Designing a Questionnaire
Three consecutive slides carry twenty separate instructions about building a questionnaire, in no stated order and with four of them repeated. Sorted into the sequence you would actually work through, they amount to a usable working method with three holes in it.
What the design guidance consists of
The design content is three slides deep and follows the two slides that argue questionnaires are the least satisfactory way of collecting anything, covered on Questionnaires and Response Rates. Before it sits the survey construction slide, which supplies the week's only build order: questions and response categories, then background information and instructions, then organisation and length, then layout and design.
Stated once before you use any of it: the week's single learning objective is to evaluate methods of data collection, and the delivery audit records that verb as undelivered. This is instruction in building one instrument, not grounds for choosing it over another.
Nine questions to ask before you draft anything
The first slide is prefaced "First of all, ask yourself:" and lists nine questions. They are reproduced below in the order printed, with what each one is really deciding and what the material does with it afterwards.
THE NINE DESIGN QUESTIONS, AS PRINTED
| Question, quoted | What it decides | What the source supplies after asking |
|---|---|---|
| "What is the question content, scope and purpose?" | The subject matter and its boundary | No method for setting scope; no link back to the research question |
| "What response format can I use to collect the best information from the respondent?" | Whether answers are fixed, ordered, scaled or open | Three formats appear five slides later, under levels of measurement. The connection is never made |
| "Does my population have knowledge of the topic?" | Whether your respondents can answer at all | No screening question, no filter, no test of competence to answer |
| "Does my population have good reading skills?" | Whether the instrument is readable by the people receiving it | No reading-level guidance, no plain-language rule, no alternative mode |
| "How should I word the question to get at the issue of interest?" | Wording | A separate slide of six wording checks, treated on the wording page |
| "What facilities and timeframes do I have?" | Feasibility of administration | Nothing further; the same concern appears in the method-selection questions four slides earlier |
| "How complex is the information I want to collect?" | Whether a questionnaire can carry it at all | No complexity threshold, and no statement of when to switch instrument |
| "Do my questions pose sensitivities e.g., asking about age, income?" | Whether people will answer honestly | Two examples — age and income — and no procedure for either |
| "What types of methods have been used in research projects similar to mine?" | Whether to follow precedent | No search method; the same concern is raised in the method-selection slide |
Nine questions are printed. The source states no count, does not number them and does not say which are answered before which.
Six design points, five distinct requirements
The second slide is the operative one. Under the heading "Questions should be", six points are printed in this order.
- "short and simple. It is far better that there are many short questions than a few long questions"
- "unambiguous"
- "as short as possible"
- "neither irrelevant nor too personal"
- "not leading i.e. one suggesting the answer" — followed by a worked example of a question that suggests the answer yes
- "designed so the questions fall into a logical sequence. This enables the respondent to understand its purpose"
Point 5 is the deck's definition of a leading question — "one suggesting the answer" — and it comes with the only worked fault example in the design material. Leading questions, and the other wording faults, are treated with worked repairs on The Wording of Questions, which is the better-developed half of this topic.
Five reminders, four of which you have already been given
The third slide is headed "Remember to" and lists five items. Set against what has already been said, four are restatements.
THE FIVE REMINDERS AND WHERE EACH WAS ALREADY STATED
| Reminder, quoted | Status | Where it was already covered |
|---|---|---|
| "Sequence questions appropriately" | Repeat | Design point 6 — logical sequence |
| "Create a clear and easy to follow layout" | Repeat | The survey construction slide — "Determining layout and design" |
| "Be careful not to create misleading questions" | New word, existing fault | Design point 5 called it "not leading"; the wording slide asks whether a question is "confusing or misleading" |
| "Use language respondents are familiar with" | Repeat | Design questions 4 and 5, and the wording slide's ban on jargon and acronyms |
| "Thank respondent for their participation" | New to questionnaires | Repeated later as an interview instruction — "thank them for their time" |
Only the last reminder adds anything to the questionnaire guidance, and it is the only statement in the whole design sequence about what surrounds the questions.
The sequence you would actually work in
The source lists twenty items and orders none of them. The sequence below is this page's arrangement of the source's own material — every item comes from the week, the ordering does not. It follows the one build order the week supplies and slots the design points and reminders into it.
The source's instructions, sorted into working order
Settle content, scope and purpose
Design question 1. Decide what the instrument is for and where it stops. Nothing in the material links this back to your research question; do it anyway.
Check the population can answer
Design questions 3, 4 and 7 — topic knowledge, reading skills, complexity. If any answer is unfavourable, the instrument changes rather than the population.
Draft questions with their response formats together
Design question 2, and the construction slide's first activity. The formats the material offers are on Levels of Measurement in Structured Questions, and the choice between fixed and open answers is on Structured and Unstructured Questions.
Apply the five distinct design rules
Short and simple; unambiguous; neither irrelevant nor too personal; not leading; in a logical sequence. Applied to each question individually, then to the set.
Work the wording
Design question 5, and the six checks on the wording slide. This is where most of the week's genuinely useful content sits.
Order the questions and fix the length
Design point 6 and the construction slide's third activity. A principle is given for order and no figure at all for length.
Write what surrounds the questions, lay it out, then rehearse
Background information and instruction, the closing thanks, a clear and easy-to-follow layout, and the rehearsal the material calls essential and does not describe.
Order, length and the covering explanation
Three design decisions deserve separate treatment, because the material's coverage of them ranges from a single principle to nothing at all.
WHAT THE MATERIAL SETTLES AND WHAT IT LEAVES OPEN
| Decision | What the source says | What it does not say |
|---|---|---|
| Order | Questions should "fall into a logical sequence", which "enables the respondent to understand its purpose" | What makes a sequence logical, and how it resolves against the instruction to ask easy questions first and difficult ones last |
| Length | "Working through organisation and length" is listed as a construction activity; many short questions are preferred to a few long ones | No number of questions, no completion time, no page limit — anywhere in the week |
| Layout | "Create a clear and easy to follow layout"; "Determining layout and design" | No spacing, numbering, grouping, page-break or response-alignment guidance |
| The covering explanation | "Writing up background information and instruction" is named as a construction activity; respondents are to be thanked | No content, no length and no wording. What a respondent should be told appears only in the interview procedure, not in the questionnaire guidance |
The pilot stage the source requires and never describes
Where personal and sensitive questions land
The material takes three different positions on personal questions across three consecutive slides, and a reader who needs sensitive data has to notice all three.
The three positions, as printed
- Design question 8 treats sensitivity as something to be considered: "Do my questions pose sensitivities e.g., asking about age, income?"
- Design point 4 prohibits it outright: questions should be "neither irrelevant nor too personal".
- The wording slide treats it as a matter of phrasing: "Is the wording sensitive? Not too direct? Not too personal?"
What none of the three supplies
- Where the line between personal and too personal falls
- Any procedure for asking about age or income, the two examples the material itself raises
- Any option to make a sensitive item optional, banded or anonymous
- Any connection to consent or confidentiality, named only in the interview slides
What to carry forward
- Twenty instructions across three slides, in no stated order: nine questions, six design points, five reminders.
- The six design points contain five distinct requirements — short and simple, unambiguous, neither irrelevant nor too personal, not leading, logically sequenced.
- Four of the five reminders repeat earlier instructions; the fifth, to thank the respondent, is the only guidance on what surrounds the questions.
- "Leading" and "misleading" are used interchangeably and never distinguished.
- Length is never stated. Decide it yourself and record the decision.
- The instrument must be rehearsed with someone else — an instruction called essential, described nowhere, and never named as piloting.
- Personal questions are prohibited on one slide, prompted on another and reduced to a wording matter on a third. No line is drawn between them.
Frequently asked questions
How many questions should a questionnaire have?
The material never says. Its only statement about length is that many short questions are preferable to a few long ones, and "working through organisation and length" is listed as a construction activity with no figure attached. No number of questions, completion time or page limit appears anywhere in the week, so the decision and its justification are yours.
What order should the questions go in?
The design slide requires a logical sequence that lets the respondent understand the instrument's purpose. A later slide, addressed to interviews, requires easy questions first and difficult ones at the end. The two orderings often conflict and the material never says which prevails, so state your ordering principle in your write-up rather than assuming a reader will infer it.
Does the material tell me how to pilot a questionnaire?
No. It requires that wording be "thought through carefully and rehearsed with someone else before you begin" and calls that essential, but the words pilot, pre-test and trial never appear, and no number of rehearsers, procedure, success criterion or revision step is given. Treat the rehearsal as a minimum and design the rest of the procedure yourself.
Can I ask about income or age?
The material raises both as examples of sensitivities to consider, then prohibits questions that are "too personal", then treats sensitivity as a wording matter — three positions in three slides, with no line drawn between them and no procedure for asking either question. If your research question needs the data, take the technique from your ethics approval process and from published instruments rather than from this material.
Do the same design rules apply to interview questions?
Yes, on the source's own statement: the design points apply whether the questionnaire is used "as a postal/digital questionnaire or as a basis for interviewing". That is one of the few places the material settles a scope question explicitly, although its interview slides then add instructions that pull against the design points on question order.
Where do the response formats come from?
The design questions ask what response format you can use and give no options. Three appear five slides later on a separate slide about structured questionnaires — nominal, ordinal and interval, plus dichotomous items — and the connection between the two slides is never made in the material. Draft the question and its response options together anyway.
References and source attribution
- Bryman, A. 2016, Social Research Methods, 5th ed., Oxford University Press, Oxford.
- O'Leary, Z. 2017, The Essential Guide to Doing Your Research Project, 3rd ed., Sage Publications, London.
- Veal, A. J. 2005, Business Research Methods: A Managerial Approach, Longman.
- Naoum, S. G. 2013, Dissertation Research & Writing for Construction Students, 3rd ed., Routledge.
- The supplied teaching source: the week 10 slide deck on collecting data, the three consecutive slides on designing a questionnaire, read with the survey construction and wording slides of the same deck.
Suggested questions for Ask KEVOS
- Sort my draft questionnaire against the five distinct design rules in this material.
- Design a rehearsal session for my instrument, given the source only says to rehearse it with someone else.
- How should I order my questions when logical order and difficulty order disagree?
- Draft the background information and instruction section for a self-completed questionnaire.
- My study needs a question about salary bands. What does this material let me say, and what do I need from elsewhere?