Build a Survey

This journey is for you if your project is a questionnaire, a rating study, or any other survey with no reaction-time trials. It walks the manual in the order a survey builder needs it and ends with a published survey and a results file on your disk.

Because you are staying on the survey side, you can ignore whole stretches of the manual: nothing in the Building Experiments group concerns you, including Timing Parameters and Feedback. When you look up individual columns, the Columns Reference sidebar marks every column that applies to surveys with a clipboard icon, so you can tell at a glance which pages are yours and which belong to the stopwatch (experiment) world.

Learn the shape of a survey

1. Trial File Fundamentals. Every Testable project, surveys included, is a spreadsheet where each row is a trial and type is the only required column. Read the core rules and the type column section closely; skim the sections on stimuli and correct responses, which mostly matter for experiments (the stim column returns later if you want images beside questions).

2. Trial Types. Take two things from this page: type is where a project forks into its two worlds, and form is the value that makes a survey row. Read the form section, then skip the four experiment values.

Write your questions

3. Forms and Surveys. The anatomy of a form row (head, body, responseType, responseOptions), how consecutive rows gather onto pages, and how to score responses. Read it fully, this is the centre of the journey.

4. Survey Response Types. Every question widget, from radio buttons and Likert scales to sliders and ranking, plus matrices, “Other” options, option randomisation, and required answers. Read the responseType table now and return to each widget’s section as you build that question.

   Caution

   responseType values are case-sensitive, and an unrecognised value silently renders the row as display-only text. A misspelt widget name produces a question nobody can answer, so check your spelling in the previewer.

5. Generating Questions from Lists. Write one row and let a list expand it into a question per item, or feed a list into a question’s options or matrix rows. Skip on a first pass; come back when you catch yourself pasting near-identical rows.

Shape the flow

Each of these three steps is optional. Read the ones your design needs and skip the rest.

6. Randomisation. The random column shuffles question order and randomBlock shuffles whole blocks of rows. Skip if your questions run in a fixed order; shuffling the answer options within a question is covered in Survey Response Types, not here.

7. Screener Surveys. Qualify participants with a few opening questions and screen out those who do not meet your criteria. Skip unless your study has eligibility rules.

   Caution

   screenout only triggers within the first 5 trials, so screener questions and their logic rows must sit at the very top of the trial file.

8. How to Inject Logic. The if and then columns branch on a participant’s answers, skip sections, or redirect. Skip unless your survey branches.

Check and configure

9. Previewing Your Trials. Run through your survey from the editor before anyone else sees it. Note the limitations table: randomisation is not applied in the previewer, but if/then logic works, so you can test branching and screenout paths.

10. General Settings. Project name, language for default UI text, colours, security options, allowed devices, and the built-in participant details form for demographics. Read the security and participant details sections; most display settings can keep their defaults for a survey.

11. Informed Consent. Four consent modes, from none to an itemised custom form with downloadable proof of consent. Skip only if consent is handled entirely outside Testable.

12. Participants & Data. Participation limits, session timeouts, and how participants arrive: a direct link, SONA, Prolific, or Testable Minds. Read the recruitment section that matches your channel and skip the others.

Collect and read your data

13. How to Read the Results Files. How result files are named, what the header rows contain, and where answers live: response holds the answer text and responseCode the position of the chosen option.

14. Results Formats. The download options compared. For surveys the wide format is usually the one you want: one row per participant, with all answers to the same question in the same column.

15. Fieldwork Insights. A live dashboard of completions, countries, and drop-off, without downloading files. Skip on a first pass; it requires a Plus, Lab, or Departmental plan or a Feature Boost add-on.

You are ready when

You are ready when your questions render correctly in the previewer, required answers and any screening or branching behave as designed, consent and recruitment are configured, and you have collected a test session and opened its wide-format CSV to confirm every question’s answers land in the column you expect. From there, share your participation link, or read Testable Minds to recruit from Testable’s own panel.