Video vs face to face interviews: The evidence and best practice

B
Authored By

Ben Schwencke

Video vs face to face interviews: The evidence and best practice

It is now very plausible that, as of 2024, more candidates are being screened using video interviews than face-to-face interviews. Almost all organizations are using video interviews at some stage in the recruitment process, but typically fairly early on. As a result, only those successful at video interviews end up attending face-to-face interviews, skewing the numbers in favor of video interviewing. However, as a communication medium, video is a relatively new way to interview candidates, and comparatively little is known about its effectiveness.

In this article, I will outline what we know about video interviewing and how it compares to in-person interviews for employee selection and assessment.

Face-to-Face vs. Two-Way Video Interviews

Face-to-face interviews employ a two-way communication structure, whereby both participants are simultaneously involved in the interview. The same goes for video interviews conducted over Zoom, Teams, Google Meet, or any other two-way video conferencing software, which we call “two-way” video interviews. The goal is to replicate the face-to-face interview experience but via video, while making the process more scalable and convenient for both interviewer and interviewee. This interview format is very common around the middle of the recruitment process, after psychometric testing but before a face-to-face interview. Alternatively, they may be used in lieu of a face-to-face interview, particularly when face-to-face interviews aren’t feasible.

One major difference between these two approaches is that research consistently shows that video interview candidates typically receive lower ratings than those assessed face-to-face. The reason is not fully understood, but it is likely that first impressions are just easier to make in person, and interviewers are easier to impress when in physical proximity. Consequently, organizations are advised to stick with one format or the other and not mix them together, lest they unfairly advantage face-to-face interviewees.

Older research also showed that candidates have historically found video interviews to be less fair than face-to-face interviews. They would state that video interviews provide fewer opportunities to connect with their interviewer, struggle with making small talk, and find it harder to express non-verbal cues, such as eye contact. Anecdotally, however, since the coronavirus pandemic and associated social distancing, I suspect that people have become considerably more comfortable with the video interview format, but more research is needed to confirm this.

Face-to-Face vs. One-Way Video Interviews

One-way video interviews, also known as asynchronous video interviews, represent a radically different approach to two-way interviews. One-way interviews are effectively recordings of responses to questions, allowing assessors to review interview recordings at their convenience. Consequently, one-way video interviews are typically used early on in the recruitment process, as they represent less of a time commitment for candidates and interviewers. Often, they are designed to filter candidates, helping to create a shortlist for two-way interviews, either in person or online.

Aside from the time-saving benefits of one-way interviews, this interview format is especially structured and repeatable, and research shows that structured interviews significantly outperform unstructured interviews when predicting future job performance. Each candidate is given exactly the same questions in the same format, enhancing the reliability of the interview process. Although this benefit can be replicated by doing structured two-way interviews, only a small proportion of organizations actually formally structure their interviews, so this represents a practical way to enforce good interview practice.

However, the downside is that one-way interviews are particularly cheatable using AI. ChatGPT or other large language models can generate scripts for candidates to read out, nullifying the benefits. Although candidates are often given a fairly short time duration to provide answers, this is usually enough time to copy through the question to ChatGPT and either create a script or at least receive some guidance. The candidate will still need to read the script out loud, however, and thus it still serves as a decent indication of their communication skills, but this isn’t really in the spirit of the exercise.

Conclusions and Recommendations

Video interviews are quickly becoming the norm, but they aren’t without downsides. When it comes to two-way video interviews, they absolutely have their place in recruitment but are not interchangeable with face-to-face interviews. Consequently, for a given stage in the recruitment process, only use video interviews or face-to-face interviews; separate them into distinct stages. One-way video interviews, until recently, have been a viable strategy for early screening, but given their vulnerability to AI-based cheating, I recommend against their use for now, at least until reliable methods of cheat detection can be established.

Ben Schwencke
Ben Schwencke

Ben is the chief psychologist at Test Partnership, with extensive experience in consultancy and research. He writes extensively on many topics, including psychology, human resources, psychometric testing, and personal development.