melgar-Quinonez and Young (2018), "Best Practices for Developing and Validating Scales for Health, Social, and Behavioral Research: A Primer," Frontiers in Public Health 6:149 — expert-judge counts and Cronbach's alpha thresholds
• SchoolFinderPH — Statement of the Problem Guide
• SchoolFinderPH — Thesis Chapters 1 to 5