Advertisement
But, Doctor, It Just Won’t Die!
Get articles like this
in your inbox
The longest running and most trusted source of information serving talent acquisition professionals.
Advertisement
Related Articles
When an old king dies and a new king is crowned, his subjects acknowledge the continuity of power by chanting, “The king is dead. Long live the king!” That is, the old King is dead, but one breath later, the new King continues the tradition in an unbroken line. The same is true of the interview. The interview is dead. Long live the interview. And, the interview tradition continues and continues and continues, ad nauseum. Odds are, even though you intuitively know that interviews are a poor selection tool, you still use interviews to screen job applicants and hire new employees. But, how do you reconcile the fact that study after study shows unstructured interviews have no predictive accuracy! That’s right. You don’t have to waste time with face time. Line up applicants and flip a coin to get the same results! But you don’t need a research scientist to tell you that. All you have to do is walk around your organization and mentally separate people you interviewed into a “keep ’em” and “lose ’em” group. If the “lose ’em” group is close in size to the “keep ’em” group, it is time to accept some personal responsibility. Your hiring interview couldn’t tell the difference – “lose ’ems” looked just like the “keep-em’s” before they were hired. The King Is Dead, Long Live The King OK, we see a large group of “lose ’ems” walking around. Why didn’t we spot these people earlier? We really got to know them during the interview. And they looked good! Well, I think some researchers just might have discovered the answer. And it’s like a highly addictive recruiter’s drug. In Volume 53 of Personnel Psychology, Murray Barrick, Greg Patton, and Shanna Haugland reported results from their study of a group of 12 experienced interviewers. Each of these interviewers had over 12 years experience, were members of SHRM, and had received extensive interview training throughout their careers. Basically, they were about as seasoned and experienced as an interviewer could be. The study is very detailed, but here is a short list of some of its main points:
Ok, so here you have it. Twelve experienced interviewers meet 73 interviewees using good interview techniques. Now the $100,000 research question, “Can very experienced interviewers predict the two personality factors that are consistently associated with job success?” A Little Background, Please Normal personality factors can be computer-clustered into five general factors called the Big Five. These are:
Of these five, two have been consistently associated with high performance in all jobs. They are conscientiousness and emotional stability. No earthshaking news here – just a recognition that your basic high producer cares about work and is sane. Interviewer Accuracy So was our professional interviewer team able to identify these critical traits? Nope! The only thing these folks were able to consistently and accurately measure were traits associated with “getting to know the applicant.” That is, extraversion, agreeableness, and openness to experience. They completely misjudged the two biggest traits associated with performance: conscientiousness and emotional stability. I think this is a primary reason why, as inaccurate as they are, interviews just won’t die! A recruiter who relies solely on interview technology achieves one objective – he or she gets to know the applicant. But knowing an applicant is a long way from determining whether that applicant will be any good at the job. Recruiters addicted to “getting to know you better” meet their own emotional needs, but entirely miss critical data. This is a “good news/bad news” joke. The good news is the recruiter may feel very good about their hiring decision. The bad news is that feeling is totally meaningless to the real objective – acquiring good performers! What is the impact on the organization? A big group of “lose ’ems”! What’s With Behavioral Interviewing? You might have noticed that interviewers used behavioral (i.e., background, behavior, and consequences) and situational (what would you do if..?) interviewing techniques. These are both well researched, and yield about 10% more predictive accuracy than casual interviews. But why didn’t they pick up the critical traits? Well, because they were not designed to do that. Behavioral and situational interview technology helps you learn whether the person has the competency to perform the job, not whether the applicant has the right attitudes, interests and motivations. Even the best of the structured interview techniques is a very poor measure of conscientiousness and emotional stability. Basically, it falls out this way:
Don’t Try This At Home Research results must always be put back into context. Personality traits are an important part of performance, but not the only thing you need to measure. You also need to measure the job’s required cognitive ability, planning ability, interpersonal skills, and job fit. You wouldn’t hire a musician without an audition, would you? Why not use that same logic to hire managers, professionals, customer service associates, executives, etc? When people aren’t familiar with the basics of measurement, they tend to build websites and systems that are visually appealing, but ineffective. In a valiant attempt to make their process more accurate, they use canned job descriptions, worthless tests and meaningless application blanks. Your Role There are literally thousands and thousands of selection studies that show the way toward better hiring practices. If you have not read them, then read about them. If you don’t have time to read them, hire an expert to do it for you. Do you see yourself as a finder of bodies for managers to interview? Or as someone who is expert in human skills measurement? More to the point, you control the front door to your organization. Are you letting in a lot of flies? Let’s all say it together now. The interview is dead! (Let’s just forget the next part.) <*SPONSORMESSAGE*>