By Marga Salvador on June 6, 2016
You've probably heard that the decision making happens during the first 7 minutes of a job interview. If you haven't, well, it does. The other 89% of the interview spent grilling the candidate is really just time confirming the perception and decision you've made. This means that questions like "Tell me about yourself," "What are your greatest strengths and weaknesses?" are utterly null.
Maybe you're thinking, that's okay, we have questions that really get to the nitty gritty of things. Many recruiters and employers, however, ask interview questions that are the situational brain twisters, like "There's a meteor shower that will hit the earth in 10 minutes, do you hide underground or head to the safe zone?" or "Estimate how many Jollibee's there are in Metro Manila." Questions like these can reveal things like logic, assessment skills, or the creativity of the candidate but for the most part, they are asked to make the interviewer feel clever or correct.
[caption id="attachment_2685" align="aligncenter" width="435"] (Source: shutterstock)[/caption]
Frank Schmidt and John Hunter did an 85 year study on how well assessments can actually predict performance. They found that unstructured job interviews can only explain 7% of a potential employee's performance (his/her work history explains 3%).
OJR: On the Job Recruiting
Your best bet is giving the candidate a work sample test which predicts a good 29% of performance. Take a piece of the job they are applying for—campaign creation for marketing, coding for IT, number crunching for finance—and have them do a bit of work. True enough, they will try extra hard to impress but this shows capacity. When you assess the test work, it also gives you an actual preview to their performance on the job.
There is a right and wrong answer
The second best predictor, which explains 26%, are tests of general cognitive ability. Instead of those awful brain twisters with no right or wrong answer, give candidates a test with right and wrong answers. "They are predictive because general cognitive ability includes the capacity to learn, and the combination of raw intelligence and learning ability will make most people successful in most jobs."
[caption id="attachment_2686" align="aligncenter" width="435"] (Source: best-job-interview)[/caption]
The structured interview
The structured interview, meaning a consistent set of questions, has also shown to predict 26% of a candidate's performance. Tied to the consistent questions is a clear criteria used to assess the answers. Oftentimes, interviews are based on mood and feel of the interviewer which has got to stop.
The two kinds of interviews are behavioural and situational. The former makes an assessment of prior achievements and whether they match what is required of the job. Situational interviews are the "What would you do if...?" ones that relate things that actually happen on the job. The structured interviews have been known to successfully assess even unstructured jobs. "A diligent interviewer will probe deeply to assess the veracity and thought process behind the stories told by the candidate."
[caption id="attachment_2687" align="aligncenter" width="435"] (Source: mirror)[/caption]
A+ for effort
The difficulty in using the formats above is that a company has to write them, test them, and stick to them. On top of that, the formats have to be refreshed periodically so that candidates don't compare notes and find the pattern. The great thing about it, however, is that you can mix and match components based on the needs of the job. You can have a cognitive test and a work sample test for one opening and a work sample test and situational interview for another.
The goal of the recruitment process is to assess and predict how well the candidate will fare on the job. It's much less of a headache to put in the effort to find someone good for the long run than to hire and fire 3 okay ones, don't you agree?