Archive
By Charlene Solomon
Apr. 1, 1996
It isn’t easy to choose the right person for a job. We all know of high achievers who don’t perform well on tests. We’ve had experiences hiring brilliant people who have had dazzling recommendations but who show up to work late or perform poorly on the job. Enter: employment tests (commonly used tools for hiring, as well as training and placement decisions).
When the hiring of a diverse workforce is of paramount importance, however, the appropriateness of pre-employment testing for a multicultural workforce takes on a different cast. The major question, of course, is: Are these tests impartial?
Human resources executives are concerned that written and verbal aptitude, personality and achievement tests may not give them the accurate information they need when applied to applicants from diverse backgrounds. For example, can employers assume that a foreign-born bilingual applicant would select the same right answers as a person would who was born in the United States and whose native language is English? What about cultural factors that may persist, even when language isn’t a problem?
Some companies use no paper-and-pencil tests. They rely strictly on interviews and recommendations. At the other end of the continuum are businesses that buy packs of tests, administer them, score them and key the results into a computer. The tests and assessment services run the gamut from inexpensive skill examinations that can be scored and interpreted in minutes to in-depth personality assessments that require extensive time and judgment on the part of both the test taker and examiner.
Regardless of the scope of pre-employment tests, some are better predictors of success than others; some are more fairer to all groups than others are. Most experts believe that tests should be only one component of the hiring process.
“In our opinion, it’s best to take a battery approach. Look at all aspects of the person—as much as is feasible—and make decisions with everything in mind,” says Scott G. Howard, president of Reid, Merrill, Brunson & Associates in Denver. The company provides tests for pre-employment, promotion and internal career-development purposes. The assessments include:
“We try to compare [potential employees] with people who are or have been successful in the position to be filled. We look at strengths and weaknesses, and make recommendations for improvement,” says Howard.
Strengths and weaknesses, similarities and differences. These are key factors in a person’s employment desirability. In this time of diversity awareness, many people find that one of the advantages of conducting assessment is that tests point out differences between people.
“We’re in an era of international competitiveness. We’re looking for a variety of people to run our businesses, and to sell and make our products,” says Charles F. Wonderlic, Jr., vice president of Wonderlic Personnel Testing in Northfield, Illinois. “It means identifying the superb, not just the norm or average.” This involves asking such questions as:
“These are questions that tests can begin to answer,” says Wonderlic.
Although generalizations are risky, most experts agree that testing is far more sophisticated today—and more valid, in most respects—than ever. Still, human resources professionals who must use tests and nevertheless are interested in hiring a diverse workforce need to know how to employ testing to their best advantage.
Job-specific tests are best.
“Some employment tests lead to significantly different passing rates for minorities when compared with majorities,” says John W. Jones, vice president of research and service for McGraw-Hill/London House, a leading assessment publisher located in Rosemont, Illinois. “Companies seeking a culturally diverse workforce prefer tests that don’t discriminate unfairly.” In addition, some employment tests clearly are job-specific, and others are more general, according to Jones. General tests originally were developed for clinical or educational settings, rather than for occupational settings. “Research shows that tests developed specifically for the workplace tend to be more accurate predictors of employees’ on-the-job performance than the more general psychological tests are,” says Jones.
Therefore, job-relevant tests that have no between-group differences are the ones to use. According to Jones, an example of this type of test is a multidimensional selection test that measures job applicants’ attitudes toward a wide variety of workplace behaviors, including integrity, dependability, service, safety and productivity. Jones says that job attitude tests usually are fair to all protected subgroups. “They’re ideal for the age of cultural diversity in the workplace,” he says.
Jones says that general psychological tests that do yield differential scores for certain subgroups might include a clinical personality test that was designed to assess personal values, family adjustment or emotional health. There also are some intelligence tests that have been scientifically validated in the workplace, but which have an adverse impact on protected groups, and therefore are controversial. In addition, there are some tests that aren’t accurate predictors of employees’ behavior, although they don’t impact any group adversely.
“I believe in valuing diversity and using the very differences that we were supposed to avoid,” says Lewis Griggs, president and executive producer of Griggs Productions, a San Francisco-based company that produces diversity training films. “I feel that any test of anything is, of course, biased by its maker—it can’t be otherwise. Still, any test is legitimate for what it tests. The individual using the results merely has to have the perspective to use that information responsibly and understand what he or she is testing for.”
For Griggs, this means giving a woman the opportunity to prove that she can lift 100-pound hay bales or giving a man the chance to be a good cross-cultural Avon salesperson in predominantly minority areas of the city. “But, isn’t it a legitimate question to ask, ‘Are you strong enough to lift the hay bales? Are you bicultural enough to be the appropriate salesperson for Avon or to sell insurance to Vietnamese immigrants?'” asks Griggs. “If we don’t ask those obvious questions, we’re putting our heads in the sand. We’re denying the real differences. We’re also denying the opportunity to discover differences that can be competitive advantages, all else being equal.”
Identify skills needed for specific jobs.
First, analyze the job. Find out which activities the person must accomplish. What does he or she have to do adequately or superbly? Focus on the things that the person has to do that require a high level of competence. Look at the necessary skills to achieve success on the job. In general, what qualities must that individual have if he or she is going to accomplish that success?
Then back off from there and remember that you aren’t just assessing a person, you’re assessing a specific person for a specific job. For instance, the job may require extensive contact with people. The employee has to perform customer-service-related activities and make customers happy. Find out what types of measures would reveal whether an employee would be good at serving customers and whether he or she would be happy doing that type of work. In this case, you would need a personality test.
In other cases, testing is obvious and is used simply to weed people out. For example, if one skill required is the ability to type a minimum of 45 words per minute and the candidate can only type 20, it’s clear that the person isn’t adequate for the job. “At that point, regardless of your status or whatever the unique attributes are that the individual brings to the organization,” says Wonderlic, “if he or she doesn’t have the minimum skills required—and what we’re really looking for is someone who has the maximum, who types 85 words per minute, who can also write, who has an understanding of the people in the community—then the test will help the human resources professional make the rejection.”
Once the job needs have been identified, the next step is to look for a company that can be an information resource—a test vendor or publisher that will help determine your company’s assessment needs and help put them into action.
Howard suggests you look at the testing company’s research. What types of studies has it done? What groups did it use in these studies? What control groups did it use? What was the level of detail?
“Ask the company to give you a description of the groups used in its studies, including the majority groups. Were they all males who have had two jobs each and who have been on the job for 15 years, or is there a good representation of the entire population? Ask which protected groups were studied,” says Howard. The answers should come back to you in easy-to-understand language that a businessperson can use, not psychobabble, he says.
Howard suggests employers that routinely hire people who either don’t have a lot of work experience or whose first language isn’t English (in manufacturing, for example) ask the test publisher if it has analyzed that specific population. See if there’s adverse impact built into the test results.
Another crucial element is to be sure that test publishers track EEO and ADA data. If the company is good, it will up-date statistics, continuously track adverse impact and update norms. (See “A Glossary of Testing Terms,”)
“A good consulting provider can help people decide what they need,” says Howard. Engage in a relationship with the provider. Some people call Reid, Merrill, Brunson & Associates (Howard’s testing firm) saying they need one thing when really they need something else, which becomes evident after they’ve described it.
You also should ask about alternative forms of specific tests. “A good test publisher will have many alternative formats for each test,” says Wonderlic. For example, Wonderlic’s tests are translated into foreign languages. One test is available in 16 languages. Wonderlic says the company could make use of the test translated into even more languages.
Language isn’t the only area in which differences can affect test taking. There are other subtle, but very important related differences among people. For example, people who come from Puerto Rico use decimal points as Americans do, and those who come from Mexico use commas, although both groups are Hispanic and speak Spanish A test should allow for these differences. “We want to make it as easy as possible for them to understand the questions,” says Wonderlic.
McGraw-Hill/London House has formed an alliance with Berlitz and other companies that do translations. Jones says this is crucial because multilingual versions of tests are becoming important. Test designers are becoming more culturally sensitive as a result. “We aren’t assuming that everyone should be forced into the English version,” says Jones.
There’s more to assessment than meets the eye, however. For instance, it isn’t simply the translation. Let’s say the human resources manager identifies a Vietnamese woman who doesn’t speak English but has an I.Q. of 140 and is perfectly capable of handling any management learning objective. She could learn at a high level and solve problems on the level of a manager. “You can assess her English speaking ability or ability to learn English,” says Wonderlic “but you also have to assess her potential for integration into the workplace. You have to do the follow-up. How do you incorporate a person who scores well on a version of the test translated into her native tongue, but doesn’t speak the language of the workplace?” He continues by saying, “We want to maintain and promote diversity. We also want to maintain and promote productivity. We want to find a way to do both.”
The test format is another important criterion. This can mean large print, Braille or audio. This is especially important because of ADA requirements.
Finally, be sure there’s scientific norming. This is a process that assures tests are equivalent across cultures. For example, all test questions should have the same meaning regardless of a person’s background.
Although there’s a lot to consider when deciding whether to use pre-employment testing when hiring members of a diverse workforce, many human resources executives deal with such considerations all the time. And these professionals don’t always have the same opinion about pre-employment testing.
Kraft General Foods considers that testing is an investment.
Charles Reid, director of diversity management at Kraft General Foods Inc. in Northfield, Illinois, always has been a proponent for pre-employment testing. The reason? “I think testing can provide information that’s valuable to an individual or the tester but couldn’t be obtained under other circumstances,” he says.
When Reid came into the business world from the education arena in 1972, he says that tests weren’t being used correctly by business and industry. The courts ruled that some tests were discriminatory. When companies stopped using many of these instruments, interviewers were thrown into a quandary. They didn’t know how to check for the sets of skills individuals had, what they were permitted to ask prospective employees, or what information they could look for to help them make valid decisions.
“A properly constructed test looks at the elements of the job and then measures the finite characteristics of the candidate that guarantee relative success. Those are the things that ought to be looked at in testing,” says Reid. “For the past 20 years, when people have asked me about tests, I’ve encouraged them not to throw them out.”
At Kraft General Foods, the company looks at testing from an entry-level perspective. It also uses tests to answer work-related questions, such as mechanical skills. The organization’s attitude is that it’s better to test in the beginning for skills that will be used on the job, than to invest the time and energy in hiring people and then discover six weeks later that they can’t do the job.
Reid describes himself as an African-American who has worked in industry for about 20 years and is trying to take an honest look at who the entrants are to the workforce and what skills they’re bringing. He supports the use of testing because he’s finding that employees are lacking skills. “When you look at blue-collar workers, nonexempt clerical workers or professional managerial workers, you’ll see that each situation is a little different. Because roughly 80% of the workforce in most companies [in a manufacturing environment] is blue collar, however, this group has a greater impact. I think that the skill level of the blue-collar workers coming in today is less than it has been in the past,” says Reid.
He says you can’t assume that just because a person has completed a four-year secondary program, he or she has adequate skills. Reid, who has also served on the National Board for the Literacy Volunteers of America, points out that at least one adult American in five is functionally illiterate. “If you don’t use paper and pencil testing, you may miss the fact that the person is unable to read. I think a paper and pencil test gives you at least an indication of whether you have a problem.”
Reid quickly adds that he isn’t intimating that just because a person can’t read doesn’t mean he or she isn’t skilled, because many illiterate people today are functioning adequately in their jobs. It’s important, however, to have that information. A verbal interview generally wouldn’t discover this fact, Reid says, because many illiterate people have outstanding coping skills.
“When you think about what’s happening in the educational system, people, especially people of color, are being undereducated. Therefore, we lack the skills that we need for a high-quality workforce,” says Reid. He goes on to explain that whenever there’s competition for the highest-quality people, it means that companies are competing for the same people. When the applicant pool includes individuals who don’t have adequate skills or who will need retraining if the company does employ them, testing is a way of establishing the level at which each person comes into the company. “It allows you to set up training programs to modify or develop skills that the individual brings to you, and to help him or her become a productive employee,” says Reid.
Cahners Publishing Company tests only for some positions.
As a general rule, Newton, Massachusetts-based Cahners Publishing Company doesn’t do pre-employment testing. The company does conduct some testing for sales positions and for certain management positions. “When we do test,” says William Stevens, Cahners’ director of human resources “it’s only to provide an additional tool to use in the employment process—in addition to such information as experience, things they may have published (for writers) and territories they covered for other companies (for sales).”
Cahners uses tests only with finalists in the selection process. If the company can narrow the field down only to 10 people, for example, it might use testing to narrow the candidate list further.
On the sales side, Cahners uses testing to find out if the individual is self-directed, motivated or needs guidance. And if guidance is indicated, the testing determines what kind is needed. Once the person is employed, the company uses the test results as a tool for the manager to help the person develop professionally on the job.
Cahners continues to use the tests from time to time to focus on training needs and to help managers understand and pay attention to certain facets of the individual’s personality. It’s helpful to know, for example, if a worker doesn’t like to work while someone looks over his or her shoulder.
Stevens, however, emphasizes that testing is merely one component. “You can talk to a recruiter and the recruiter will say, ‘You can give me all of the test results you want, but there’s something inside me that gives me a good feeling or a bad feeling about somebody. It’s that gut feeling,'” says Stevens. “On tests, an individual can have a good day or a bad day, but in the interview process, if it’s done properly, there’s enough diversity that a person who doesn’t do well with one manager might do better with another. In a test, you don’t have that opportunity for variation.”
Stevens believes that tests won’t weed out particular cultures. In fact, he says that the problem for recruiters is having enough cultures to draw from in the beginning. “The problem is getting [diverse individuals] into the loop, and it’s more difficult in different parts of the country,” he says.
Cincinnati-based Procter & Gamble also believes that any testing procedure should be supplemented with other hiring tools. The company hires only at the entry level. “The use of testing at Procter & Gamble is of paramount importance,” says Lynwood Battle, manager of corporate affirmative action. “Selection is critical because we grow all our management from inside the company.”
Like the other companies mentioned, Procter & Gamble uses tests as part of a total assessment of job candidates. “It has a lot of key indicators that offer a profile of the successful candidate,” explains Battle, “and only a part of that process is a paper-and-pencil cognitive test.”
For the previous four years, according to Battle, the company has recruited minorities at a rate of 20%, which is significantly higher than the percentage of available minorities who are receiving college degrees. The process must be working because, according to Battle, women and minorities are moving up at a rate commensurate with their representation in the population.
He believes that skill tests generally are valid, although he says, “I can’t emphasize enough that the test (a 45-minute cognitive test) is only one aspect of the total assessment process.”
The assessment of a candidate will take several days, beginning at the college placement office on campus. It can proceed for a period of several weeks. The paper-and-pencil test is administered along with the other components of the company’s job-hiring procedures. Only after all of the information is compiled and evaluated is the prospective employee reviewed. “The determination is based upon a total assessment process,” says Battle.
It seems clear that within the proper framework, testing is a valuable tool for human resources professionals. Tests give reliable data about skills, and will point out deficits as well as strengths. When employers go outside the skill-testing arena, it becomes important to choose assessment professionals who can help meet their specific needs and focus on diversity.
Note: Kraft General Foods is now known as Kraft Foods Inc. Charles L. Reed is currently the director of ethnic external relations, Kraft Foods Inc.
Personnel Journal, April 1996, Vol. 75, No. 4, pp. 131-140.
Schedule, engage, and pay your staff in one system with Workforce.com.