I am currently reading a great book, “Thinking Fast and Slow” by Daniel Kahneman, an Israeli-American psychologist and Nobel laureate. He is notable for his work on the psychology of judgment and decision-making. Once again reading Kahneman’s work reaffirmed why the employment interview is such a dangerous hiring tool.
All organisations use the face-to-face interview when selecting new staff. There are many types of interviews ranging from loosely structured conversational interviews, to highly structured interviews. Some organisations use a telephone interview first to check details in addition to the Application Form as a screening tool. Doing so helps to reduce the shortlist.
The most popular interview is the unstructured interview. It is highly unreliable and runs the risk of the interviewer asking questions that are not legally defensible. Unstructured interviews are not consistent and extremely open to personal bias – more on this shortly.
The unstructured interview ranks the lowest of any tool used in the selection process, yet it is the most popular! The validity of an unstructured interview is between .05 and .15 – that means, at very best, about one in every six interviews will be accurate in identifying the “real” person. These are very poor statistics when you are putting the expense of an annual salary on the line.
The purpose of an interview is to collect information that will show to what extent the applicant has the knowledge, skills and abilities sought after for the role. The applicant’s responses to the interview questions will provide this information. Effective interviewing therefore depends on the type of questions that are asked and how they are interpreted and rated.
Do not use vague, unstructured questions such as:
- Tell me about yourself.
- How would your previous employer describe you?
- What are your career goals?
- What are your strengths and weaknesses?
These types of questions are open to personal opinions and anyone can give an opinion. An employment interview should be built around the key 6-8 performance factors for a position (competencies). One or two behaviourally based interview questions are usually developed for each performance factor.
The intended purpose of the employment interview is to gather information to enable us to make a judgement call and subsequent hire, or non-hire decision. However, when poorly structured and carried out by one individual, our judgements and decisions become “clouded”. We are influenced by what we think and feel as opposed to what we know. Kahneman’s book and studies gives you great insight into this concept.
Some common interview biases we should be alert to are:
- Central tendency bias– here an interviewer leans towards the middle ground when rating the individual’s answers on all performance factors.
- Leniency/ strictness bias– an interviewer consistently giving either only high or only low ratings.
- Halo effect bias– allowing the rating of one performance factor to influence the ratings given to other performance factors.
- Similarity bias– assessing applicants that are like your-self more favourably. Like usually hires like – diversity in the workplace is good.
- Contrast effect– assessing applicant in relation to the applicants previously interviewed – mentally ranking the overall candidate experience based on the previous one(s).
- First impression– making a decision in the first few minutes of the interview. Just because this person dresses well, talks and present professionally, does not mean they can perform on the job well.
- Biases and stereotypes– allowing personal biases (generalisations) to influence ratings of applicants – i.e. certain ethnicities are lazy, women with young children are unreliable etc.
- Overly sensitive to negative information. Many interviewers look for reasons to reject rather than a reason to hire.
To avoid interview bias it is recommended that all interviews be conducted using 2 or more interviewers. Do not lump all interviews into a single time block as this encourages the contrast effect. Always score each performance factor, and total up overall score immediately after concluding the interview using a consensus discussion.
A huge volume of scientific information has been gathered on the poor validity of employment interviews. Structuring the interview, using multi raters, and asking the same behavioural questions to all candidates will greatly increase your selection validity to.50 (a flip of a coin). Not perfect I know, but better than 1 in 6.
We will rarely get it 100% right, but you can create incrementally validity to the structured interview by adding a valid psychometric assessment, a finely tuned application form and a diligent background check. Then you are likely increase your odds to around 75% of getting it right. I can live with that.
You can learn more about conducting job interviews in Rob McKay’s new hard covered book, “No More Square Pegs: How to Hire Winners for Your Business.