Skip to content

Attention check

Also called instructional manipulation check, IMC

An item with a knowable correct answer, used as one signal that somebody is reading rather than clicking.

The usable form is a plain instruction: “For this item, please select Disagree.” It tests whether an instruction was followed. Trick questions, obscure knowledge and tiny visual details test something else, and they fail people whose reading, language or screen is different rather than people who are inattentive.

Decide before fielding what a failure means: end participation, or flag the response for review. Write the rule down, because deciding after you see the outcome is how an inconvenient result becomes an excluded one. One failed check is a signal to look, not proof that every other answer is worthless.

Pair it with other evidence. Response timing, straight-lining and internal consistency together support a decision that no single item can.

Where it goes wrong

An exclusion rule invented after analysis. Report the rule and the number of exclusions alongside the final sample size.

See also

Straight-lining
Selecting the same answer across every row of a question battery.
Screener
The questions at the start that decide whether somebody belongs in the study, and route them accordingly.
Completion rate
The share of people who started your survey and reached the end; its inverse is drop-off.

All 35 terms

Start

Put it into practice in your next survey.