Chally home page
Home Contact Us Site Map Help and Support Login

The New Art Science of Selecting Employees - Part 2

Employment Testing:
Facts versus Myths

Question 1Who uses testing?

Answer: Nearly every segment of business, industry, and government uses testing. Originally, testing was mandated in many government jobs as a deterrent to political patronage. In 1988 the American Society of Personnel Administrators reported in its monthly newspaper, Resource, that 84% of responding members included testing in their employment decision.6 The trend continued upward from there. Human Resource Executive Magazine surveyed 497 human resources executives in 1992 and reported, "Employers as a whole expect to conduct even more tests during the next 12 months."7

Question 2What kind of hiring improvements can I expect to get with testing?

Answer: When professionally done, testing has demonstrated the capability to reduce turnover and mistakes of hiring those who must later be fired, identify skill-based strengths and weaknesses for training purposes, and provide many other benefits. In fact, Frank Schmidt and John Hunter, industrial psychologists at the University of Iowa and Michigan State University, respectively, have completed studies indicating that hiring based on ability would have raised US industrial output by $150 billion.8

Question 3Why are organizations regaining an interest in testing?

Answer: Human Resource Executive Magazine reported that, "Executives cited a variety of reasons for increasing their use of pre-employment and assessment tests, including the introduction of the Americans with Disabilities Act, business expansion, and the need to select quality candidates."9

One reason for the recent resurgence in testing is that studies of the various selection tools have consistently found that tests are the most valid predictors of job performance. As shown in the chart below, one of the broadest studies reported the comparative validities of a variety of predictors for entry level jobs. This study demonstrated that testing is far more effective than traditional selection techniques.

Predictor Validity

Ability Composite (Test) .53
Job Tryout .44
Biographical Inventory .37
Reference Check .26
Experience .18
Interview .14
Training & Experience Ratings .13
Academic Achievement .11
Education .10
Interest .10
Age .01

* Validity refers to the ability of each predictor to correctly forecast subsequent success on the job. The higher the validity number, the better the predictor is at forecasting future success.

Question 4How can the use of testing help employers avoid discrimination charges and allow them to build a more effective work force?

Answer: Testing can help by providing proof that rates at which minority applicants pass are not significantly less than that of mainstream candidates ("no adverse impact") or by proving that the test is valid, i.e., related to job performance. Validation can statistically demonstrate that a test does predict a candidate's effectiveness on the job and, thus, the test is "consistent with business necessity."

Question 5Does the definition of "test" mean just a paper and pencil measurement?

Answer: No. The courts and the EEOC guidelines have defined "tests" to also include all those procedures that make up a personnel decision, including background checks, interviews, and supervisory performance appraisals. Of these, however, testing can be most easily and statistically demonstrated to be valid.

Question 6Can the traditional job interview be validated?

Answer: According to James Sharf, a former staff psychologist of The Federal Equal Employment Opportunity Commission: "It is no secret that interviews as typically practiced are the most frequently used of all personnel decision-making tools available. But stated in terms of the business objective of hiring the person who can do the job, use of the typically unstructured interview for decision-making purposes is far less likely to result in making decisions with demonstrable validity than would be more reliable objective sources of information."

Question 7 Is testing always useful?

Answer: No. There is no point in testing if you need to hire nearly every applicant or when you don't really know what factors in the job are most important. The H.R. Chally Group, for example, prefers to complete a statistical validation whenever possible to determine which factors are most important to do the job.

Question 8Can testing be misused?

Answer: Yes. Some people, especially those who did not believe in testing before they had the benefit of a good testing service, tend to be overly impressed with the results and often place too much emphasis on them in making the employment decision. They may tend to ignore other sources of information about the applicant which are also needed to determine if he or she can do the job. Testing is not an end in itself. Rather, it is an integral part of the entire personnel selection procedure that should be used in the successful prediction of job performance.


Find out why so many companies depend on
Chally's Online Pre-Employment assessment selection testing services
for sales, management, and career development positions.
Learn More about Chally Assessments



To Part 3:
- Three Criteria to Establish A Legal Selection Technique
- The Bottom Line
- Footnotes

 

FOOTNOTES:

6 American Society of Personnel Administrators, Resource, 1988.

7 "Testing Report," Human Resource Executive, May 1992.

8 Brimelow and Spencer, p. 82.

9 Human Resource Executive, p. 46.