Good Validation Matters in the Selection of Top Performers
(0) Comments
By Bill

In today's litigious business environment organizations want to know that the tests they are using in their selection process are legal.

Your answer is, yes, job-relevant, validated employment tests are a fast, fair, accurate, and legal means to better selection, placement, and promotion decisions. The better the match between a person's personality traits, skills, knowledge, abilities, and attitudes, and the requirements for the job, the more likely the person in that job will work productively and successfully. The proper use of employment tests is good for the employer and good for the employee or applicant. There are two legal criteria that employment tests must meet. Employment tests must be (1) properly validated, and (2) job-relevant. If they meet these criteria, then it is legal to use them.


What does it mean when we say a test is "job relevant?"


It means that here must be a demonstrable link between what the employment test measures and what the job requires. This means that the test publisher is responsible for making sure that the employment test measures the skill or attitude that it is designed to measure, through a process called validation, and the company that uses an employment test is responsible for making sure that the job description demonstrates the need for behavior or attitudes that the employment test measures.


Many test vendors say that they believe in construct validity. Still others will offer up that they prefer criterion validity as the best method. You will often hear a vendor say that good business sense suggests that concurrent validation on a group of people in your organization is the best approach. Their concurrent study is done by benchmarking as few as 3 to 10 employees in a job.

This approach is questionable at best!

It is Important to Know That Correlation is Not Causation

Suppose a well known magazine published a nice twenty-question hiring test. Furthermore, suppose your prospect or client gave that test to their high producers and averaged their scores. Is that validation?

Correlation means there is an association between two variables, for example, high producers tend to be good talkers. But correlation is not enough. Test developers and publishers need to find Causation. Does being a good talker cause high production (causation)? Or do high producers tend to be good talkers (correlation)? Your prospect or client should toss all of their tests in the closet unless they know, for CERTAIN, that the content they test for causes productivity.

Styles as measured by the MBTI, DISC, Enneagram, Social Styles, and so forth, might occur more often among certain job holders, but life is too complex to assume style causes productivity. This is critical to remember because, while hiring managers might embrace an intuitively attractive test today, if is does not predict performance, it WILL FAIL over time.

Organizations need to know that "what is measured" equals "on-the-job performance." Remember that blue eyes and blond hair might be correlated, but one does not cause the other.

Validation is the means of determining whether a test accurately predicts job performance, not simply identifying common personality traits.

Validity is not supposed to be some risky venture to determine if something or anything correlates with job performance. It is supposed to be validation.

The Guidelines to prove this are clear. They suggest using either Content (competencies - the nature of the job) or Criterion (performance on the job) validation, not benchmarking.

An acceptable process of validation is explained in the nearly 4,000 words in the Uniform Guidelines on Employee Selection Procedures. These guidelines include prerequisites for conducting a job analysis that include fairness, representative samples, personality traits, critical knowledge, skills and abilities, statistical sampling, statistical procedures, and other important criteria.

It is important to understand that even people with "the right" personality traits can fail in the job because they have the wrong skills.

Here's where it pays to use your common sense. Don't you think that if you could identify the "high performers," it is likely they would have different personalities?

If a test is really valid, wouldn't it be nice to know whether the candidate's test results were significantly different from the low group or that the candidate's results were statistically similar to the top performers in the validation study?

What Do "Validation" and "Reliability" Mean with Regard to Employment Tests?

 "Validity" refers to how well an employment test measures what it is supposed to measure. A validation study is a systematic gathering of data and information to support a claim that an employment test is valid, or (in other words) that it measures what it says it measures.

"Reliability" refers to the consistency in performance of an employment test: does it measure the same knowledge, skill, or ability every time it is used?

Employment tests should be carefully documented so that people and companies that want to use them can examine the way they were developed and validated, and for what purpose they are intended.

There are professional guidelines and standards for how an employment test is validated. The U.S. Equal Employment Opportunity Commission publishes Uniform Guidelines on Employee Selection Procedures that includes standards that a proper validation study should meet. This does not mean that EEOC "approves" a validation study or an employment test. The EEOC does not review, approve, or give stamp of approval to specific tests.

The American Psychological Association has published Standards for Educational and Psychological Testing that serve as guidelines for psychologists to use in the development and use of tests. These standards do not have the force of law, but psychologists who are involved in the development and use of employment tests should be thoroughly familiar with them.

Validation of the Assessment

For example: a vendor may say that in a national survey conducted on salespeople and earned commissions, that there is at best a weak correlation between profile scores and commissions earned. Scores obtained on a national level would be unsatisfactory predictors of commissions earned."

What assessment users should know is that it is important to know that sales commissions are consistently the number one indicator of sales performance, and if and when a vendor openly admits that their test scores are an unsatisfactory predictor on a national level and you may want to ask yourself why you would want to use this assessment.

Theory of the Test

Start with one simple question. Were the procedures used in validation consistent with generally accepted professional standards such as those described in the "Standards for Educational and Psychological Tests?" A reputable test publisher will generally make such a statement somewhere in its brochures or validity manuals.

Be cautious of any personality test that claims to have been written by a professional, and then immediately tries to lead you to the conclusion that it was professionally developed without referencing any validation or reliability studies. The two concepts do not necessarily go hand in hand.

What I would have to really question in relation to such a test would be whether or not the validity studies would meet the requirements of the "Uniform Guidelines on Employee Selection Procedures" as they pertain to the professional standards for validity studies.

Let's say that a company has designed a test that measures communication styles and that the personality assessment is very effective. The validation studies for any assessment instrument are only an objective measure that evidences that the test actually measures what it purports to measure, and in this particular case it is communication styles.

Let's say that this particular personality test is later given certain external modifications so that it can also be sold as a pre-employment assessment. The personality test is still backed by validity studies, but unless new validity studies are done, there are no validity studies to support the use of the assessment for its intended purpose as a pre-employment assessment.  In this example, the intended use is quite clear (to measure communication styles).

The Uniform Guidelines on Employee Selection Procedures specifically state that the evidence of validity and utility of the selection procedure must support its operational use.

Cutting Through the Quagmire

Suppose we accept that 1) Causation is the only way to accurately predict performance and, 2) our Test Content is a pre-cursor to job performance. Our next question is how to prove or validate the test and arrive at good cut-off points. This takes a thorough knowledge of statistics and experimental design.

For example, we have to define what to predict. We have to find "hard" data that is hard to fake, something we know we can trust.

Okay, let's suppose we have the right kind of hard data. What's next? We need to compare test scores with on-the-job performance. We can do this several ways:

1. By giving the test to everyone who applies, hiring them all, waiting until we get performance data and comparing test scores with job performance.

2. By giving the test to people already in the job and comparing test scores with job performance.

Of course we'll need to correct for "restriction of range," that is, the people who are IN the job will be more similar than people who APPLY for the job. So what? Well, for one thing, we might not see the same kind of big differences between high and low producers that we would see between applicants.

For example, we might find there is very little difference in skill between the top 10 in the NBA and the bottom 10. On the other hand, we would probably find a very large difference between the top 10 spectators and the bottom 10.

Flawed decision-making in hiring leads to misguided job standards, using tests that are not validated properly, hiring the wrong people, and rejecting the right ones. It is a major reason why Congress passed the Civil Rights Act and the Department of Labor wrote the 1978 Uniform Guidelines on Employee Selection Procedures.

Tests Add Consistency and Uniformity to the Hiring Process

Employment tests ask the same questions, in the same order, every time. They are scored and the results are presented in exactly the same way. This sameness lets the employer focus on the applicant's responses. Even the best interviewer can have a bad day and forget to ask all the questions of each candidate. Tests don't have bad days.

Tests Save Your Prospect or Client Valuable Time

Applicants can answer hundreds of test questions without taking up an interviewer's time.  Employment tests use the applicant's time, not the company's time. When there are many candidates for the same position, tests help an employer narrow the number of people who will be considered to those who meet certain job requirements -- and do so in a time- and cost-effective way.

Tests Demonstrate Respect for the Applicant or Employee

Employment tests give each applicant an opportunity to demonstrate their job-relevant skills or attitudes in a fair, unbiased way. Giving all applicants and employees this same opportunity is a demonstration of fair-mindedness and respect on the company's part.

Tests Help Your Prospects and Clients Avoid Bias in Hiring Procedures

There are two kinds of bias that should be considered. The law forbids discrimination, sometimes referred to as bias, in hiring. According to the U.S. employment law as promulgated by the Equal Employment Opportunity Commission, all aspects of the hiring process, such as interviews, application forms, reference checks, background investigations, physical examinations, drug testing, job-relevant tests, etc., are to be fair and not biased against any individual or group.

A second kind of bias arises from the natural human tendency to allow impressions, pre-conceived notions, and variations in interpretation to affect decisions and actions. This is almost always an unconscious, unintentional bias on an interviewer's part in the hiring process.

The best way to guard against the second kind of bias exerting an unintended influence during the hiring process is to rely on procedures and tools that are objective and consistent. Properly validated employment tests that are job-relevant help companies avoid the influence of the second kind of bias in the hiring process. There is usually no such evidence that can be shown to support the idea that the interview, reference checks, or any other part of the hiring process is equally fair and unbiased.


No Comment Available.

Add A Comment