ResearchCall

Survey workbench · dry run · no calls
Test mode — example data, not a real study
All eight stations are open. Ethical locked settings stay hidden. Network disabled · fixture transport · no real calls.

2. Instrument

4 settings shown · an agent asks 1 · 1 are recorded without effect

Text in double quotes is spoken word for word. Everything outside the quotes is rephrased by the agent — a behaviour measured in a real call.

Some settings on this station are kept but not yet read by anything. They are marked, because a control that quietly changes nothing is worse than one that is missing.

Randomisation spreads position and drop-off effects unless a fixed order is methodologically required.

Randomised draws a fresh item order per respondent, filters respected.

Parallel forms only make sense when memory effects or a parallel test are planned.

Nothing reads this value yet: Two equivalent forms need paired items; the item grammar has no pairing yet.

One line per rule: if q1 = no skip q4, q5. The triggering item must come before the ones it skips and must have fixed categories; the rule is fixed before fieldwork.

Compiled into a filter condition on every skipped item.

One line per item: id | hypothesis | format | "wording" | option | option. Formats: dichotomous, scale, scale_reversed, choice, open, creative. Options: scale=5:very unsatisfied..very satisfied · categories=yes,no · free (may be rephrased) · verbatim (word for word) · probe=2 or probe=unlimited (open items only) · rule=<analysis rule>. Example: q1 | H1 | dichotomous | "Do you use the bus?"

Becomes the questionnaire the dry run actually asks.