Using Employment TestsÂ©
Suzanne Simpson Ph.D., C. Psych.
Well – developed and valid tests can contribute greatly to an employer’s ability to predict success on the job. On the other hand, tests that do not meet accepted professional standards can prove costly in terms of the effort of replacing workers who fail to perform, the stress created for all concerned, lost reputation and most importantly lost productivity on the job.
Employment tests are coming under increased scrutiny as potential barriers to employment for groups protected under Human Rights and Employment Equity legislation. Employers must be prepared to demonstrate that their employment practices do not discriminate against such groups. Employers, therefore, must be prepared to demonstrate that the tests they use are reliable, valid and non-discriminatory and based on generally accepted standards for testing.
There are hundreds of extremely well-developed tests, however, there are many that do not meet generally accepted standards for test development and validation. Many of these tests have wide distribution and high powered marketing networks, and to the untrained, the tests "look like" they could do the job. Often, more effort has gone into packaging these tests, than conducting the proper research to ensure that they meet at least minimum standards for validity and reliability. Unfortunately, while there are widely accepted standards for test development and application (e.g. Guidelines for Educational and Psychological Testing, Canadian Psychological Association, 1987), there is no regulatory body which reviews tests or serves as a consumer protection group to ensure that tests meet required standards.
Validity and Reliability
The bottom line is that it is a "buyer beware" situation. If your organization is thinking of using employment tests to screen applicants, or to assess employees to support employment decisions, consider carefully how the test was developed. The organization producing the test should be able to demonstrate that it is reliable, valid and non-discriminatory. Reliability, in layman’s terms, is the extent to which a test yields stable and consistent results. Validity refers to the extent to which the test actually measures what it was intended to measure. For example, for a test claiming to measure "sales ability", the test producer must demonstrate that there is a strong and defensible relationship between test scores and sales success. Finally, there should be evidence that the test is non-discriminatory – i.e. it is equally valid for all groups protected under Canadian human rights and employment equity legislation in comparison with majority groups.
All of this is not meant to scare employers from the use of employment tests. Rather to provide a warning that should tests not meet generally accepted standards, they could have the exact opposite effect to what was intended – the wrong person hired, or the wrong employment decision made. Testing is a highly complex field. If there is uncertainty as to whether a test or an assessment programme meets the required standards, seek the assistance of an Industrial/Organizational Psychologist. They are the only group who have the training and the qualifications necessary to evaluate whether a test meets appropriate standards and will do the job you want it to do. In addition, if the Psychologist is Registered by their province, they are bound by the standards set by regulatory bodies within their province to ensure that their clients use defensible testing practices.