Interviews

Introduction

Remote or in-person contact during an interview is undoubtedly the most natural way of getting to know people. The purposes of the interviews are multiple. They can, for example, be used to collect information and descriptions for a selection, serve as a support for an orientation advice, or be part of a performance review.

Generalities

The interview process varies greatly. Some interviews only last five minutes, others an hour or two. Some interviews are individual: one interviewer and one interviewee. In others, the candidate faces two or more people, in still others, several candidates are interviewed at the same time with a single assessor. How the latter forms his judgment and what quality this judgment is is an essential point to consider.

Interview techniques are also practiced in the clinical environment, where pioneering debates allow interesting comparisons with approaches based on the use of statistics and tests.

Different forms of interviews

Five types of interviews can be differentiated based on their mode of operation:

• Non-directive
• Semi-directive
• Directive
• Structured
• Group

Non-Directive

In this type of interview, the interviewer leaves the person free to express themselves as they see fit. Of clinical origin, the non-directive interview is widely used in psycho-sociological surveys.

The interviewer adopts a position of empathy: intuitive understanding of the person facilitated by tolerance and open-mindedness, or of neutrality: listening/openness allowing the other to express themselves.

Non-directive interview is difficult to practice in a selection interview, where certain topics must be covered. It is the default technique used in day-to-day management, during one-on-one and group conversations.

The non-directive interview is considered a poor predictor of future professional behavior. Judgment, feelings, etc., are subject to very clear variations that concern both interviewers and interviewees. On the other hand, this type of interview complicates the comparison of candidates since there is no interview grid and each evaluator asks different questions to each new interviewee.

Semi-Directive

The person conducting this type of interview is directive in the questions they ask and non-directive in the answers they expect. This makes it possible to deepen certain points of a person’s biography. The specific questions asked by the interviewer aim to gather as much information as possible on targeted topics.

Unlike the non-directive interview, semi-directive interviews allow the evaluator to target relevant information with specific questions. However, if the questions asked are not the result of a job analysis, then the same problem arises as for non-directive interviews, namely, the comparison of candidates with each other. The evaluation is essentially based on the judgment of the evaluator, carried out in a short time and without precisely defined evaluation criteria.

Directive

In a directive interview, the interviewer asks precise questions, to obtain precise answers, the questions are closed and leave no opening to the candidate. This type of interview is implemented when verifying purely factual data, for instance, from the resume or social media. A direct interview has the advantage of being able to standardize the collection of information to a minimum. However, it can become a communication barrier and may prevent the person from fully expressing themselves, depriving the evaluator of some information.

Structured

Prior to the interview, a work analysis is carried out in order to identify the behaviors or performances essential to succeed in the position to be filled. The structured interview includes an interview guide in the form of a series of questions to ask and an evaluation grid with a clearly defined list of points to be evaluated. This grid is applied without exception in all interviews for a particular position. The term "structured interview" includes behavioral interview and competency-based interview techniques.

Group

In a group interview, the candidate responds to several interviewers at the same time, or a group of candidates responds to one os several interviewers at the same time. The idea of ​​group interviews is to cross-check impressions between interviewers in order to reduce interview bias and also to save time.

Group interviews have been studied by meta-analyses which show that the presence of several interviewers does not increase the validity of the interview and even gives less valid results than individual interviews.

Critics

Of clinical or sociological origin, the first three types of interview are not based on an analysis of the work. Studies show poor reliability and validity as well as high sensitivity to biases and distortions.

Numerous studies have been carried out highlighting the variability of reliability of interviews and particularly unstructured interviews. The degree of agreement between several evaluators remains low. Disagreements seem all the more pronounced when they concern a prognosis rather than a description. The multiple causes of disagreement between evaluators can be summarized: incompetence of one of them, inclination to defend ideologies corresponding to the groups to which they are affiliated, and finally, differences in the way each of them organizes information. As a result, the validity of unstructured interviews is also highly variable. Meta-analyses carried out since the 1980s give validities ranging from 0.38 to 0.63.

Research shows that interviews are all the more faithful and more valid when they are structured. These interviews are preceded by a job analysis, which makes it possible to determine the necessary behaviors. It can be supplemented by the analysis of critical incidents. The interview guide is written on the basis of the job description. It allows the different interviewers to have a canvas. In Campion's study the agreement between raters was 0.88. The validity linked to the criterion was 0.40 and 0.56 after correction of the dispersion of non-fidelity of the criterion.