Virtually all organizations use interviews as a selection device for most jobs. Generally candidates are interviewed by at least two people before being offered a job. HR specialist and the person who will be the candidate’s immediate supervisor conduct these interviews. For managerial and professional jobs, it is common for the candidate to have a third interview with a higher level manager, such as a division head.19 because the interview is so popular, one might think it is a highly useful election device. But this is not always the case. We shall consider the reliability and validity of the interview.
Reliability of the Interview:
In the interview context, reliability is consensus, or agreement, between two interviewers on their assessment of the same candidates. This is called Interpreter Reliability. Research shows that it is rather weak.
Validity of the Interview:
The predictive validity of the interview is very low. Research in 1970s and 1980s suggested that the average validity of the interview for predicting job performance was as low as 0.14. 20 However, recent research has suggested that some individual interviewers may be valid, whereas many others are not. Past research has also lumped various types of interview procedures together, but very recent research suggests that some interview procedures can be quite valid.21
What can go wrong in the typical interview to cause many interviewers to make inaccurate predictions? It seems that interviewers often commit judgmental and perceptual errors that can compromise the validity of their assessments.
Similarity Error: Interviewers are positively predisposed to candidates who are similar to them (in hobbies, interests, personal background). They are negatively disposed to candidates who are unlike them.
Contrast Error: When several candidates are interviewed in succession, raters tend to compare each candidate with the preceding candidates instead of an absolute standard. Thus an average candidate can be rated as higher than average if he or she
comes after one or two poor candidates and lower than average if he or she follows an excellent candidate.
First Impression Error: Some interviewers tend to form a first impression of candidates rather quickly, based on a review of the application blank or on the first few moments of the interview. Thus, this impression is based on relatively little information about the candidate. Nevertheless the initial judgement is resistant t change as more information or contradictory information is acquired. In addition, the interviewer may choose subsequent questions based on the first impression, in an attempt to confirm the positive or negative impression. 22
Traits Rated and Halo Error: Halo error occurs when either the interviewer’s overall impression or strong impression of a single dimension spreads to influence his or her rating of other characteristics. For instance, if a candidate impresses the interviewer as being very enthusiastic, the interviewer might tend to rate he candidate high on other characteristics, such as job knowledge, loyalty and dependability. This is especially likely to happen when the interviewer is asked to rate many traits
Types of Interviews: Interviews can be classified by their degree of structure, or the extent to which interviewer plan the questions in advance and ask the same questions of all the candidates for the job. Three types of interviews, based on three degrees of structure, can be defined:
Unstructured Semi structured Structured (interviews)
Unstructured Interviews:
Here questions are not planned in advance, and interviews with different candidates may cover quite different areas of past history, attitudes, or future plans. They have low interrator reliability and lowest validity. Because questions are not planned, important job related areas may remain unexplored, and illegal questions may be asked on the spur of the moment.
Semi structured Interviews:
Involves some planning on the part of the interviewer but also allows flexibility in precisely what the interviewer asks the candidates. Semi structured interviews are likely to be more
.
Structured Interviews:
Research shows conclusively that the highest reliability and validity are realised in the structured interview. In a structured interview, questions are planned in advance and are asked of each candidate in the same way. The only difference between interviews with different candidates might be in the probes, or follow-up questions, if a given candidate has not answered the question fully. Interviews that feature structured questions usually also provide structured rating scales on which to evaluate the applicants after the interview.
Stress Interview:
Sometimes where the job requires the jobholder to remain calm and composed under pressure, the candidates are intentionally subjected to stresses and strains in the interview by asking some annoying or embarrassing questions. This type of interview is called stress interview.
Placement:
Placement refers to assigning rank and responsibility to an individual, identifying him with a particular job. If the person adjusts to the job and continues to perform per expectations, it means that the candidate is properly placed. However, if the candidate is seen to have problems in adjusting himself to the job, the supervisor must find out whether the person is properly placed as per the latter’s aptitude and potential. Usually, placement problems arise out of wrong selection or improper placement or both. Therefore, organisations need to constantly review cases of employees below expectations / potential and employee related problems such as turnover, absenteeism, accidents etc., and assess how far they are related to inappropriate placement decisions and remedy the situation without delay.