Research Methods guide, Unit 2, step 8 of 16Guide overview Next step

Step 8: Designing Questionnaires and Surveys

By the EconTutorials team. 12-minute read with 6 practice questions.

MatchesSurvey-based dissertations · IB Extended Essay and A-level NEA primary data · market research projects

In 30 seconds

Start from your objectives: every question should map to an objective or a variable you will analyse. If it does not, cut it.
Use closed questions (tick boxes, rating scales) for things you will count and test; open questions for reasons and ideas you will code.
Avoid leading, double-barrelled, loaded and vague questions, jargon, and overlapping answer options.
Likert scales (strongly disagree to strongly agree) work best as several items averaged into one score.
Always pilot with a few people like your participants, then revise before the real launch.

Video lesson coming soonA short animated lesson for this step will appear here.

Plan before you write

The quickest way to a useless survey is to start writing questions. Instead, make a table with three columns: objective, variable, question. When you have finished, check that every objective has at least one question and every question serves an objective.

Objective	Variable	Question
Measure paid work	Weekly hours worked (ratio)	“In a typical term-time week, how many hours do you work in paid employment? (enter a number; 0 if none)”
Compare marks	Average mark last year (ratio)	“What was your overall average mark last academic year (%)?”
Control for background	Year of study (ordinal)	“Which year of study are you in? 1 / 2 / 3 / 4+”
Measure stress	Stress scale score	Five Likert items (Step 7)

Chapter 8 of Research Methods Made Simple covers this in full, with the complete companion student questionnaire and a pilot checklist.Get the textbook · $15

Question types

Type	Example	Use for
Dichotomous	Do you have a paid job? Yes / No	Simple facts
Multiple choice	Main type of job: retail / hospitality / tutoring / other (please state)	Categories
Numeric	Hours per week: ____	Exact amounts
Likert item	“My job leaves me too tired to study.” 1 strongly disagree … 5 strongly agree	Attitudes and constructs
Ranking	Rank these reasons for working from 1 to 4	Priorities (keep lists short)
Open	“What, if anything, helps you balance work and study?”	Reasons, ideas, surprises

Writing good questions

Problem	Bad	Better
Leading	Don’t you agree that long shifts harm your studies?	How, if at all, do your shifts affect your studies?
Double-barrelled	Is your job well paid and flexible?	Two questions: one on pay, one on flexibility
Vague	Do you work a lot?	How many hours did you work last week?
Overlapping options	0–10 / 10–20 / 20–30 hours	0–9 / 10–19 / 20–29 / 30 or more
Loaded / sensitive	How much do you earn?	Use bands, and offer “Prefer not to say”
Jargon	Does work cause role conflict?	Do work and study ever demand your time at the same moment?

Worked example

Fix this question: “How often do you skip lectures because of your job and do you regret it? Never / Sometimes / Often”.

Problem 1Double-barrelled: it asks about skipping and regret at once.

Problem 2Vague options: “sometimes” means different things to different people.

FixQ1: “In the last four weeks, how many lectures did you miss because of paid work? 0 / 1–2 / 3–5 / 6 or more.” Q2 (only if Q1 > 0): “How much do you agree: ‘Missing these lectures has affected my marks’?” 1–5.

Order, length and layout

Open with easy, neutral questions; put sensitive questions and demographics near the end.
Group questions by topic, with short headings.
Use skip logic so people without a job do not see job questions.
Keep it short: under 10 minutes for most student surveys. Response and quality fall as length rises.
Check it works on a phone; most respondents will use one.

Piloting

Give the draft to five to ten people like your participants. Time them, then ask: Was anything unclear? Did any answer options not fit you? Was anything uncomfortable? Then look at their answers: if everyone chose the same option, the question may not discriminate. Revise, and note the changes in your methodology.

In your dissertationPut the full questionnaire in an appendix, describe the pilot and what it changed, and report the response rate and how you handled incomplete responses. Examiners look for all three.

Common mistakeCollecting hours as bands (0–9, 10–19…) when you plan a correlation or regression. Bands turn a ratio variable into an ordinal one and lose information. Ask for the number; you can always group it later, but you cannot ungroup it.

Practise until it sticks: Research Methods Made Simple Practice Questions has 1,004 questions, each with a worked answer.Get the practice book · $9

Practise: 6 questions

Try each one on paper first, then check the worked answer.

EasyQ1. What is wrong with: “Do you find your lecturers helpful and approachable?” Show worked answer

It is double-barrelled: helpful and approachable are different qualities. Split it into two questions.

EasyQ2. Give one advantage of closed questions and one of open questions. Show worked answer

Closed: quick to answer and easy to count and test statistically. Open: capture reasons and unexpected answers in respondents’ own words.

MediumQ3. Fix the answer options: “Age: 18–20 / 20–25 / 25–30 / over 30”. Show worked answer

They overlap at 20, 25 and 30. Use 18–20 / 21–25 / 26–30 / 31 or over (and “under 18” if relevant).

MediumQ4. Rewrite the leading question: “Most students say working improves their CV. Has your job improved yours?” Show worked answer

Remove the social pressure: “How, if at all, has your paid work affected your CV?” or a neutral Likert item: “My paid work has improved my CV” (1 strongly disagree to 5 strongly agree).

MediumQ5. Why should demographic and income questions usually go at the end? Show worked answer

They can feel intrusive. Placed first, they may put people off before they engage; placed last, respondents have already invested time and are more likely to answer or skip them without abandoning the survey.

ExamQ6. Explain why piloting a questionnaire is important and what you would do in a pilot. [5] Show worked answer

Piloting finds unclear wording, missing options and technical problems before real data are wasted [1]. Test it on 5–10 people like the target group [1]; time how long it takes [1]; ask what was unclear or uncomfortable [1]; check answers for questions everyone answers the same way, then revise and record the changes [1].