Desktop version

Home arrow Psychology arrow The Wiley Blackwell handbook of the psychology of recruitment, selection and employee retention

Criteria for Evaluating Interview Responses

Every interviewer would like to be as accurate as possible about the potential of the candidate in question, so employing multiple criteria to evaluate interview responses should be high on the interviewer’s list of priorities. One of several techniques that can facilitate the accurate evaluation of a candidate’s responses is understanding the interview context within the framework of the Realistic Accuracy Model (RAM) (Funder, 1995). RAM advocates that there are four moderator variables that facilitate accurate interpersonal judgements of others. The first is judgemental ability (also known as the good judge). The RAM model has shown that some individuals are simply better judges of personality than others due to their ability to perceive and use cues from the target correctly. Past research (Akert & Panter, 1988) and theory (Funder, 1995) suggest that the good judge has an extraverted personality and wide experience in social settings (Christiansen, Wolcott - Burnam & Janovics, 2005; McLarney, 2003). Such experience undoubtedly gives the judge more knowledge about how personality is revealed in specific behaviours. Organizations should consider carefully before choosing an interviewer as the accuracy of the interview is largely dependent on that person. The most senior individual in the organization may not be the best interviewer (Funder, 1995). However, interviewers can become better judges by training and rating videotaped targets, thus increasing the accuracy of their judgements and decreasing biases or errors (Wexley, Sanders & Yukel, 1973).

A second moderator of interpersonal judgement according to RAM is termed the ‘good trait’. Research shows that properties of the trait being judged also affect the degree to which an accurate interpersonal judgement is likely (Funder & Dobroth, 1987). RAM posits that some attributes are difficult to judge, while others are relatively easy (Funder, 1995). An attribute such as ‘is socially at ease around others’, which is revealed by frequent positive social interactions, is easier to judge than a quality such as ‘daydreams or ruminates frequently’. For this trait the interviewer must infer its presence from verbal statements from the target or, even more difficult, from distracted responses. Funder and Dobroth (1987) revealed that the more visible the trait or cues of the trait, the higher the levels of inter-judge and self-other agreement. With this knowledge in hand, interviewers can be confident in the accuracy of their ratings for highly visible traits such as the degree to which the individual is conscientious, agreeable or even dependable. And for those hard-to-judge qualities or traits, such as the candidate’s potential to sabotage others’ work, the interviewer should focus on going beyond the interview (e.g. reading letters of recommendation, utilizing assessment centres) to increase the quality and quantity of information on which to base the judgement.

Table 9.1 The optimal usage of various interview formats.

Interview Format


Predictive Criteria

Optimal Use



Standardized questions, no follow-up questions, ratings forms

Skill sets, future job performance

To facilitate reliable comparisons between candidates

Campion, Palmer & Campion, 1997;

Levashina, Christopher,

Morgeson & Campion, 2014


Spontaneous questions, small-talk, informal format, follow-up questions

Personality characteristics, integrity, counterproductive behaviour, organizational citizenship

Assessing the candidate’s potential for engaging in counterproductive behaviour and person- organization fit

Blackman, 2002a


Multiple interviewers and one applicant

Skill sets

Reduces interviewer bias, perceived by applicants as a fair interview

Prusha, 2014; Dipboye, Gaugler & Hayes, 1997

Group Interview/ Multiple Applicant Interview

Multiple applicants with a single interviewer

Skill sets, future job performance

Efficient pre-screening device to make broad cuts in applicant pool

Tran & Blackman, 2006


Conference Call

Interview over the phone or via a video-conference monitor

Skill sets, future job performance

Efficient pre-screening device to make broad cuts in applicant pool

Chapman & Rowe, 2002; Sear, Zhang, Wiesner, Hackett & Yuan, 2013; Straus, Miles & Levesque, 2001;

Blackman, 2002b

The third moderator variable is termed ‘good information’. This variable refers to the quantity and quality of information available to the judge of personality. At a basic level, the longer you have known or have experience with targets the more accurate you will be in assessing their traits and qualities (Funder & Colvin, 1988). More specifically, in an experimental setting, Blackman and Funder (1998) showed that self-other agreement steadily increases as acquaintanceship with a target subject via videotaped, 5-minute episodes increases. Other studies support this finding (Cloyd 1977; Colvin & Funder 1991; Funder & Colvin 1988; Paulhus & Bruce 1992; Paunonen & Jackson, 1988).

The quality of information likewise plays a role in the assessment process. Blackman and Funder’s (1998) research specifically supports the utility of this moderator variable. When comparing the accuracy of ratings made by interviewers who conducted a structured interview with who those used an unstructured format, the interviewers using the unstructured format were significantly more accurate in assessing the job candidate’s personality and job-relevant traits. This research indicates that the unstructured interview, in which the candidate’s behaviour is spontaneous, is generally more informative about personality than meeting the candidate in a very structured format where personality is less likely to vary (Blackman, 2002a).

The implications for interviewers are that part of the interview should be unstructured when the candidate’s behaviour is free to vary, thus producing high quality information on which they can base their judgements. The unstructured format will increase the likelihood that the interviewer will make an accurate judgement about the candidate job-related qualities. Evidence suggests, however, that interviewers should refrain from using interview techniques that yield poor quality information, such as the telephone conference or, in some cases, the multiple-applicant interview with more than four candidates participating (Blackman, 2002a; Chapman & Rowe, 2002; Straus, Miles & Levesque, 2001; Tran & Blackman 2006).

The final moderator variable in RAM is coined ‘the good target’. Colvin’s (1993a, 1993b) research found that some individuals are easier to judge than others. This ultimately increases the likelihood that candidates will be judged accurately. ‘Judgable people’ are those Colvin found to be consistent in their actions, cognitions, words and deeds in different situations. For this reason, their future behaviour is easy to predict and the judge of personality more likely to be accurate in the assessment.

Blackman and others have identified the challenge of candidate behavioural inconsistency. This behaviour may be an indication that the individual is trying to conceal a negative aspect of their personality or work-related traits (Funder, 1995; Blackman, 2008). Follow-up interviews are an ideal way to observe a candidate’s behavioural patterns, while eliciting more information from which to make an accurate assessment (Blackman, 2008).

Interviewers who are able to increase the accuracy of their assessment of the candidate during the interview process are doing a lot more than hiring the most qualified candidate. Accurate interview assessments are also a preventative measure that an organization is taking to reduce the number of employees who might engage in counterproductive behaviour, such as absenteeism, weight, volatile behaviour and workplace violence (Blackman, 2008; Ones, Viswesvaran & Schmidt, 1993).

When interviewers incorporate into the interview process their knowledge of the four moderator variables that curb accurate interpersonal judgement they increase their likelihood of hiring the best candidate. Regrettably, due to the complexity of the characters involved and the interaction between them, there is no single way to accurately predict a job candidate’s standing on every job-relevant attribute during an interview. But if the duration of the interview is increased, a larger breadth or quantity of information becomes available on which to base a judgement. Even when conducting a panel interview, if one interviewer misses a candidate-prompted indicative behavioural cue, another interviewer may spot it. With multiple judges, a more accurate picture of the candidate can be made (Prusha, 2014).

Once knowledge of how to increase the accuracy of one’s interpersonal judgements has been achieved, interviewers should supplement this knowledge with structured response rating scales, such as the Behaviorally Anchored Ratings Scales (BARS). BARS were developed by Smith and Kendall (1963) as a superior appraisal method to subjective graphic rating scales. BARS focus on identifying important employee behaviours that are relevant to successfully completing a particular job, rated on a numerical scale. Standard graphic rating scales, on the other hand, do not focus on the specific behaviours, but on personality characteristics and subjectively determined work habits. An example is: ‘Employee answers telephone within three rings.’ A standard rating scale might state it more loosely: ‘Employee answers telephone promptly and efficiently.’

Behavioural anchors are typically developed using a critical incident technique or task analysis. BARS have been to shown to be more accurate and reliable than mixed standard scales when evaluating job applicants (Benson, Buckley & Hall, 1988). The method has been shown to reduce the adverse impact of various ethnic groups in a college admission study (Sinha, Oswald, Imus & Schmitt, 2011). An advantage of BARS is that applicants and employees view the rating process as fairer than trait-rated scales (Latham & Seijts, 1997). In addition, this method is more legally defensible when evaluating employee performance than other graphic rating scales (Benson, Buckley & Hall, 1988). And if the interviewer plays a role in the development of the BARS, convergent validity has been shown to increase while errors or biases such as the halo decrease (Friedman & Cornelius, 1976). However, a disadvantage of BARS is that administrators need to update the behavioural anchors as the job requirements change or the work context evolves.

< Prev   CONTENTS   Source   Next >

Related topics