Marcus Andrey Study
Ler em português Library

Framework · Research · 3 min read

Screener questionnaire

A short questionnaire that decides who joins a study: criteria from the research questions, past-behaviour questions that can't be gamed, and quotas so six people aren't six copies of one.

  1. 1Criteria

    Inclusion and exclusion, each traced to a research question.

    Behaviours, not demographics
  2. 2Past behaviour

    Ask what they did, with four or five plausible options.

    "How did you send your last invoice?"
  3. 3Funnel

    Broad first. Reveal the topic late, in plain words.

    The tool appears among other options
  4. 4Open + quotas

    One open question to read, and how many of each kind.

    "4 who stopped, 2 who stayed"
  5. 5Pilot

    Try it on a few people, then a short call.

    Typos, ambiguity, broken skip logic
A screener lets the right people in, even when the wrong ones want in.

Use it when

  • You recruit participants for interviews or usability tests, yourself or through an agency.
  • An agency sends you its own screener and you need to check it against your brief.
  • Your last study filled up with friendly, tech-savvy volunteers who weren't your users.

How to use it

  1. Start from the research questions. Write the inclusion criteria (what makes someone a good fit, mostly behaviours) and the exclusion criteria (people in UX, marketing or IT, recent participants, professional testers). Screen only for what affects your questions. Output: two short lists, each line traced to a research question.
  2. Ask about past behaviour. Turn each criterion into a question about what people did, not what they would do. Offer four or five plausible options and mark which ones continue. Output: closed questions with no obvious right answer.
  3. Funnel the topic. Start broad and reveal the subject late, so people can't tailor their answers. Use plain words, no internal jargon. Output: questions in order, broad to specific.
  4. Add one or two open questions. Ask people to describe a recent time in a sentence or two. A one-word answer may predict a one-word session. Output: an open question you read before booking.
  5. Set quotas. Write how many of each kind you need: the differences that change the answer, and the people usually left out. Ask about access needs to adapt the session, never to filter. Output: one line of quotas.
  6. Pilot it. Try it on a few people to catch typos, ambiguous wording and broken skip logic. A 2–5 minute call after the written screener tells you how someone talks. Output: a tested screener.

Worked example

A team studies why freelancers stop using an invoicing app for returning clients. Six sessions, quotas: 4 who stopped, 2 who stayed, no more than 2 from one trade.

Question Options Continue if
Which of these describes your work in the last 12 months? Employed full-time · Freelance or self-employed · Student · Retired · Other Freelance or self-employed
How many clients did you invoice more than once in the last 3 months? None · 1 · 2–4 · 5 or more 2–4 or 5 or more
How did you send your last invoice? A spreadsheet or document · An invoicing app · My accounting software · Someone did it for me Any; tag for quotas
Do you work in any of these fields? Marketing · UX or product design · Software development · None of these None of these
In a sentence or two, tell us about the last time invoicing took longer than you expected. Open Answer is specific

The word "invoicing app" appears only in question 3, among other options, so nobody knows which tool the study is about.

Watch out for

  • Yes/no questions. "Do you use invoicing software?" announces the answer. Give plausible options instead.
  • Predictions. "Would you use…?" gets a guess. Ask what they did last month.
  • Overscreening. Matching every trait of the ideal user gives a screener nobody passes.
  • Percentages from the sample. Six purposive participants show what happens and why, not how often.

Template

Fill one row per criterion, then write the quotas.

Study:                        Sessions:          Length:
Research questions:
  RQ1:
  RQ2:

Inclusion (behaviours):        Traces to:
Exclusion:                     Traces to:

Q1  (broad, past behaviour)
    Options: a) ...  b) ...  c) ...  d) ...  e) Other
    Continue if:               Stop if:
Q2  ...
Q3  ...
Open: "In a sentence or two, tell us about the last time you ..."

Quotas: __ of ___ , __ of ___ , no more than __ from one ___
Access needs (to adapt the session, not to filter):
Piloted with:                  Fixed:
Source
NN/g, Maddie Brown (2024)
Where to find it
UX Research: Screeners and sampling, Recruiting participants

"See on Amazon" links are affiliate links. As an Amazon Associate I earn from qualifying purchases, at no extra cost to you.

Saved only on this device.

↑ ↓ to move · Enter to open · Esc to close