Beware claimed data

A third of respondents expressed a view on fictitious legislation.

The questionnaire had asked a question, so respectable participants supplied answers.

Some inferred. Some guessed. Some chose whichever wording felt less embarrassing.

The resulting percentage described an attitude partly manufactured by the instrument recording it.

Commercial research performs gentler versions every day. Customers rate obscure features because “I never noticed it” is unavailable. People rank purpose statements that played no role in purchase. Prompted brand selection becomes “awareness” once the chart needs a title. The question appears neutral while quietly insisting that the feature, purpose or brand already occupies some space in the respondent’s mind.

Claimed data still reaches feelings, intentions and private experiences observation cannot see. It arrives entangled with memory, wording, social desirability and nearby answer options. Decimal points tidy the result without removing any of those conditions.

Ignorance, irrelevance and non-exposure are substantive findings. They may shrink segments and damage certainty, which is exactly why researchers are tempted to omit them.

A market containing millions with no opinion can be more commercially interesting than a forced preference between two polished claims. Weak salience, irrelevant positioning or a category customers barely distinguish may be the actual discovery. That absence can change product, media and message decisions more radically than a decimal-point preference manufactured on demand.

The uncomfortable respondent says, “I don’t know.”

The honest table leaves them there.

Behavioural principles

Behavioural ideas at play in this post

Short, plain-English explanations of the principles behind this post, with links to related books and examples in the archive.