By Jim Schutze
By Rachel Watts
By Lauren Drewes Daniels
By Anna Merlan
By Lee Escobedo
Now, a doctorate degree and successful entrepreneurial venture later, Helm creates and sells personality tests for companies. He uses established tests like the Myers-Briggs Type Indicator in conjunction with his own home-grown tests for companies looking to hire or promote management, as well as crisis management for companies that have a valued but unstable executive.
"The reason these tests work is because people vary on any trait you care to name," he says. "For individual characteristics, genetic variation says that some people will have a little, some people will have a lot."
Any quality, from basketball skills to a propensity to sexually harass a co-worker, will be "normally distributed" within any population; some good, some bad, and most in the middle. When plotted on a graph, a normal distribution will look like a bell--what is commonly known as a bell curve.
Testing someone's skill--how fast a secretary can type, for instance--is easy. But can something as intangible as propensity to steal be measured in a test?
Yes, says Helm. In his experience 5 to 10 percent of people will not steal for any reason. Others will steal any chance they get. The rest--the fat part of the bell curve--will steal given certain circumstances.
"We call it rationalization," he says. "We all have our own view of the world, and everyone's is different. We have constructed a 'rational' set of arguments to support our worldview. As a test constructor, I can tap into that worldview and plug into those rationalizations with questions."
Each test is designed to pinpoint particular attributes; Helm has a 56-question test designed to glean insights into a propensity for stealing, drug use, and sexual harassment. They can be pretty easy to beat if you know how they work.
Questions in that test come in two forms--validity questions to gauge whether a subject is lying and substantive questions. They're pretty easy to tell apart, if you know what to look for. An example, taken from Helm's test, would be: "I have never raised my voice in anger." Anyone who says they never have is probably lying.
Most of the questions, however, are designed to discover problematic behavior. Helm says the basis of the test is centered on the tendency for applicants to vary their responses based on what they think the tester wants to hear and what they think will be the easier answers to defend.
"I have a very pragmatic approach," Helm says. "What is this test telling me? What steps did this person take to look better?"
For example, a sexual harassment question should always be answered with a strong negative. But when Helm adds qualifiers, some people will betray their worldview by soft-pedaling the answer. For example, the question: "Patting a co-worker or subordinate on his/her shoulder or putting one's arm around his/her shoulder is OK as long as one is sincere and not making any kind of sexual advance."
With so many qualifiers, a subject may not see that the right answer is a strong negative. Two or three questions with ambiguous answers about sexual harassment leads to a pattern, and that could lead to trouble for the applicant. Theft tests are the same when applicants admit to taking small amounts of supplies or money from the office.
So who would ever answer a question about theft by admitting they stole at past jobs?
"Sometimes I look at the responses and say, 'God Almighty! Why would you say something like this?'" Helm says. "And the answer is, you think everyone feels this way."
And that leads to the most ironic complaint about honesty tests: The best way to beat one is to lie while answering. Being too honest while taking an honesty test is a good way to get bad results.
"You get these tests that ask for a black-and-white view of the world," says George Alliger, an employment consultant and former professor for the State University of New York-Albany who has published research studies regarding the tests. "All overt tests about theft or drugs ask about previous behavior, and any admission--even to being tempted--is considered a negative and counts in the dishonest direction."
It is impossible to call Alliger an unfair critic. He is quick to point out that there is "reasonably strong" evidence that these tests do indeed weed out those who may steal.
"That doesn't mean they're appropriate," he adds.
His problem lies not with potential crooks who are shut out of jobs but with honest people who have been shafted on the tests. He tells a story borrowed from a published psychologist of a nun who scored abysmally low on an integrity test because of her refusal to deny she was tempted by sin.
The moral? "It's their own honesty that gets them in trouble," he says.
Others have issues with the phrasing of questions, which leads the truly honest to answer incorrectly. Maltby says that those who are very ethical, like the nun, have broader definitions of dishonesty and are more likely to fail because of "world-perception questions."
"The obvious thinking is that if you're a crook you want to rationalize the behavior by saying everyone does it," he says. "But if you have high ethical standards, you will fail the test. You'll see a lot of dishonesty in the world."