If Job Boards Don’t Work, Why Do Companies Screen by Computer?
Warehouse manager at operations desk on computer Photo by: John McBride & Company Inc.
I first heard Wharton Business School Professor Peter Cappelli speak grim truths about today’s job market at a conference in 2009. We then featured him in a story on unemployment.
“Maybe for the first time ever, wages are actually going down,” he told us (and you). “One of the unbreakable truths in economics was that wage rates didn’t fall.” But now that they have fallen, said Cappelli three years ago, “The model that most U.S. companies seem to be relying on now is a very short-term churning model. Business is up? Scramble to hire. Business is down? Lay people off.”
The demoralizing moral of the story for job seekers: “It’s more like looking for temp work than it is like looking for an old-fashioned job.”
Given sound bites like that, it’s no wonder we kept an eye on Cappelli. We were particularly struck by a column of his in the Wall St. Journal: “Why Companies Aren’t Getting the Employees They Need” and the book that it inspired: “Why Good People Can’t Get Jobs.” That book led in turn to our recent story about the futility of online job search.
Given the popularity of posts by Nick Corcodilos, another key protagonist in our job search story, we thought you might want to hear from Cappelli as well. In light of the apparent futility of computer job searches, we posed the question: Is nothing gained by screening applicants online? Here’s the response:
In my book, I describe the common experience of people who appear to be highly qualified for jobs who nevertheless are screened out of the hiring process by computerized application systems. The idea behind these systems is to filter out the candidates who do not have the basic qualifications for a job and then turn those who do over to recruiters for a closer look.
In practice, though, employers have tried to get those systems to take over the entire hiring decision, getting rid of the human recruiters altogether. The hope is that a long list of requirements, often generated by hiring managers with high expectations, will mean that the perfect candidate will come out the other side. What happens instead is often that no candidates get through the screening process. The few that do usually are already employed somewhere else doing a job with exactly the same job title as the one being filled.
So when we hear that practices based on “big data” are coming to the hiring process, it sounds like more of the same. “Big data” might more accurately be called “data mining,” where we are hunting for relationships in data without a lot of guidance as to where to look. In fact, it is a fundamentally different process than what goes on now and one that is better both for employers and for job candidates.
Most of us grew up with hiring processes that seemed pretty informal: A personal contact leading to some interviews and then to a job offer, little or no data analysis involved. So it is surprising to hear that this was not always the case. A generation ago, a candidate for a white collar job in a big company would have found themselves going through days of testing – IQ tests, skills tests, interviews with psychiatrists, you name it. That died off in part because companies were no longer making lifetime hiring decisions, so the benefits of careful screening weren’t so big, and because the new threat of being sued for discriminatory hiring practices caused them to back away from formal hiring practices in favor of informal practices that were more difficult to track.
Now a lot of those hiring tests are making a comeback, in part because they can be done online at significantly lower cost. Perhaps surprisingly, the interest has begun not with management jobs but at the lower end of the labor market, in call centers. One reason why is that these call centers hire so many people, largely because of high turnover, that it could pay to put in place the initially expensive screening systems. The other reason is that it is relatively easy to track individual performance in these jobs, which in turn makes it easier to use data mining techniques to determine what constitutes good performance.
The reason that data mining is a good thing is because it is trying to establish what actually predicts good job performance. The earlier applicant tracking approaches typically assumed that only applicants who had already done the job in question would be good candidates or they relied on the instincts and gut feel of the hiring managers: “We need someone with a PhD for this job.”
Data mining, in contrast, is trying to identify in practice what factors actually predict good performance. In part, it is finding things that recruiting and staffing experts from the 1960s already knew: factors like IQ and personality can help find the better candidates, for example. But it is also finding new relationships, like the fact that having done the job before may not matter much in predicting success.
There is always a risk of messing this up, of course, and the easiest way would be for employers to rely on only one result from data mining and conclude that they should only use IQ, let’s say, to assess candidates. But in practice, even the best of these statistical relationships account for only a small amount of candidate success. A lot of information is needed to make good hires, and a lot of human judgment is required as well, especially where the different pieces of information do not point in the same direction. On the other hand, looking at actual relationships between individual attributes and performance is a big step beyond going with gut feelings.
The idea that hiring should be about the plucky applicant who persuades a human recruiter may sound appealing. But with thousands of applicants for most every position and a lot of money on the line, it is not surprising that employers are once again turning to more sophisticated analyses to try to solve the problem.
This entry is cross-posted on the Rundown– NewsHour’s blog of news and insight.