Rapport building is usually the first step of a job interview. Even when ensuing interview questions are standardized and job-relevant, it’s typical to start with a few questions that seek to ease tension and establish a friendly connection between the interviewer and interviewee. But questions remain: what is the purpose of this, and how does this affect how the interviewee is rated? On one hand, ability to establish good rapport may be indicative of a socially-competent candidate. On the other hand, if the interviewer forms a strong intuitive opinion about an applicant, it may color subsequent scores on the actual job interview questions. So, is rapport building good or bad? Should the practice be continued or phased out?
THE VERDICT ON RAPPORT BUILDING
Researchers (Swider, Barrick, & Harris, 2016) conducted an experiment using 163 simulated structured interviews of students who were preparing for actual job interviews. Unique to this experiment, interviewers rated the interviewees on first impressions (after rapport building), in addition to scoring the structured interview questions. For comparison, additional trained experts who did not witness the rapport building were also used to score the interview responses.
Results indicate that scores provided based on initial impressions alone were correlated with the expert ratings of the actual interview. This demonstrates that rapport building is capable of measuring some aspect of the job-relevant qualifications. At the same time, these initial impression scores were found to be partially measuring something beyond job-relevant qualifications. This is the type of thing that interviewers want to avoid, because it “contaminates” their assessment of an applicant. Further, this contamination introduced by first impressions was also statistically different from other ways that interviewees manage their image, for example by physical appearance or impression management. This means that the rapport building is contributing something very unique to the evaluative process, which may have the potential to making applicant evaluation more inaccurate.
Finally, researchers found that when interviewers had more favorable first impressions of the interviewees, they gave them higher scores on subsequent interview questions. Indeed, interviewers seemed capable of letting first impressions misguide their ability to accurately rate people. However, this effect was more pronounced for early interview questions. By the end of the interview, first impressions played a less prominent role in influencing ratings.
This research shows that rapport building and first impressions are partially job relevant, but partially not job-relevant. The part that is not job-relevant can throw off the accuracy of interview ratings. What can organizations do about this? The authors recommend training interviewers to follow a rapport-building script, and to rate responses as though they are part of the interview. Because first impressions affect early job interview questions to a higher degree, the authors suggest not scoring the first actual interview question, and they emphasize the need to train interviewers to remain open-minded throughout the entire interview. This approach would allow interviewees to overcome poor first impressions.
On the other hand, rapport building cannot simply be eliminated. The authors suggest that it is still useful because it does, in fact, predict some job-relevant characteristics, such as social competence. Also, it may help the participants feel at ease and make subsequent questioning go more smoothly. Also, they say, rapport building may be an opportunity to recruit applicants.
For job applicants, the authors warn not to underestimate the importance of rapport building. They say that those who struggle in this area may want to pursue coaching on small talk, role-playing, or even hand-shaking.
Swider, B. W., Barrick, M. R., & Harris, T. B. (2016). Initial impressions: What they are, what they are not, and how they influence structured interview outcomes. Journal of Applied Psychology, 101(5), 625-638.