Rethinking Interviewing and Personnel Selection

T. Oliveira

• Publisher: Palgrave Macmillan
• Year: 2015
• ISBN:
  9781137497352
• Topic: Business
• Subtopic: Management
• Index: Business

The case studies in Rethinking Interviewing and Personnel Selection find support for Herriot (1993, 2003) and Fletcher's (1997, 2003) claims that the selection interview is a social process which may gain from a degree of semi-structured interaction with candidates.

1

What Selection Theory Claims

Mainstream selection theory is normative in claiming that an interview should be premised only on overt criteria; that managers as selectors should rely only on inference from candidates’ attributes and avoid sensing, feeling or intuition in decision-making. Such theory is the ‘highway code’ of selection methods. In particular, no interviewer should ‘go it alone’ and trail ‘off road’ in semi-structured or unstructured dialogue with candidates that may involve questions not put to all of them in the same way.

A paradigmatic example of this is Robert Dipboye (1996), who has advocated that selectors should (1) focus on knowledge–skills–abilities needed for the job, (2) ask candidates the same questions and use the same rating scales, (3) develop scoring keys for evaluating applicant answers in behavioural terms, (4) use more than one interviewer, (5) eliminate extraneous conversation with the applicant, and (6) explain to candidates that they cannot ask questions.

While Dipboye (1996) nonetheless admitted that the dilemma is how to do this without becoming ‘Orwellian’ in a quest for standardisation, such theory has a central problem in that its own advocates admit that managers as selectors tend to neglect it. The neglect appears to be so widespread as to suggest that it cannot be due simply to incompetence or lack of training. Concern in normative theory about how managers should select tends to displace not only the degree to which they may think and act differently from the theory, but also why they do so.

This chapter assesses the scope and limits of normative theory and the long-standing concern of some of its advocates such as Dipboye (1992, 1994, 1996, 1997) with knowledge, abilities and skills for explicitly defined job fit. Yet, it also suggests that while frequent reference is made to person–organisation fit, there is a need to deconstruct the concept in terms of different operational and organisational contexts and to relate this to what may be different stages of a selection process, with differing explicit or implicit rationales.

It illustrates that there are limits to simulations of selection, often with students, of what criteria are important to attract or retain candidates in organisations as well as that the range of criteria that normative theory recommends that managers as selectors should consciously correlate not only is difficult but may be impossible.

The normative mission

In identifying the interview as the focal point of the selection process, Guion (1965) stressed that it needs to be conducted in as ‘constant’ a manner as possible for all candidates who should be asked the same questions in, as much as possible, the same way. But he also claimed that ‘the responsible interview is not merely another source of data. It has a unique function in that it is where data from various sources, including the interview itself, are integrated’ (ibid., p. 39) and that this final ‘integrating judgement’ should include both explicit and implicit factors.

But normative selection theory rarely has followed Guion’s recognition of a distinction between explicit and implicit factors, or that how managers come to selection decisions may combine both conscious and unconscious cognitive processing. Nor has the theory acknowledged that the range of criteria which it advocates that managers should consciously hold in mind may be not only an invitation to inferential overload, but impossible.

For example, as outlined in Box 1.1, a pioneer of structured interviewing, Rodger (1952), combined seven main sets and sixteen sub-sets of criteria for selection. Munro Fraser (1978) reduced the main sets of criteria to five, but then offered nineteen sub-sets. If there is a five-point scale for ranking each criterion, whether a selector could consciously correlate or integrate these in the manner claimed by Guion (1965) either during an interview or in post-interview evaluation is open to question. Robertson (1994) is blunter in suggesting that this is not only improbable, but impossible since such integration would imply not only hundreds, or thousands, but billions of correlations.

Box 1.1

Inviting inferential overload

Normative selection theory advocates conscious and consciously retained criteria in candidate choice. Thus, an early advocate of the theory, Rodger (1952), identified seven sets and relevant sub-sets of criteria to evaluate applicants:

•		physical make up – appearance and physical health;
•		attainments – general education, vocational training and professional qualifications;
•		general intelligence – overall cognitive ability measured in psychometric tests;
•		special aptitudes – specific abilities and attainments;
•		interests – spare-time activities, sports and hobbies;
•		disposition – motivation, personality and acceptability to others;
•		circumstances – family life and general way of life.

Munro Fraser (1978) reduced these seven sets to five, each of which also had its own sub-sets:

•		impact on others – appearance, speech and manner, and health;
•		qualifications and experience – general education, vocational training and professional training;
•		innate abilities – verbal, perceptual, numerical, mechanical and spatial;
•		motivation – level of goals, realism and consistency in following them up;
•		emotional adjustment – acceptability, sense of responsibility, reliability and potential leadership.

Rodger therefore offered seven main sets and sixteen sub-sets of criteria. Munro Fraser offered five main sets and nineteen sub-sets.

Whether any manager could consciously hold all of these in mind is open to question. For them to rank each on a five- or ten-point scale and then integrate them by the end of an interview not only would invite inferential overload but may be impossible (c.f. Robertson, 1994).

Principles, norms and practice

Besides which, in contrast with the claim of normative selection theory that in principle all interviewing should be highly structured, a range of evidence indicates that semi-structured or unstructured interviews may be the norm in practice (e.g. Robertson & Makin, 1986; Anderson & Shackleton, 1990, 1993; Dipboye, 1992, 1994, 1996, 1997).

In a comprehensive meta-analysis, McDaniel et al. (1994) found that when the criterion was suitability for training performance, the validity of structured and unstructured interviews was similar. Yet Herriot (1993, 2003) and Fletcher (1997, 2003) have stressed that the selection interview is a social process which may gain from a degree of semi-structured interaction with candidates. This is also consistent with the earlier claim of Bakhtin ([1935] 1981) that open-ended dialogue may give rise to new insights and meanings, which informs the case of Altink et al. (1997) for open-ended discourse rather than only one-sided questioning, assessment and judgement by selectors.

Furthermore, according to Hackman and Oldham’s (1980) Job Characteristic Model, the interviewer may gain better results by semi-structured procedures for candidate fit in terms of needs for autonomy, variety, sense of purpose and task significance. This suggests that semi-structured, non-consequential or lateral questioning may provoke responses from interviewees which may be more revealing than those within a structured approach, to which we return.

Anderson and Shackleton (1993), like Dipboye (1996), have recommended that selectors should be trained for interviewing so as to rationally gather and process information in order to avoid bias, and then objectively rate the applicant’s answers. They define four major responsibilities for the interviewer: (1) evaluate each item of information; (2) allocate it an appropriate weight in decision-making; (3) combine multiple sources of data in order to (4) reach a final accept–reject decision.

They also make five key recommendations in relation to the context and environment of an interview. First, its conduct should combine formality, comfort, tranquillity and courtesy. Second, general documentation is needed in terms of previous examination results and curriculum vitae, the results of a candidate’s psychometric tests and references received before the interview. Third, there should be mutual understanding of the aims of the interview. Fourth, this should be matched by structured question generation and hypothesis formulation. Fifth, there should be a follow-up assessment on the validity of predictors.

There is substance in some of this. Managers may be concerned to give information in a structured manner to be able to assure themselves that all candidates grasp what the job means in both operational and organisational terms since they do not want to select and train people who then claim that they did not understand what a job entailed and consider this a breach of psychological contract, which is an issue touched on in this chapter and relevant to some of the main findings from case studies reported later.

Yet whether there can be mutual understanding of the aims of the interview, based on Dipboye’s (1996) recommendation that a candidate should not ask questions, is less than clear. Anderson and Shackleton’s (1993) first reco...

Table of contents

• Cover Page
• Half Title Page
• Title Page
• Copyright
• Contents
• List of Case Studies and Boxes
• List of Figures
• List of Tables
• Preface
• Acknowledgements
• About the Author
• Introduction
• 1 What Selection Theory Claims
• 2 Who Knows for a Fact?
• 3 Yet How Do We Know?
• 4 What’s the Logic?
• 5 Where’s the Proof?
• 6 Why Dismiss Intuition?
• 7 Interviewing and Psychological Contract
• 8 Tacit Knowledge and Implicit Learning
• 9 Rethinking Selection Theory
• 10 What Managers Have in Mind
• 11 Power and Panel Interviewing
• 12 So Where Now?
• Annex: Sets-within-Sets of Criteria in Panel Interviewing
• Bibliography
• Index