Date of this Version
Paul, M. (2020, June 17). Umbrella summary: Employment interviews. Quality Improvement Center for Workforce Development.
What are employment interviews? Employment interviews are defined as “a personally interactive process of one or more people asking questions orally to another person and evaluating the answers for the purpose of determining the qualifications of that person in order to make employment decisions” (Levashina, Hartwell, Morgeson, & Campion, 2014, p. 243). Interviews are often loosely described as being either structured or unstructured, varying in the extent to which interviewers have freedom and flexibility in carrying out the interview (Huffcutt & Arthur, 1994). Within this broad dichotomy, there are many structural aspects on which interviews can differ (Campion, Palmer, & Campion, 1997; Levashina et al., 2014): Initial rapport building (not permitted, limited or pre-planned, limitation on time or content, or no guidance) Transparency (no information provided to candidates about questions or competencies that will be assessed, information provided on competencies, questions provided in advance, or questions and competencies provided in advance) Consistency of questions across interviewees (same questions in same order, array of questions to choose from, topics to cover without guidance on questions, or no guidance) Use of prompting, probing, and follow-up questions (not permitted, only limited or preplanned probes or follow-ups permitted, unlimited probes or follow-ups encouraged or required, or no guidance) Use of ancillary information (e.g., application forms, work histories, or transcripts) Timing of questions from applicant (at the beginning, end, or throughout) Consistency of interviewer(s) across candidates Notetaking (extensive or brief, required or optional, actual responses or evaluations of responses, or during or after the interview) Rating timing (after each answer or at the end) Number of ratings (multiple ratings or one global rating) Types of rating scales (multiple, detailed anchors; single anchor; numbers or adjectives; or no scales) Types of anchors for rating scales (example answers; narrative descriptions of answers; evaluations of answers, such as excellent or good; or comparisons among candidates, such as answer given by X% of candidates) Timing of discussion about candidates and their answers (between interviews or after all of them) June 17, 2020 Scoring (combining ratings mathematically, within and/or across raters; using equal or differential weights; using consensus to resolve differences and arrive at final judgments) In addition to differing on structural aspects, interviews can also differ on several other factors. Though professional and legal guidelines indicate that interviews should be developed on the basis of a job analysis, this is often not the case. Interviewers are often left to develop or find their own questions, based on their best judgment of what will work. When job analyses are performed, the methods vary, resulting in different types of information on which to base questions. Traditionally, interviews have most often been conducted face to face, but with advances in technology, they are also being done using audio or video. They can be synchronous (candidate and interviewer are communicating in real time) or asynchronous (candidate and interviewer are participating at different points in time, such as through the use of recorded questions and answers). A variety of question types can be used, mostly categorized as situational, behavioral, credentials, experience, job knowledge, or self-perceptions (Campion et al., 1997; Janz, 1982). Situational questions ask candidates to describe what they would do in a hypothetical situation, such as, “What would you do if an angry and dissatisfied customer confronted you; how would you resolve their concern?” Behavioral questions ask candidates to describe how they behaved in a specific situation in the past, such as, “Describe a time when you encountered a conflict while working on a team; how did you handle that?” Credentials questions assess achievements or qualifications through questions such as, “What certifications do you have?” Experience questions explore work history through questions such as “What were your primary responsibilities in your last job?” Job knowledge questions assess job- or industry-specific technical understanding through questions such as, “What marketing strategies would you consider using for our product?” Finally, self-perceptions questions inquire about candidates’ self-assessments through questions such as, “What are your strengths and weaknesses?” Interviews are conducted at different points in time and for different purposes, with some being used for initial screening and others being used as a later step, after the candidate passes other stages of the hiring process. Some interviews are very brief, lasting less than 5 minutes and some are longer, lasting up to several hours; the average is around 30 minutes (Thorsteinson, 2018). They are conducted either by individuals or panels, and when there are panels, sometimes different interviewers fulfill certain roles (e.g., taking notes, asking primary questions, asking follow-up questions). Some interviewers receive extensive training on how to conduct an interview, whereas others receive minimal or no training. Common types of training content include background and purpose of the interview, job requirements, rapport building, how to write questions or use existing questions, asking probing or follow-up questions, legal considerations, notetaking, how to rate responses, rating biases and errors, and using results to make hiring decisions (Campion et al., 1997). Finally, interviews vary greatly in terms of what they assess, as discussed in the next section. What do employment interviews measure? Interviews are a method of gathering information and are therefore not inherently intended for measurement of any specific competencies. Nonetheless, there are broad categories of underlying constructs or competencies that interviews tend to assess. Examination of interview content has shown that, overall, interviews most commonly target candidates’ personality and applied social skills (e.g., oral communication and interpersonal skills) (Huffcutt, Conway, Roth, & Stone, 2001). More highly structured interviews, however, focus more on applied mental skills, direct job knowledge, applied social skills, and organizational fit, whereas less-structured interviews tend to focus more on cognitive ability, background credentials (education, training, and experience), certain aspects of personality, and physical attributes (Huffcutt et al., 2001). Additional insight into what interviews measure can be gained by looking at associations between interview scores and scores on other measures. Overall, interviews have a somewhat strong association with cognitive ability (Roth & Huffcutt, 2013). More specifically, behavior interviews (focused mainly on on-the-job behaviors and job experiences) are strongly associated with measures of job knowledge, social skills, job experience, and situational judgment and are moderately associated with measures of cognitive ability (Salgado & Moscoso, 2002). In contrast, conventional interviews (focused mainly on credentials, experiences, and self-perceptions) are most strongly associated with measures of social skills (though less strongly than behavior interviews), cognitive ability, and various aspects of personality (Salgado & Moscoso, 2002). Thus, using different approaches to uncover what interviews measure, similar findings have emerged, indicating that more structured, jobfocused interviews tend to target more applied and job-related competencies, whereas unstructured, conventional interviews tend to target cognitive ability and personality. It is possible that these differences are less a function of structural differences, per se, and more due to the fact that structured interviews are typically developed on the basis of a job analysis, leading them to be more focused on job-specific qualifications, rather than general qualities like cognitive ability and personality. Regarding the two most common types of structured interview questions—situational and behavioral questions (i.e., future behavior and past behavior)—there is uncertainty about what they measure and their relative strengths. When behavior and situational questions are intentionally written to assess the same job requirements, there is only a moderate correlation between them, suggesting that they are measuring something different (Culbertson, Weyhrauch, & Huffcutt, 2017). More research is needed to understand this finding. Though there is still much to learn about technology-mediated interviews, the early findings show that when interviews are done via telephone or video, the scores tend to be lower than for face-to-face interviews, though it is not clear why (Blacksmith, Willford, & Behrend, 2016). The lower scores may reflect a more accurate assessment or a more biased assessment of candidates. More research is needed to see if the pattern of differences is reliable and if so, to examine why. Owing to the interpersonal and often unstructured nature of interviews, they appear to be affected by a number of factors that can go beyond the job requirements. Candidates naturally want to put their best foot forward and make a good impression. Some impression management strategies may be pertinent to the job, but some are not. For example, though having a professional appearance may be important in a job, it is strongly related to interview ratings, much more so than it should be for most jobs (Barrick, Shaffer, & DeGrassi, 2009). Unfortunately, physical attractiveness is even more strongly related to interview ratings (Barrick et al., 2009). The good news is that the connection between both types of appearance and ratings depends on interview structure. The association is extremely high for low-structure interviews but is significantly lower when structure is high (Barrick et al., 2009). Other impression management strategies, such as self-promotion and ingratiation, are moderately to strongly related to interview scores (Barrick et al., 2009; Peck & Levashina, 2017). To the extent that these behaviors are relevant for the job, they may serve as useful indicators of potential performance, but they may also result in interviewers being unduly influenced by unimportant factors. As is the case with candidate appearance, the role of these impression management factors is mitigated with more structure (Barrick et al., 2009). Why are employment interviews valuable? Interviews are valuable because they are fairly strong predictors of training and job performance across a wide range of jobs and industries (McDaniel, Whetzel, Schmidt, & Maurer, 1994). Most of the specific structural aspects have not been individually tested through meta-analyses, but there are a few features that have been more extensively studied. Specifically, interviews are very strong predictors of performance when they are based on a formal job analysis (Wiesner & Cronshaw, 1988) and are more structured (Huffcutt & Arthur, 1994; McDaniel et al., 1994). Broadly speaking, behavioral questions are more effective than situational questions, though both are strong predictors (Taylor & Small, 2002). For jobs of moderate complexity (which would include child welfare workers and supervisors), there are no differences in the validity of the two types of questions (Huffcutt, Conway, Roth, & Klehe, 2004). Both types of questions are more effective when they are scored with descriptively anchored rating scales (Taylor & Small, 2002). In contrast, when behavioral and situational questions are written to assess the same job requirements and asked of the same candidates, situational questions are better predictors of job performance than behavioral questions (Culbertson et al., 2017). In other words, when more extraneous factors are controlled, and question type is therefore isolated as the primary difference, situational questions outperform behavioral questions. More research is needed to determine the conditions under which each type of question is better and why. Interviews are more strongly predictive of job performance when interviewers receive training and take notes and when the same interviewer(s) are used across applicants (Huffcutt & Woehr, 1999). Interviews are equally effective when conducted by a single individual or a panel (Huffcutt & Woehr, 1999). When panels are used and the interview is structured, predictive validity is higher when the panel uses consensus to arrive at final ratings, versus making independent ratings and statistically combining them (Wiesner & Cronshaw, 1988). The length of the interview (in minutes) has not been shown to make a difference (Marchese & Muchinsky, 1993). Finally, there are no meta-analyses examining the connection between interviews and turnover, but there is a modest connection to job tenure (McDaniel et al., 1994). Relative to most other hiring methods, interviews garner the most favorable overall reactions by applicants and are seen as most favorable on several specific aspects, including interpersonal warmth and logical fit for the job, though they rank much lower in terms of the perceived scientific evidence behind them (Anderson, Salgado, & Hulsheger, 2010). Early findings for technology-mediated interviews show that applicants’ reactions are less favorable for phone and video interviews than for face-to-face interviews (Blacksmith, Willford, & Behrend, 2016). More research is needed to explore different types of technology-mediated interviews, especially as technology advances and becomes more widely accessible. Despite the significant amount of research on interviewing, there are still many characteristics that need to be more widely studied in order to reach confident conclusions about what strategies are optimal. QIC-WD Takeaways ► There are at least two dozen structural factors on which interviews can differ. ► Interviews that are more structured and job focused tend to assess more applied and job-related competencies, whereas unstructured, conventional interviews are more likely to assess cognitive ability and personality. ► When interviews are done via telephone or video, the scores tend to be lower than for face-to-face interviews. ► Interview ratings appear to be unduly influenced by candidates’ professional appearance and physical attractiveness; both are mitigated by imposing more structure on the interview. ► Impression management strategies such as self-promotion and ingratiation are moderately to strongly related to interview scores. These connections are lower when the interview is more structured. ► Interviews can be useful hiring tools that lead to higher job or training performance among new hires, particularly when they are based on a job analysis and are structured. ► Situational and behavioral questions are both predictive of job success, particularly when they are scored using descriptively anchored rating scales. ► Interviews are more strongly predictive of job performance when interviewers receive training and take notes and when the same interviewer(s) are used across applicants. ► Interviews are equally effective when conducted by a single individual or a panel. When panels are used and the interview is structured, predictive validity is higher when the panel uses consensus to arrive at final ratings, versus making independent ratings and statistically combining them. ► The length of the interview (in minutes) has not been shown to make a difference. ► Interviews are modestly related to job tenure. ► Interviews are not intended to improve turnover, and there are no meta-analyses assessing that connection. Because they lead to better performance, it is possible that interviews may reduce involuntary turnover caused by poor performance, but research is needed to test that question. ► Relative to most other hiring methods, interviews garner the most favorable overall reactions by applicants. Applicants’ reactions are less favorable for phone and video interviews than for face-to-face interviews. ► An interview should not target knowledge, skills, or situations that will be covered in training or learned on the job. ► As with all strategies used to make hiring decisions, interviews are considered tests and are therefore subject to certain professional and legal guidelines. It is recommended that agencies consult with an expert for assistance.