This Is a Test
The ways of screening job candidates today run the gamut from background investigations to gene tests. And unless they're used with care, they can land you in court
You mean you don't know the genetic makeup of your work force?
Believe it or not, some big companies are using gene testing to screen out job applicants vulnerable to expensive and debilitating diseases. Although the practice is unusual, extreme, and legally dubious, it indicates the lengths to which some employers will go before hiring a new employee, and the range of screening methods available. It's now possible to know almost everything about job applicants, and consultants peddling background checks, drug tests, psychological profiles, medical exams, and more insist you should know everything.
None of those tests can take the place of a good interview. But some, used carefully, can supplement an interview and give you more complete information. In some instances, you may have a strong legal justification -- even a duty -- to use a particular type of screening method. In others, to do so could set you up for a lawsuit. Before you buy a test or hire a firm to conduct background checks, examine the common screening options and the questions -- legal, practical, and ethical -- that surround them.* * *
Everyone understands the importance of checking identity, education, employment history, and references against information the job applicant provides. But some companies, pressured for time, drop their guard. Witness Hatteras Hammocks.
In October 1991 the $10-million Greenville, N.C., company abruptly lost its controller. Executive vice-president Jay Branch interviewed several candidates for the position, but one man distinguished himself from the others. Recommended by someone Branch knew, the candidate came with a great rÉsumÉ -- B.A. in accounting, M.B.A. from Indiana University, CPA, and several years as controller at a big local corporation. The following March, Branch hired him after a quick call to the previous employer. After a few months it became obvious that the new controller couldn't do the job, and by October Branch had fired him.
His replacement quickly saw something wrong with the company's books. The man with the fancy rÉsumÉ had embezzled $60,000 from Hatteras Hammocks. A quick background check turned up no record of either of his degrees or his CPA credentials. The police arrested him and he confessed. Now his former employer, the large corporation, is scrutinizing its books. Branch, who saw his own name forged on several checks, says, "The whole episode taught me the necessity of checking thoroughly, no matter how good a recommendation is."
The threat of lawsuits, however, has changed the nature of the response to reference checks. Take the manager who, uncomfortable confronting an employee with a job poorly done, indicates satisfaction with the employee's work in reviews and in raises but later gives a negative reference to prospective employers. A court may consider those reviews and pay hikes proof that the manager was happy with the employee's work, and conclude that the bad reference was false and damaging. To protect against defamation suits, many companies stonewall any reference seekers. Some prospective employers try to wheedle information by promising to keep it "off the record." That may work, but only with the foolhardy or misinformed. "Off the record" won't protect a respondent against a defamation claim.
Voca Corp., a $70-million company based in Columbus, Ohio, with 2,500 employees in six states, has wide experience in screening job applicants. Hilary Franklin, director of human resources, says Voca asks references to suggest further references and asks those for still others. "As you get two or three times removed," Franklin explains, "you get more detailed, honest information."* * *
What if a repairman you hired were to burglarize a customer's home? If you hadn't checked his background to be sure he had no criminal record, you could be guilty of negligent hiring. If you hire a receptionist, however, you probably don't need to run such a check -- and probably shouldn't. You should always measure what's appropriate to know for the job instead of dwelling on the abstract.
Many of Voca's employees, for example, work closely with vulnerable clients -- the company provides care to people with mental retardation and developmental disabilities. To ensure the safety of its residents, the company wants to know about any incident of abuse in an applicant's history. Because Voca's care givers drive clients to appointments, they must have a clean driving record as well. The company informs all applicants that any job offer it makes is contingent upon a satisfactory background check.
Before Voca staff members run the check, they ask applicants if there is anything they would like the company to know about and give them an opportunity to explain. Then they fingerprint the applicants. Voca does not use those fingerprints in the background check. Rather, "it's our way of announcing the check, of demonstrating we're serious," Franklin says. "And it's pretty darn effective."
By running the check only after making a job offer, Voca protects itself from charges of discrimination. The company has set unambiguous standards for the information background checks turn up, and it sticks to those standards. For instance, no one convicted of a felony and no one with a history of abuse, neglect, or mistreatment will be hired to work with clients. Finally, Voca keeps specific findings confidential to protect against defamation. If, at your company, one person handles both the background checks and the hiring, that individual must be careful to consider only job-related information when hiring.
To facilitate background checks, you should get complete information from applicants -- full name, social-security number, driver's license number, address, employment history with no unexplained gaps -- and ask them to sign a release giving you permission to confirm it all.
If you're too shorthanded to do extensive background checks internally, you can hire outside investigation firms, which can pry into every imaginable area of an applicant's past. Companies usually order extensive checks only for sensitive, high-level positions -- top managers or people who will handle large amounts of money. More common checks include those of the applicant's workers' compensation, credit, and criminal records. But even those checks raise legal questions.
Some companies want to know an applicant's history of workers' comp claims in order to detect malingerers. And although workers' comp records are not public, there may be legal ways for a private investigator to obtain that information. However, under the Americans with Disabilities Act (ADA), you may not do a workers' comp check before offering someone a job.
As for credit and criminal checks, the government protects certain classes of citizens -- racial minorities and women, for instance -- against discrimination. Members of some protected classes are more likely to have bad credit or to have been convicted of a crime. If you consistently base hiring decisions on those criteria, you open yourself up to discrimination lawsuits. You would have to demonstrate that the information was related to the job, and in many cases that would be hard to prove. If you're hiring a driver, a drunken-driving arrest is relevant; if you're hiring a security guard, criminal history is relevant; if you're hiring a telemarketer, neither is.
Look into state laws, too, before conducting such checks. Many states don't allow you to check records of arrests, but most -- not all -- allow you to check convictions. Some states require you to get the applicant's approval before running checks, and some protect certain information. Speak to a labor lawyer before you do anything. And if you hire an outside firm to do your digging, demand a degree of legal savvy from it. Many display a cavalier attitude that could land you in court.* * *
Preplacement tests are administered after a company makes a job offer contingent upon a clean result. That may be the safest time, legally, to test for drugs, but some employers wonder if it's the least effective, since many applicants now expect a test.
Voca, for one, prefers to give preplacement drug tests. Its care givers must respond quickly in an emergency, so the company has good reason for ensuring they're clean. (After seven years of testing, it consistently finds that 4% of applicants test positive.)
Voca explains up front to all applicants that any job offer is conditional upon a clean drug test, and applicants sign a form indicating they've read and understood the policy. The company offers any applicant who tests positive an opportunity to retest at the company's expense. And Voca is consistent in its testing: it tests all candidates for care-giver positions. For accurate readings and to further protect itself, the company sends its tests to a certified laboratory. While that is more pricey, certification will help you if anyone challenges a test result in court.* * *
The ADA, which applies to companies with 25 or more employees (and beginning in July 1994, those with 15 or more), prohibits any pre-job-offer questions about medical conditions. You may ask only whether the applicant can perform the functions of the job. The Equal Employment Opportunity Commission has permitted certain agility tests, such as those used by police and fire departments, but the line between those and medical exams remains unclear. Would it be legal to ask applicants to read an eye chart? To perform a strength test? We won't know for certain until the ADA shakes out.
After you make an offer, you can, under federal law, require a complete head-to-toe physical and access to all medical records. However, you cannot legally use any information that is not job related, says Mark Rothstein, director of the Health Law and Policy Institute at the University of Houston. Eleven states explicitly limit exams to job-related information.
If you do test for job-related fitness, don't set blanket policies -- for instance, that anyone with a back problem is unfit. Rather, check that each person can lift the weight the job requires. Try to make that distinction, too, when you talk about the applicant. Inform managers only of ability, not of condition. "You can say, Joe Smith has a 25-pound lifting limit," Rothstein explains, "not, Joe Smith has a 25-pound lifting limit because he's got a slipped disk." Otherwise, overcautious managers might restrict Joe from jobs he's capable of performing and thus leave themselves open to a lawsuit. For the same reason, keep medical records confidential and separate from ordinary personnel files.* * *
Employers have used psychological tests since the 1950s. The oldest of those tests, developed for clinics and adapted for employers, screen for emotional disorders. Some tests contain prying questions about religious beliefs and sexual habits. In California (a state with unusually broad privacy rights), a job applicant sued a company after being given such a test. (He won, but the case is under appeal.) Such tests for emotional disorders should be used only for security positions -- so you don't give a gun to someone dangerous, for example.
The psychological-testing industry blossomed after 1988, when the federal government banned the use of lie-detector tests in most employment situations. Publishers filled that void with integrity tests. The simplest "core" integrity tests look at security issues only -- theft, drug abuse, and violence -- and cost as little as $8 to $16 a test, depending on volume. Others may test for security plus productivity or customer-service attitudes. Some even claim to measure the likelihood that a new hire will have accidents on the job or will quit.
Those tests try to predict job applicants' propensity to steal, for instance, by matching their test results with those of known thieves, asking obvious questions, such as, How often do you tell the truth? and less obvious ones, such as, How often do you make your bed?
Do those tests work? In a way. Just because one person's responses match those given by thieves does not mean that person will steal; rather, it means he or she is more likely to steal. You play the odds and, in the long run, may improve your chances of reducing theft.
Some integrity-test publishers scare employers with estimates of thefts by as many as 30% of all employees. But they may be counting people who take home pens or make personal calls from the office. Other experts say the rate of serious theft is as low as 5%. Whatever the situation in your company, be warned: even the best integrity test will produce many false positives -- people unjustly suspected and rejected.
Beyond integrity tests, publishers have developed all sorts of "personality-assessment tools" to predict how an applicant will do the job and fit into your organization. Those tests may judge a prospective salesperson's aggressiveness, for instance, or an accountant's attention to detail.
Few tests work in all situations. If you are considering using psychological tests, how should you choose? Get references from others in your industry, a state psychologists' association, or a local university's professor of industrial psychology. Then contact publishers and ask to see their validation studies, which are like trial runs. The trial group should resemble your work force in job duties and in demographics.
Also ask the publisher if the test has ever been challenged in court and how it fared, and ask what the publisher will do to help if someone challenges your use of the test. Conscientious publishers will provide you with technical assistance and will send a staff psychologist to testify, if necessary, at no charge.
Remember, tests are often wrong about an individual. For effective hiring, make testing only one part of an overall assessment that includes interviews and reference checks.* * *
Publishers of personality tests talk persuasively about how much more accurate their tests are than interviews. And a good test probably does work better -- more accurately for you, more fairly for the employee -- than a bad interview. But a good interviewer can get richer information. It takes time to develop interviewing skills, though, and many managers prefer to invest their money in a quick solution.
Managers also like the reassuring formula a personality test provides. A bad interview seems like purposeless small talk, but a test imposes structure on the process. Ed Ryan has an answer for that. His consulting firm, MPR, based in Chicago, teaches clients to set benchmarks for each job in the same way that some personality-assessment tests do. If you're hiring a receptionist, he says, ask yourself who's the best receptionist you've ever come across. What made that receptionist so good? Categorize those qualities into certain overall behavioral traits -- sense of responsibility, attentive to detail, good at relating to people.
Ryan then teaches clients to conduct "pattern" interviews to identify those traits in applicants, using such questions as, How do you feel when someone questions your statements? and, How important is it what other people think of you? Those are similar to the questions personality tests ask, but Ryan believes they work better in interviews. Testing forces the subject to choose between true and false or among multiple choices. If you ask the same questions in an interview, you can get subjective responses that, Ryan claims, tell you much more.
The Registry, based in Newton, Mass., which provides software engineers and other personnel to large corporations, uses the sort of pattern interviewing Ryan describes. The Registry combines its interviews with other techniques designed to test qualities it looks for in salespeople and recruiters.
An applicant for a job at the Registry meets with the hiring manager, who describes the position and reviews the applicant's background. If the applicant is promising, the manager will call in a recruiter or salesperson to describe the job firsthand. Next, the candidate interviews with the team he or she hopes to join, to determine personal fit. Each interviewer makes notes on an interview guide, which spells out questions to ask -- for instance, "Why are you successful?" -- and the type of responses to listen for -- ones that reveal "a process or methodology; logical, repeatable steps."
Job candidates also interview over the phone with the directors of training and development in the Registry's Washington, D.C., office, and with one or two training managers in other branches. Those phone interviews test applicants' persistence, a trait particularly valued by the company. The managers are hard to reach and often won't return calls. "We'll tell candidates, 'Now interview with Meredith Cohen in our corporate office,' " CEO Drew Conway says. "It may take five calls to reach Meredith. Do they just leave a message? We want them to find her and get her on the phone. The ability to identify and track down the right person is a part of our day-to-day job."
Before making a final decision, the Registry administers a personality test, which was designed by an industry insider, so Conway believes it accurately identifies the qualities that make success more likely. He doesn't believe the profile will tell him which candidates will be successful -- indeed, one of his best salespeople performed badly on the test. Rather, he thinks it indicates how naturally the qualities needed for sales and recruitment come to a candidate.
As valuable as that information is, it can't make Conway's decision for him. "Lots of companies use the profiles as an excuse not to go through the interview process. But to do good work, you have to invest the time and energy."
YOU ARE WHAT YOU EAT
"Whenever Thomas Edison was about to hire a new employee, he would invite the applicant over for a bowl of soup. If the person salted his soup before tasting it, Edison would not offer him the job. He did not hire people who had too many assumptions built into their everyday life. Edison wanted people who consistently challenged assumptions."
-- From Thinkertoys, by Michael Michalko, Ten Speed Press, 1991
THE PAPER-BAG TEST
Amy's Ice Creams, in Austin, wants counter help with imagination. Scoopers joke with customers and perform for them. Amy's screens for that elusive aptitude by handing applicants a plain paper bag and asking them to improvise. One applicant wrote his qualifications on the bag, then made it the basket of a model hot-air balloon and floated it in the doorway. The bag test started one day in 1985 when the store ran out of standard applications and asked an applicant to jot down his personal information on a bag. He elaborated, and Amy's never has gone back to the old form.
PSYCHOLOGICAL CHEAT NOTES
Publishers insist that tests are hard to fool, but well-read applicants may know a few tricks. The Organization Man, by William Hollingsworth Whyte (Simon & Schuster, 1956), a classic of organizational ethics, includes an appendix titled "How to Cheat on Personality Tests."
PRINT THIS ARTICLE