Cut the Confusion: Pre-Employment Test Validity and Reliability
Author: Scott Fiore
By Ira Wolfe, Guest Blogger for TriStarr Staffing
Two of the most important factors that determine if a pre-employment test is “legal” to use are validity and reliability. And there is no shortage of confusion about what they mean and do. So let’s start there.
Validity refers to the accuracy of a measure. A measure is valid if it actually measures what it says it measures. For instance a thermometer is valid for measuring temperature, not wind velocity, even if someone claims otherwise. In employee testing, a pre-employment attitude test measures a candidate’s attitude toward work. It does not measure motivation or personality or intelligence. If you want to test for personality job fit, then you must use a test validated for personality job fit.
Here’s the rub. Valid does not infer legal and legal does not infer validity. For instance, “do you plan to have children” may be a perfectly valid question, depending upon the requirements of the job. If hired, the employee may add to health care costs or be unable to perform certain duties if she becomes pregnant. There is a high likelihood and therefore correlation that both situations will be true. Therefore the question may be valid. Unfortunately, the question cannot be asked – it’s on the list of forbidden fruit for employers.
Likewise, it might be legal to ask an English speaking applicant to describe his ability to speak Spanish, but if it has nothing to do with the job, it’s not a valid question to ask. Interview questions must be job relevant. If the candidate doesn’t have to be multi-lingual, the questions are not job relevant. In other words, to be legal, the questions must be valid. But valid questions aren’t necessarily job relevant.
The second critical factor is called reliability. Reliability refers to how dependable or consistent a test measures a characteristic. If a candidate takes the test a month ago, and repeats it today, will he or she get a similar test score? A test that yields similar scores is said to measure the characteristic reliably. Admittedly, the test taker’s temporary psychological or physical state as well as environmental factors can influence the results. These and other similar factors are sources of chance or random measurement error in the assessment process. The degree to which test scores are unaffected by measurement errors is an indication of the reliability of the test. Reliable assessment tools produce dependable, repeatable results.
Test-retest reliability is one of two important types of reliability. Test-retest indicates the repeatability of test scores with the passage of time, or the stability of the characteristic being measured. Some constructs are more stable than others. For example, an individual’s reading ability is more stable over a particular period of time than that individual’s anxiety level. Therefore, one would expect a higher test-retest reliability coefficient on a reading test than you would on a test that measures anxiety.
Validity and reliability are terms used interchangeably and in the wrong context by many managers and human resource professionals. I hope this article offered relief from the confusion and helps guide better decisions when selecting pre-employment tests.
[Note: Success Performance Solutions & TriStarr Staffing use pre-employment tests that meet EEOC guidelines for validity and reliability.]
By Ira Wolfe, Guest Blogger for TriStarr Staffing
Two of the most important factors that determine if a pre-employment test is “legal” to use are validity and reliability. And there is no shortage of confusion about what they mean and do. So let’s start there.
Validity refers to the accuracy of a measure. A measure is valid if it actually measures what it says it measures. For instance a thermometer is valid for measuring temperature, not wind velocity, even if someone claims otherwise. In employee testing, a pre-employment attitude test measures a candidate’s attitude toward work. It does not measure motivation or personality or intelligence. If you want to test for personality job fit, then you must use a test validated for personality job fit.
Here’s the rub. Valid does not infer legal and legal does not infer validity. For instance, “do you plan to have children” may be a perfectly valid question, depending upon the requirements of the job. If hired, the employee may add to health care costs or be unable to perform certain duties if she becomes pregnant. There is a high likelihood and therefore correlation that both situations will be true. Therefore the question may be valid. Unfortunately, the question cannot be asked – it’s on the list of forbidden fruit for employers.
Likewise, it might be legal to ask an English speaking applicant to describe his ability to speak Spanish, but if it has nothing to do with the job, it’s not a valid question to ask. Interview questions must be job relevant. If the candidate doesn’t have to be multi-lingual, the questions are not job relevant. In other words, to be legal, the questions must be valid. But valid questions aren’t necessarily job relevant.
The second critical factor is called reliability. Reliability refers to how dependable or consistent a test measures a characteristic. If a candidate takes the test a month ago, and repeats it today, will he or she get a similar test score? A test that yields similar scores is said to measure the characteristic reliably. Admittedly, the test taker’s temporary psychological or physical state as well as environmental factors can influence the results. These and other similar factors are sources of chance or random measurement error in the assessment process. The degree to which test scores are unaffected by measurement errors is an indication of the reliability of the test. Reliable assessment tools produce dependable, repeatable results.
Test-retest reliability is one of two important types of reliability. Test-retest indicates the repeatability of test scores with the passage of time, or the stability of the characteristic being measured. Some constructs are more stable than others. For example, an individual’s reading ability is more stable over a particular period of time than that individual’s anxiety level. Therefore, one would expect a higher test-retest reliability coefficient on a reading test than you would on a test that measures anxiety.
Validity and reliability are terms used interchangeably and in the wrong context by many managers and human resource professionals. I hope this article offered relief from the confusion and helps guide better decisions when selecting pre-employment tests.
[Note: Success Performance Solutions & TriStarr Staffing use pre-employment tests that meet EEOC guidelines for validity and reliability.]
Share This Post