What is Recency Effect?
The recency effect is a cognitive bias where people are more likely to remember or choose items they saw most recently — typically the last items in a list or the most recent information presented. In surveys, it can skew responses toward later options or questions.
The recency effect describes the tendency for respondents to give disproportionate weight to information presented near the end of a sequence. In questionnaires this shows up as higher selection or recall for items listed last, stronger influence from the most recent question when answering later ones, or heavier reliance on recently primed concepts. Recency effects often contrast with the primacy effect (favoring the first items); which one dominates depends on how the survey is delivered (written, spoken, or digital), the response format, and respondent memory constraints.
Usage example
A council questionnaire asks residents to rank ten local priorities listed in a fixed order. Even though all items are relevant, several priorities at the end of the list receive disproportionately more selections — a likely sign of a recency effect.
Practical application
Why it matters: recency effects can bias results and misrepresent true priorities or opinions, particularly in multiple-choice lists, ranking tasks, and long questionnaires. To reduce this bias, designers can randomise question or option order, rotate lists across respondents, group similar items thoughtfully, or place the most important items in multiple places. In multilingual or multi‑mode projects, pay attention to delivery mode (spoken surveys tend to amplify recency) and ensure randomisation and order-preserving features work consistently across languages. Mitigating recency bias improves data quality and helps organisations make fairer, more reliable decisions based on survey responses.
FAQ
How is recency effect different from primacy effect?
Recency favors the last items seen, primacy favors the first. Which appears depends on mode and memory load — spoken or interviewer-led surveys often show stronger recency effects because people rely on their short-term memory, while self-administered written surveys often show primacy.
Will randomising answer choices always remove recency effects?
Randomisation usually reduces order bias by spreading position effects across respondents, but it can introduce other issues (loss of comparable order, cognitive load). Use randomisation thoughtfully and combine it with pretesting and split-sample checks.
Do recency effects affect open-text answers?
Yes. The most recent question or prompt can prime respondents and shape the language or topics they use in open-text responses. Clear question sequencing and neutral prompts help reduce unwanted priming.