Dissertation Questionnaire: How to Design and Use One

A questionnaire looks like the easy route to data: write some questions, send a link, collect answers. In practice, it is one of the easiest instruments to get wrong. A badly worded question produces answers you cannot interpret, and no amount of clever analysis fixes it afterward.

This guide walks through the design of a dissertation questionnaire from research question to analysis, with examples of weak and strong wording and a checklist for piloting.

Start With Your Research Question, Not the Questions

Before you write a single item, list what you need to find out. Every question in your questionnaire should connect to a specific research question or hypothesis. If you cannot say why a question is there, remove it.

A useful exercise is to build a simple table. In one column, put each research question. In the next, list the variables you must measure. In the third, write the questionnaire items that measure them. Gaps and redundant items become obvious.

Decide Whether a Questionnaire Is the Right Method

Questionnaires suit research that needs standardized answers from many people: attitudes, behaviors, experiences, and demographic patterns. They work less well when you need depth, when the topic is sensitive, or when participants may not understand the questions without explanation. In those cases, interviews or focus groups may fit better. Your methodology chapter must justify the choice.

Choose Your Question Types

Different question types give different data.

• Closed questions (yes/no, multiple choice) are quick to answer and easy to analyze, but limit what participants can say.
• Likert scales (for example, strongly disagree to strongly agree) measure attitudes and are widely used. Use consistent scales and labels.
• Ranking questions ask participants to order items by importance. Keep the list short.
• Open-ended questions allow detailed answers but take longer to code. Use a few, and place them near the end.

Consider using validated scales where they exist. They have already been tested for reliability and validity, and they let you compare your results with earlier research. You will need to cite them and, where required, get permission to use them.

Write Clear Questions

Most questionnaire problems come from wording. Follow these rules:

1. One idea per question. Avoid double-barreled items such as "How satisfied are you with the content and delivery of the course?"
2. Avoid leading wording. "How much did you enjoy our excellent tutorial?" nudges the answer.
3. Use plain language. Avoid jargon and abbreviations unless your respondents all know them.
4. Be specific. "Do you study often?" is vague. "In a typical week, how many hours do you study outside class?" is not.
5. Cover the full range of answers. Include "not applicable" or "prefer not to say" where it makes sense, and make response categories mutually exclusive.
6. Put sensitive questions later, after participants are comfortable.

Before and After

Weak: Do you agree that the library's opening hours and study spaces are convenient and well-maintained?

Stronger:

• The library's opening hours suit my study schedule. (1 = strongly disagree to 5 = strongly agree)
• The library has enough quiet study space. (1 = strongly disagree to 5 = strongly agree)

The stronger version splits one double-barreled item into two clear, measurable ones.

Order and Layout

• Open with an introduction explaining the purpose, who you are, how long it takes, and how the data will be used.
• Start with easy, non-threatening questions.
• Group related items together under clear headings.
• Place demographic questions at the end, unless you need them for screening.
• Keep it as short as your research allows. Long questionnaires lower completion rates.

Ethics and Consent

Before you collect data, you will usually need ethics approval. Plan for these elements:

• An information sheet explaining the study, what participation involves, and who to contact.
• Informed consent, often a tick box at the start of an online form.
• A statement that participation is voluntary and participants may withdraw.
• Clear information about confidentiality or anonymity, and how data will be stored and for how long.
• Extra care if participants are under 18 or in vulnerable groups.

Do not start collecting data before approval is confirmed. Your department's ethics process may take weeks, which is why it belongs early in your timeline.

Pilot Before You Launch

A pilot test is the cheapest protection you have. Give the questionnaire to five to ten people similar to your target group, and ask them to tell you:

• Which questions were unclear or confusing.
• How long it took to complete.
• Whether any options were missing.
• Whether any question felt uncomfortable.

Revise based on their feedback, and ideally repeat the pilot if you change a lot. Report the pilot in your methodology chapter; examiners value it.

Sampling and Distribution

Your sample determines how far you can generalize your results.

• Probability sampling (such as random sampling) allows stronger generalization but needs a list of the population.
• Non-probability sampling (convenience, snowball, or purposive) is common in student research. It is practical, but you must acknowledge its limits honestly.

Decide how many responses you need before you launch, based on the analysis you plan to run. Many statistical tests have minimum sample sizes. Then choose a distribution method that reaches your group, such as email, learning platforms, social media, or in-person handouts, and plan reminders. Response rates are almost always lower than students expect.

Preparing Data for Analysis

When responses arrive:

1. Export the data and keep an untouched backup.
2. Clean it: remove duplicates and incomplete responses according to a rule you decide in advance.
3. Code responses numerically, and create a codebook that documents each variable.
4. Check for errors and unusual values.
5. Run the analysis you planned, and report what you did.

For guidance on what to do next, see our guide to choosing the right statistical test and how to write the findings chapter.

Reliability and Validity

Show that your instrument measures what it claims to measure. For scales, report internal consistency (commonly Cronbach's alpha). Explain how you checked that questions matched the concepts, and note any limitations, such as self-report bias or a non-random sample. Our article on validity and reliability in research explains the terms.

Common Mistakes

• Writing questions before deciding what you need to measure.
• Skipping the pilot.
• Using double-barreled or leading questions.
• Making the questionnaire too long.
• Failing to get ethics approval first.
• Collecting too few responses to run the intended analysis.
• Attaching the questionnaire without referring to it in the text. Place a copy in an appendix and mention it.

Summary

A strong questionnaire starts with your research question, uses clear and unbiased wording, is piloted, and is distributed to a sample you can defend. Plan the analysis before you collect data, and document every decision for your methodology chapter.

Where to Go Next

• Want feedback on your methodology chapter? See our dissertation editing service.
• Write up what you collect: Dissertation Findings Chapter.
• Build your methods section: How to Write a Dissertation Methodology Chapter.
• Work through design choices with our academic coaching, or see the pricing.

From choosing a topic to polishing a final draft, CampusScribe's subject-specialist editors and coaches help with dissertations, capstone projects, presentations, and more - whether you need topic guidance, structural feedback, or a final edit before your deadline.