A survey question about attitudes toward a specific demographic group produces measurably different response patterns depending on whether it's administered by an interviewer perceived as belonging to that group or not, despite the interviewer reading from an identical, standardized script in both cases. This is an interviewer effect — a well-documented phenomenon where a respondent's answer shifts based on the interviewer's perceived characteristics, entirely independent of anything the interviewer says or does differently.
Why this happens even with a perfectly standardized script
Respondents adjust their answers, often unconsciously, based on inferences about how a particular interviewer might react to or judge a given response — a mechanism closely related to social desirability bias, but specifically tied to the perceived identity of the person asking, rather than being a generic tendency toward socially favorable answers regardless of who's asking. A standardized script controls the words used; it doesn't control the respondent's inferences about the specific person delivering those words.
Where this effect is most pronounced
Interviewer effects show up most reliably on topics where the interviewer's visible or inferred characteristics are directly relevant to the question's subject matter — attitudes toward a specific group, when the interviewer is perceived as a member of that group; professional or technical topics, when the interviewer's perceived expertise or role could plausibly shape how a respondent expects their answer to be received or evaluated.
Why mode effects are a related but distinct concern
Separate from interviewer identity, the survey mode itself — telephone, in-person, online self-administered, mail — has its own documented, measurable effect on response patterns, generally with self-administered and online formats producing somewhat more candid responses on sensitive topics than interviewer-administered formats, likely because self-administration removes the interpersonal dynamic that interviewer effects specifically depend on. Combining data collected through different modes within a single study or tracking series without accounting for this can introduce a genuine confound between actual attitude change and a shift in survey mode.
What actually manages both kinds of effects
Standardized interviewer training that minimizes variation in tone, pacing, and non-verbal cues reduces but doesn't eliminate interviewer effects, since much of the effect operates through respondent inference about interviewer identity rather than through anything the interviewer actively does differently. Maintaining consistent survey mode across waves of a tracking study, and being cautious about combining data collected through meaningfully different modes, addresses the mode-effect side of the same broader concern.
What this means for designing and interpreting interview-based research
- Consider whether interviewer characteristics could plausibly interact with a study's specific subject matter before finalizing interviewer assignment
- Standardize interviewer training and administration style as thoroughly as possible on sensitive topics, while recognizing this reduces rather than eliminates the underlying effect
- Keep survey mode consistent across waves of a tracking study, or explicitly account for mode as a variable when it must change
- Be specifically cautious drawing conclusions about attitude change when mode or interviewer composition has shifted alongside the timeframe being studied
Interviewer and mode effects aren't evidence of sloppy fieldwork — they're well-documented, structural features of how survey responses are actually generated, and managing them requires deliberate methodological attention rather than simply trusting that a standardized script guarantees standardized answers.