|In this article:|
|The psychology of psychological testing|
But after triumphs and failures, good nights and bad, you realize there is no 100% guarantee. Sure, the lullaby tape works... sometimes. And the new storybook works... if your child feels like reading it that night.
While we all know in our hearts there is no silver bullet, the idea is compelling. And it might have something to do with why more organizations these days are considering the idea of pre-employment testing for their IT candidates.
Whereas 18 to 24 months ago you had a month for the hiring process, you¹re now looking at a two-week period, maximum, says James Essey, president of TemPositions Group of Companies, a temporary staffing firm in New York City. In the last six months, even that timeframe has collapsed, with companies making job offers right after the first interview, he says.
In fact, the number of large- and medium-size companies that require job applicants to take a psychological test rose from 39% in 1997 to 48% last year, according to the American Management Association, which surveyed 1,100 of its members. The tests assessed everything from managerial abilities to personality traits to career interests. Meanwhile, nearly 65% of the respondents said that they test applicants' job skills, nearly the same percentage as the year before.
Tests provide applicant insight
There are tests available in the IT world that measure how deeply job applicants really understand the technology areas listed on their resumes. For instance, TeckChek, a leading provider of technical proficiency tests, offers tests for over 100 skill areas, including C++, Oracle databases, and SAP R/3. Applicants sit down at a PC for about an hour and respond to multiple-choice questions, many of which use actual computer code. The test taker is permitted to select up to three of five possible answers, some of which are more insightful than others. This technique is supposed to help demonstrate the candidate's range of knowledge.
In addition, the test is adaptive--the questions get harder or easier based on answers to the previous questions. Once the candidate hits his maximum level of proficiency, the test ends.
Once TeckChek scores the tests, employers are given both raw scores and percentile rankings that compare the candidate's performance to that of all other professionals who have taken the test. You also get a detailed profile of the candidate's strengths and weaknesses.
According to Alan Epstein, president of TeckChek's parent company, Knowledge Testing Enterprises Inc., in Los Angeles, the tests provide employers with insight that just isn't available during a regular interview. For instance, a candidate might say he's certified in Lotus Notes. Well, says Epstein, "that's as if you and I applied to Harvard, and a testing service says we both passed our SATs. It provides assurance of some minimum level of competence, but it doesn't say much about the candidate's relative level of competence and his strengths and weaknesses."
Consider the possibilities--no more in-depth interviews! No more extensive reference checks! No more costly hiring mistakes!
Applicants are people, too
Well, not quite. Unfortunately, hiring involves a lot more than skills assessment. "The person could get the best results in the world and still be a jerk to work with," says Lina Fafard, a branch manager at Montgomery West, an executive search firm in Torrance, Calif. "They can code on water, but if the chemistry isn't there, they won't be hired."
And therein lies the danger of testing. "It can be too tempting to say, if you're not above the 50th percentile, we won't hire you, and certainly some of our clients use it that way," Epstein admits. In fact, some of TeckChek's clients have done away with technical interviews all together.
However, even Epstein says TeckChek is just one piece of the hiring puzzle. "If you look at our contracts, there is specific language that says, 'don't rely on the test score exclusively.' It should be one of a package of tools, including interviews and reference checking," he says.
For instance, a typical company that uses this type of test might otherwise pull its best database professionals off line to conduct technical interviews. Not only are these employees unproductive during this time, but if you have three or four people conducting the interviews, they're applying different standards, because it's highly unlikely the company has defined a set of standards for questions to ask and what answers will be acceptable.