Recruiting science: The structured interview - On efficiency, validity, and diversity.

Viktor Nordmark
1 November, 2021

In a recruiting context, how many times have you heard the following phrase from someone in your proximity?

”I’m fully capable of deciding whether a candidate is a good fit for a job just by casually talking to him/her.”

This blog post is specially crafted for those who identify with the above statement or simply want to know more about how to predict future job performance.

In the real world, you are about as likely of identifying the best fit by casual conversation as you are by arranging a candidate lottery. And yes, that goes for all, no matter how much of a people person you might be. The tendency to overestimate gut feelings is as common as it is detrimental in hiring decisions.

Even if miracles do happen and it’s true that you can find great people with no further assessment than a sensible conversation. It’s highly unlikely that this practice, over time, will lead to something other than bad hires. And bad hires is not only bad for business, it’s disastrous for business. Just look at some of these numbers:

1.Up to 80% of employee turnover has bad hiring decisions to blame.

2.Tony Hsie, CEO of Zappos, estimates that bad hires have cost his company “well over $100 million”.

3.The US Department of Labour says that the cost of a bad hire is at least 30% of an employee’s yearly salary.

And there’s more to it than just direct economic effects. For example, bad hiring can also lead to a lousy culture, decline in morale, lower productivity, and poor customer satisfaction to name a few things.

To substantiate and make your hiring decisions more objective, what you need is pre-hire assessments with high scientific validity. Implementing them does require some added work, but if you’re thinking of skipping a few steps, just go back and re-read the numbers in the previous section again.

Fortunately, assessment methods have been rigorously examined by behavioral scientists for the past 100 years. One of the most recognized contributions to our understanding of this area is a meta-study containing findings from over 80 independent studies.

Validity and Utility of Selection Methods in Personnel Psychology Practical and Theoretical Implications of 85 Years of Research Findings, Psychological Bulletin, Vol. 124, №2, 262–274

In terms of validity for different methods of predicting future performance, below is a representation of the combined findings from these 85 years of research.

Structured vs. unstructured interviews

As evident, the structured interview is one of the very best methods available for predicting future performance. What differs a structured interview from an unstructured interview is, as the name implies, mainly a rigid, standardized structure that makes one interview comparable to another.

In clear terms, what this means, is that every candidate is treated in the exact same manner, asked the exact same questions in the exact same order. The responses are then analyzed against a predetermined grading rubric that makes it possible to score each response in an objective way.

For example, a common way of standardizing the evaluation of responses is to ask the candidates to structure the responses according to the STAR model (Situation, Task, Action, Result)

Three things that structured interviews really got going:

Low bias

In an unstructured interview, it’s very easy to favor candidates similar to oneself, which over time, will lead to a homogenous workforce with narrow perspectives.

Great at indicating performance

Structured interviews are more focused on competencies, which by default is far more objective and less dependent on gut feelings.

High candidate satisfaction

Google’s internal research shows that rejected candidates are about 35% happier after participating in a structured interview compared to those who did not.

Drawbacks to the structured interview

Everything comes at a cost, and the structured interview is no exception to that. These are the main drawbacks you need to consider:

Resource intense

Interviewing one candidate requires one recruiter. If you have a massive influx of candidates, it can be challenging to give everyone an equal chance.

Logistically complex

With meetings in set time-slots comes an array of issues. People will reschedule, cancel, ghost and be late etc, etc.

Not entirely bias-free

A soon as two humans meet, bias will inevitably be present and might influence the evaluation no matter how structured the interview is.

A take at improving the structured interview

When it comes to high volume recruiting there’re not a lot of options when it comes to good pre-hire assessments that are both scalable, user friendly, and deliver a high validity.

As great as the structured interview is at indicting performance, it’s not the best when it comes to high volume recruitment due to its time-consuming nature.

With the recent advancements in artificial intelligence and Natural Language Processing, some companies have started to explore the possibilities of automating the structured interview.

The advantages are pretty straightforward:

-Up to 90% resource savings compared to a manual process

-Virtually bias-free (no human interaction)

-Great candidate experience and convenient for both candidates and recruiters

And in terms of structure, it’s hard to be more structured than a machine trained to perform the same task in the same set order.

For the next step, the interview is coupled with a scoring engine that looks into the performance of the candidate’s interview and assigns a score based on the requirements for the position.

To gain more knowledge about how this works in practice, just shoot us a message or schedule a live demo right here.

Viktor Nordmark
1 November, 2021

Feautured Post

AI Recruiting: A Guide for People People

Read post