The Selection Interview:
Comparing 'Can Do' Versus 'Will Do' Questions
The interview remains the most popular selection method, used by virtually all organisations at least at some stage in their selection procedure. It is now widely accepted that an interview which follows a pre-defined structure is a more valid predictor of job performance than an unstructured interview. Recent research has focused on identifying specific aspects of structured employment interviews which may influence how valid the interview is.
In the current study, by Paul Taylor and Bruce Small (both academic psychologists), the authors review a large number of studies which look at what effect the type of questions used in the interview has on predicting job performance.
The two types of question styles that they focus on are:
-
Past behaviour questions (e.g. 'Recall an example of when you were faced with. What did you do?')
-
Situational Questions (e.g. 'Suppose you were faced with the following situation. What would you do?') These are the hypothetical questions, in which, from our experience of interview training, interviewers have considerable faith.
The results of their research showed that:
-
Both these types of structured interview were reasonably valid predictors of job performance.
-
The past behaviour interview was a better predictor of on-the-job performance than the situational interview.
-
The validity of the interviews was further improved by using descriptive rating scales.
Past behaviour might provide a better measure of job performance than future behaviour because it tests candidates' 'typical' rather than 'maximal' performance. Tests of 'maximal' job performance only assess a candidate's:
-
Declarative knowledge (knowledge of facts, rules, principles and procedures).
-
Procedural knowledge and skills (applying declarative knowledge to perform a task / procedure).
Tests of 'typical' performance assess the above, but also test the candidate's motivation to apply their knowledge and skills to day-to-day job performance. Therefore, a candidate in a situational interview might have the required declarative and procedural knowledge and skills to perform the job well. However, once they are in the job, they may lack the motivation or opportunity to perform to the best of their ability.
Human Assets Comments
The distinction between tests of maximal and typical performance is important for recruiters to consider, in particular when deciding which selection tools to use.
We recommend the use of structured interviews which ask candidates questions about their past behaviour and experiences. The questions posed by the interviewer should be developed around the behaviours required by an effective performer in the role. Interviewees who report that they have responded effectively in a past situation will have demonstrated both the necessary knowledge and skills, and also that they were sufficiently motivated to apply their knowledge/skills in that situation.
For a (past behaviour) structured interview to be most effective, we also recommend the following:
-
Well-designed competencies: The behaviours for effective performance that are assessed in the interview are typically grouped into competencies. They must be clearly defined, relevant to the current and future target role, and easily assessed.
-
Interviewer Training: The past behaviour interview will be most effective when interviewers are fully trained in best practice assessment and questioning skills. In particular, some time is needed to move interviewers away from their favourite hypothetical questions. The argument that hypothetical questions elicit hypothetical answers can now be couched in terms of maximal performance, with the empirical findings of Taylor and Small as further support for interviewers to change their ways!
-
Descriptive Rating Scale: The points on the rating scale used by assessors should be clarified by a description (e.g. '5' = Strong positive evidence of the competency).
If you would like to find out more, please contact our consultants on +44(0)20 7434 2122 or by email enquiries@humanassets.co.uk
Reference
Taylor, P.J. & Small, B. (2002) Asking applicants what they would do versus what they did d A meta-analytic comparison of situational and past behaviour employment interview questions. Journal of Occupational and Organisational Psychology, 75, 277-294.
Newsletter: 2003