Everyone knows that what you say in a job interview matters. What is less commonly appreciated is how much the way you say it and the way you look saying it matters too. Martín-Raugh and colleagues (2022) conducted the first meta-analysis to systematically examine which nonverbal cues most influence interviewers’ judgments, drawing together 63 studies and 4,868 participants spanning over 70 years of research.
The Theory Behind It
The authors ground their review in dual-process theory, which distinguishes between two modes of thinking. System 1 is fast, automatic, and implicit, the kind of processing that generates rapid impressions before conscious deliberation kicks in. System 2 is slower, deliberate, and analytical (Martín-Raugh et al., 2022).
The argument is that interviewers, even when trying to evaluate candidates fairly and systematically, are continuously running System 1 in the background. Nonverbal cues, appearance, eye contact, posture, facial expressions, feed directly into this fast, implicit processing, shaping evaluations in ways that may be invisible to both the interviewer and the candidate.
What the Numbers Show
The correlations the meta-analysis produced are striking in their magnitude. Professional appearance showed the strongest association with interview performance ratings (ρ = .62) — a remarkably large effect for a variable that has nothing to do with a candidate’s actual qualifications or answers. Eye contact came second (ρ = .45), followed closely by head movement (ρ = .43) (Martín-Raugh et al., 2022).
To put these figures in context: a correlation of .62 between professional appearance and interview ratings suggests that how a candidate looks accounts for a substantial portion of how they are judged — independent of anything they say. These are not subtle background influences. They are major determinants of perceived performance.
Does Structure Help?
One of the most important questions in interview research is whether structured interviews — those with standardised questions, scoring rubrics, and formal evaluation criteria — reduce the influence of irrelevant factors. The logic is intuitive: if interviewers are following a script and rating against defined criteria, there is less room for snap judgments to intrude.
Martín-Raugh and colleagues (2022) found that the moderating effect of interview structure was largely absent. The influence of nonverbal cues persisted regardless of whether interviews were structured or unstructured, conducted in person or remotely, or short or long in duration. Across all these variations, the pattern held: how candidates presented themselves nonverbally continued to shape how they were rated (Martín-Raugh et al., 2022).
This is a meaningful finding. It suggests that structural safeguards, while valuable for other reasons, do not reliably neutralise the pull of nonverbal impressions.
A Gender Dimension
The meta-analysis also uncovered a troubling pattern in how nonverbal cues operate differently for women and men. Several cues — including facial expressions and professional appearance — showed stronger associations with interview ratings for female candidates than for male candidates (Martín-Raugh et al., 2022).
The authors interpret this as evidence that interviewers apply gender-based stereotypes when evaluating nonverbal behaviour. Women appear to be judged more heavily on the basis of how they look and present themselves, with deviations from expected norms carrying greater consequences for their ratings. This is not a marginal effect — it reflects a systematic difference in how the same nonverbal signals are weighted depending on the candidate’s gender (Martín-Raugh et al., 2022).
Why It Matters
Job interviews are among the most consequential social evaluations most people will face. They determine access to employment, income, and opportunity. The assumption underlying their use is that they measure something meaningful about a candidate’s suitability — their knowledge, skills, judgment, and potential.
What Martín-Raugh and colleagues (2022) demonstrate is that interviews are also measuring something else: how quickly and favourably an interviewer’s System 1 responds to a candidate’s physical presentation. Professional appearance, eye contact, and head movement are not proxies for competence. They are, at least in part, proxies for familiarity, conformity to professional norms, and — in the case of gender differences — stereotyped expectations about how different people should present themselves.
None of this means interviews are worthless. But it does mean that organisations serious about fair and accurate hiring need to reckon honestly with what their interviews are actually measuring — and who bears the cost when they get it wrong.
Reference
Martín-Raugh, M. P., Kell, H. J., Randall, J. G., Anguiano-Carrasco, C., & Banfi, J. T. (2022). Speaking without words: A meta-analysis of over 70 years of research on the power of nonverbal cues in job interviews. Journal of Organizational Behaviour. Advance online publication. https://doi.org/10.1002/job.2670
