<figcaption> Credit: © GETTY IMAGES/MATTHIAS TUNGER</figcaption>
Credit: © GETTY IMAGES/MATTHIAS TUNGER

With more than a hundred job openings at any one time, the 5,000-employee Applied Biosystems is constantly looking for ways to more efficiently sort the wheat from the chaff. This includes giving candidates specific open-ended questions, whose answers are standardized and assigned a score based on factors such as core skills, leadership, and cultural compatibility. As a check on the validity of the approach, director and higher-level candidates that make it through this round have also recently been asked to take a psychological test. Of the approximately 50 candidates of this grade that the Foster City, Calif., company has assessed since it started the practice, only a few failed the test, says Angela Peters, the company's director of staffing, confirming the move towards using more consistent and objective measurements in the company's screening process.

Applied Biosystems is not alone in striving to bring additional rigor...

"Sitting on a park bench and hiring every seventh person would be as effective as the traditional hiring process." -Jim Hazen

A closer look

For a smaller biotech with relatively few new hires per year, there is less utility in formal assessments and third-party interviews, says Colleen Hamilton, director of human resources at Biorexis, a company of 48 employees based in King of Prussia, Penn. Biorexis hires about five people each year and relies on traditional interviewing. Some companies have taken a step toward formal assessments by developing their own behavioral interviewing techniques.

At Kalypsys, a 115-employee pharmaceutical company based in San Diego, the interview team asks the candidates open-ended questions such as, "Describe a time that you led by example. What was the result?" The team looks for a three-part answer: a description of the situation, the action taken, and the result. "We look for examples of past behavior to indicate how an individual will respond or act in the future," says Audrey Schellinger, director of human resources at the company.

In contrast to assessments used for hiring, those used for team-building tend to be descriptive rather than evaluative. "In team-building, we are not too concerned about skill sets; we just look at the interaction of personality," says Hazen. One of the goals is to promote team function by helping employees identify and adjust to different work styles. "Over the course of 27 years I've worked in the pharmaceutical industry, I've had a dozen different tests administered to me," says a former Pfizer executive, now employed at a biotechnology start-up company. These included such venerable names as Myers-Briggs, Johari Window, and Situational Leadership. Such assessments can help retain good employees and help them move forward in the right direction, she says. "That's more cost-effective than starting with a new person from scratch."

From 1993 to 1994, when Merck reengineered its purchasing department into a strategic procurement organization, it used a behavior assessment tool to decide which employees would move to the new unit. "It was a stressful period, because we were going through a dramatic change," recalls Jim Segal, now an associate director at the unit. The tool, called System for Identifying Motivated Abilities (SIMA), indicated that Segal could think strategically and deal with rapid change - desirable traits for employees in the new unit. "The end result was, I made the transition," he says, while some members of the team were assessed less capable of strategic thinking and were reassigned to different jobs.

Costs for assessments vary widely. HPI ranges from $25 to $125 depending on the depth of detail required in the report. Kenexa charges from $5 for an online test to more than $2,000 for an in-depth phone interview followed by a consultation with an analyst. Third-party behavioral interviews could cost several thousand dollars per hire. Ben Dattner, a psychologist and founder of Dattner Consulting, a New York City-based assessment services provider, calculates the cost of an unwanted turnover to be equal to an entire year's compensation. By improving the "hit rate" for employment, assessments and structured behavioral interviews could have a payout between 10 and 100 to 1, he estimates. Other potential benefits are reduced hiring costs (unsuitable candidates can be screened out before the interview) and improved productivity.

Bridgeville, Penn.-based Development Dimensions International says its behavioral interviewing technique helped lower employee turnover from 21% to 17% at Covance, which was hiring 1,500 to 2,000 people a year, and reduced hiring costs by 20% at Sanofi-Aventis. Other major assessment companies report similar benefits from their products. "We saw a significant increase in sales activity and retention rates with our new hires," says Phil Stewart, director of human resources at York, Penn.-based dental products company Dentsply Caulk, which used Kenexa to develop a preemployment assessment. Dentsply Caulk was impressed with the results and implemented the assessment company-wide.

Testing the tests

The widespread use of assessments raises some concerns, however. The industry is currently unregulated, and some tests may not comply with stringent conditions for fairness and nondiscrimination laid out by the Americans with Disabilities Act, Equal Employment Opportunity Commission, Department of Labor, or Civil Rights Act. Further, to be effective, a test should be both reliable (proven to be repeatable) and valid (proven to measure what it is designed to measure). Many tests fail to meet these standards, says Hogan.

Some employers also use the wrong kind of test altogether, says Hazen. For instance, they might use the Myers-Briggs assessment in employment decisions. "This instrument was designed as a marriage counseling tool and never intended for use in hiring," he says. On online cognitive tests, cheating could also be an issue, since the test-taker's identity is hard to confirm. In personality tests, some candidates may try to beat the system by providing dishonest answers, although many assessment tools now include a "candor score" to detect this. Given these considerations, says Hazen, "assessments should not be more than a third of the hiring decision."

cshekhar@the-scientist.com

Interested in reading more?

Magaizne Cover

Become a Member of

Receive full access to digital editions of The Scientist, as well as TS Digest, feature stories, more than 35 years of archives, and much more!
Already a member?