It depends upon who you ask. Some think they are not a valid representation of the candidate. Others believe it gives a good snapshot of certain skills which can reflect the type of employee they will be. It’s possible that both answers are correct, it just depends upon which tests and how the tests are used.
Many companies use drug testing for pre-employment, but it’s not a clear indicator of a candidate’s job performance. If you do opt to use more extensive testing, an article in Forbes magazine offered the following suggestions:
Is the test relevant? Clearly you don’t want to give a Linux server technician test to an administrative assistant candidate. Likewise your Linux server tech candidates don’t expect to see a spelling and basic math test. Be considerate of each candidate’s talents and areas of expertise.
Are the results valid? If a position requires using a Mac version of a particular software on the job, but testing is administered on a Windows version, the results will be inaccurate. It can also present problems for any new hire and those in their department when it comes time to train. Be aware of requirements ahead of time.
Is the test reliable? This means if the same candidate takes the test multiple times, the results are the same. Do research on the test ahead of time. When using an outside firm for testing, check to see they use proven methods and resources for their services. Otherwise it may not be a good predictor of on the job results.
Does it meet local, state, and federal EEO laws? If not, just don’t use it. If there is a question, consult with a specialist in employment law.
Testing doesn’t have to be extensive. When used properly with your online interview system, the findings on a pre-employment test can help reduce employee turnover, increase productivity, improve morale and even reduce the cost of hiring.