Nearly all selection processes involve the famous trio: application form, references and interview. Both the selectors and the candidates seem to like it that way, which probably explains why it continues with some modification despite the evidence about its many drawbacks.
For many the selection interview is an impression management battleground, where selectors buy books on “killer questions” to “get beneath the skin” of the interviewee. The latter buy books of “smart arse answers” to confuse the interviewer. It is, of course process of smoke-and-mirrors, subterfuge and play-acting: which is why certain individuals (narcissists and psychopaths) like them and do well, and often get selected.
We know that older, more socially confident, manipulative people rather like interviews. They shun tests and assessment centres, preferring to show off their well-honed charm to less sophisticated interviewees. They can do “apparent sincerity”, and glib, politically hot and sensitive phrases positively role off their tongue
So why are interviews so poor at predicting performance? There is a long and growing list of factors that collectively explain this.
- Interviewers differ in insight, skills, preferences: some have got skill and insight and not others. Training helps but some are naturals with real insight.
- Interviewers are not as logical and rational and able to deal with complexity as they suppose; interview confidence is often negatively correlated with interview competence.
- Interviewers’ motives, attention and need for justification of their decisions differ widely which leads to problems of reliability: very different candidates are preferred.
- Interviewees nearly always portray a positive (unrealistic) impression by self-promotion and self-enhancement; it is called impression management, dissimulation and faking.
- Interviewers often look for, and pay attention to, novel and quirky and negative information which they weigh too heavily in their judgements.
- Interviewees are increasingly being coached on how to behave in interviews, so you are measuring performance skills not the real person.
- Variations occur in how interviewers use any rating scale used during or after the interview; some give extreme scores while others never do.
- Interviewers make up their mind far too before the interview or too soon into it: they are highly selective in the things they notice, remember and rate. They can’t weigh and integrate information well like an AI program.
- Interviewers cannot cope with the complexity of integrating the information they collect: people are capricious and inconsistent, which interviewers cannot understand or resolve inconsistencies….and people are inconsistent
- Interviewers are susceptible to forming a first impression and ignoring later data: the well known first impressions effect which can be firmness of handshake, perfume used or the subtle exposure of jewelry or tattoos
- Even relatively little pieces of negative information (failures) has a great impact: Reasons to reject (i.e. select out) factors have disproportionate weight compared to select-in factors.
- Interviewers have their own (wrong, unproved, bizarre) implicit personality theories (i.e. red heads are intelligent, rugby players are good team workers).
So many have come up with a list of recommendations such as:
Prepare all questions on a thorough, up-to-date job analysis.; Ask exactly the same questions of each candidate; Do not allow questions from the candidate until after the interview (when the data have been collected); Rate each answer using multiple rating scales.; Use detailed anchored rating scales and take detailed notes; Use multiple interviewers where possible.; Use the same interviewer(s) across all candidates and provide extensive training to enhance reliability; Use statistical rather than clinical prediction.
Other recommendations include: Restrict the range of the interview to selection (not recruiting) and focus on the most job-relevant; Rely on multiple independent reviewers; Train interviewers in the process of the selection interview.
In many ways this is old hat but still ignored
Small things that make a big difference
Three is a wonderful literature based on good studies albeit with a “fun-and-games” approach that show how much little things impact on interview ratings
Consider three unrelated areas:
Smell: Does perfume, after-shave or body odour have an impact, consciously or unconsciously, on interview ratings? There have been a few studies in this area. Some guinea-pigs on a interview training course were asked to rate smells, finding the most and least popular. The next day they interviewed (unbeknown to them) an actor, who played the same role for each selector while wearing a popular, unpopular or no perfume. And guess what? When wearing the preferred perfume the actor was rated as more socially skilled, conscientious, and trustworthy,
There seems some evidence that women are more sensitive to these cues than men. And, of course, BO is a big no-no. Note the claim that Ms Markel chose Diana’s personal perfume to lure the young Prince.
Handshake: The body language gurus have described all sorts of handshakes, such as the “dead fish”; the “knuckle grinder” and “the stiff-arm thruster”. The “glove” or “politician” handshake is where one person tries to cover the other’s (opponent’s) hand completely. The desirable shake? Firm, with a strong and complete grip with (moderately) vigorous shaking for around 5 seconds while maintaining total eye contact.
Voice: There are several features in a voice: pitch, variability, loudness. Some people seem to have “nice voices”. Some actors are famous for their “rich and dark” or “sexy and sultry” voices. It is their signature: their meal ticket. And so it has been shown at selection. The attractiveness of the voice has an important, subtle and consequential impact.
And accent? Don’t even go there for embarrassment? Play people tapes with exactly the same responses to questions, even in the same voice, and you find a “strong accent effect”.
Automated Digital Interviews
Traditional interviews are very expensive in terms of time and money. Usually the interviewee has to travel, sometimes long distances to be interviewed by one or more individuals who have to make time for the process. Unless audio-recorded, which is rare and seen as a sign of paranoia, there is no reliable data about what went on during the interview which could be challenged in a court of law.
Even with structured interviews where people are asked the same questions it is possible that the verbal tone and non-verbal behaviour of the interviewer could be importantly or subtly different eliciting different reactions in interviewees.
The advent of automated interviews claims to offer many advantages over more traditional interviews. Usually this involves an interviewee responding on camera to a set of standardised, pre-recorded (text, audio or video) questions. These interviews are recorded and analysed by a whole range of techniques ranging from state-of-the-art AI technology or more modest techniques. It is possible to do an analysis of the verbal, vocal and non-verbal aspects of the recorded answers. Indeed, this methodology has become so widespread that various HR technology providers offer their services to do this.
From the recorded interviews it is possible to measure a very wide-range of particularly interesting variables ranging from facial expressions to vocabulary and word-speed to latency of response. They might even be used to measure body temperature changes, and in time, other physiological methods. These interviews are recorded and thus may be used at any time to link data from them to any subsequent behaviours.
Everything about their supposed advantage depends on a number of things. First, standard problems with interviews: the honesty of the interviewees, impression management, acquiescence all of which make the interview unreliable in the sense that the data are inaccurate and unreliable. Second, that the questions asked are perceptive and high yield. Third, that of the many variables measured, the most relevant (to the selection task) are chosen. Just because it is recorded is not good enough
One paradoxical finding at the moment is however reactions to digital interviews. A number of studies have shown that candidates do not like them and that they lead to a negative perception of the organisation they are applying to. Participants report feeling less in control and that they have an impression of lower social presence and fairness. Interviewees express concerns with privacy. Yet in a study where students responded online to voice recordings, results showed that they engaged in less deceptive impression management, provided shorter answers and felt that they had fewer opportunities to perform at their best.
Early studies in this area also showed that the more positive the interviewee was about the usefulness and ease of use with the technology the happier they were with it. Thus it seems that as this technology becomes more sophisticated, user-friendly and wide-spread the more it will be accepted by the candidate
Faking. Lying and Dissimulation
The biggest issue is that people lie in interviews (as well as on application forms and questionnaires) Max Eggert has listed a number of typical types of lies.
White-Lies: “I am a totally committed team player”. “I have excellent social skills and the ability to read people”. “I am utterly trustworthy and loyal”.
Altruistic Lies: These attempt a cover-up, but look as if they are helping others. So rather than say they left their last job because their manager was a bully, or the company was patently dodgy, they say they resigned to look for new challenges.
Lies of Omission: The most common and easiest of lies. People might omit details of school or university grades because of they had poor marks. Whole periods of their life are obfuscated. The most common lie concerns dates, often to disguise the fact that the candidate seemed to spend a surprisingly short amount of time in a succession of jobs. It is no more not less than concealment.
Defensive Lies: The defensive lie is one that conceals by generalizations or vagaries. Ask a person about their previous boss’s management style, their reason for leaving or their health record and you are often faced with a string of vague expressions such as “like others in the company”; “much the same as my co-workers”; “at that time”. Ask vague questions you get defensive lies.
Impersonation Lies: This is also called the transfer lie and occurs mostly where people take credit for others’ work. Statements such as “I doubled sales over the year”; or “I was responsible for a budget of over three million.” All others in the hierarchy are forgotten in these lies. And it is difficult to establish the facts often as to who exactly was responsible for particular successes (and disasters which are, of course, omitted).
Embedded Lies: This is a clever subterfuge to confuse the interviewer. So “I really enjoyed my time in Oxford” could refer to a first job in the City of Dreamy Spires where s/he was a mere underling. The idea is to suggest than an experience, qualification or achievement was very different from the actuality. “It was good fun being with the BBC” could mean practically anything from “I once went to a show there” to “They filmed at my school”.
Errors of commission or fact: This is lying 101. They are explicit, verifiably, false claims. It is about claiming qualifications you don’t have; starting up or working for companies that never existed; skills that don’t exist. It is the most blatant form of lie.
The Personality of Interviewers
How does the personality of the interviewer may play a very significant role in the whole interview process. So do different types/personalities make different decisions on the same people given the same criteria?
Extraverts usually enjoy interviewing. They are sociable; eager to be amused and entertained and entertaining. Extraverts probably talk too much and listen too little. They may not do their preparation as thoroughly as they should. They may be impatient and inattentive in long interviews.
Introverts make diffident, interviewers. They pause more, seeming hesitant, when they are processing information. They often find the process tiring and intimidating and feel more for those candidates who are similar to themselves. They usually take the data gathering more seriously and see the whole interview less as a social occasion than a semi-scientific exercise. Certainly, the introverted candidate probably gets a “better deal” (more favourable hearing) from the introverted interviewer.
Poorly adjusted (neurotic) interviewers are, by definition, hyper-sensitive to real and imaginary threats. They are stress-prone and may see people in general as threatening. Neurotic interviewers can easily feel threatened by the potential “mover and shaker”. They listen carefully to the candidates’ answers to questions about work-life balance, diversity, counselling and other issues.
Stable interviewers, like stable employees, are better news. They are less moody and better able to weigh the information. They worry less about what might go wrong and cope with all the little dramas at interviews well. They tend to be calm, focused and rational.
Agreeable interviewers are warm, empathic and trusting. They are for the most part likeable. They understand that interviews can be stressful. They are concerned about making the candidate comfortable, relaxed and able to be their real selves. They are slow to chide and swift to bless and believe they get the best out of others by giving them a chance.
Less agreeable and likeable interviewers believe you understand people best by “putting them on the spot”. They treat the interview as a “Paxman-inspired” political interview. They cross-examine individuals, often pushing them to give details of success and failure which their CV overlooks. They are hard to please: cynical, tough, world-weary and they care little for interviewee comfort.
Conscientious interviewers are not only conscientious about how they approach the task of interviewing, but also what they are looking for. Hard work is a virtue. Some are even prepared to “trade off” ability for the work ethic: preferring the loyal plodder to the capricious wunderkind. Conscientious interviewers are concerned that the applicant follows orders, obeys rules, and has a sense of duty.
Less conscientious interviewers want to have fun. They tend to be less achievement-orientated, less careful and with a much weaker work ethic. All that “postponement of gratification” stuff never worked with them. They prefer what the Freudians call “the pleasure principle”. They seek out playmates more than solid and reliable colleagues.
The ambition and achievement needs of interviewers are also relevant. Paradoxically, both the low and high ambition interviewer may be intimidated by the obviously ambitious candidate. Those with low ambitions can feel intimidated by newly minted MBAs who want to be on the board at 30 and retired at 40. The highly ambitious see a potential threat.
What of the abilities of the interviewer? How are bright, educated interviewers different from their less talented peers? Another paradox: the clever prefer discriminating questions, the dim prefer “clever” questions. Brighter people tend to have a bigger vocabulary and think fast. They ask good questions which sort the wheat from the chaff.
The less bright and less educated interviewers might rehearse “killer” questions that make them appear intelligent, even if they cannot process the answers. They can be intimidated in group interviews and behave badly. They often have “crackpot” theories, refreshingly evidence-free about desirable characteristics in candidates.
The interview is a social process. It can be a sophisticated intellectual theatrical show; a hall of mirrors; a game of bluff and counter-bluff. There is no doubt that there is a lot of ‘gut feeling’ going on in both parties, despite all their training.
So the moral of the story? First, acknowledge that the interviewers’ make-up (ability and personality) do inevitably play a part. Second, try to work out how specific interviewers react to particular candidates. Third, use multiple interviewers but particularly those with the ability and personality profiles found among those. Forth, follow the rules and guidelines with respect to training interviews. Fifth, supplement interview data by references and tests. Sixth, the cost of errors (known to all those who divorce) is high so invest time, money and effort in trying to do a better job
About the Author
Adrian Furnham is Professor of Psychology at BI: Norwegian Business School. He is particularly interested in assessment, particularly the development and evaluation of psychometric tests and has recently completed a co-authored book on the Psychology of Spying.