The problems with employee integrity tests

Henry Ford used to send committees to his employees’ houses to check their behavior. Good employees would avoid excessive alcohol, keep their houses clean and do things “the American way.” These were his conditions for paying them a $5 wage.

Ford’s practices were extreme, but it’s not surprising that employers want to trust their people. For more than 60 years now, employers have used integrity testing to avoid hiring ‘high-risk’ candidates. These tests became more popular in the US after the Employee Polygraph Protection Act (EPPA) banned lie detector tests.

Employee integrity tests are meant to measure honesty, dependability and work ethic. They take two forms: overt and covert. Overt integrity tests refer directly to dishonest and counterproductive behaviors (theft, cyber-loafing, absenteeism etc.) Covert testing is personality based. They assess integrity by proxy (e.g. conscientiousness.)

Are integrity tests effective?

There’s a large body of research with interesting results. Overt employee integrity tests have been shown to be valid and somewhat better in predicting job performance than personality tests or unstructured interviews. Covert integrity intesting, on the other hand, can predict absenteeism better than overt tests. There’s also evidence that employee integrity testing is generally less biased and more cost-effective than other forms of assessment. And there’s positive feedback from employers who state that integrity tests have reduced worker’s compensation claims among new hires.

So, it seems employee integrity tests can add value to the hiring process. Does that mean employers should use them? As with all assessment methods there are a few more questions employers should ask before deciding to use an integrity test:

Are employee integrity tests legal?

There was a time when employee integrity testing asked about people’s religious beliefs and sexual orientation. Those tests were challenged in court. Problems could also arise from tests that ask candidates whether they were accused or convicted of a crime.

Tests could also be invasive in a subtle manner. For example, some personality-based tests ask candidates to rate statements like “I experience extreme mood swings.” These statements try to assess dependability. But, they can also be viewed as an indirect effort to diagnose bipolar disorder. Thus, the test is discriminatory under the American with Disabilities Act (ADA). The pre-employment use of the Minnesota Multiphasic Personality Inventory (MMPI) was challenged in court for this reason.

The law, generally, restricts the questions that integrity tests can ask. Massachusetts has banned integrity testing altogether.

What to do:

Always ask test providers whether their test complies with applicable laws and request proof, if possible. They should be able to show evidence that they haven’t observed any adverse impact against protected groups. It might also be a good idea to have a lawyer review tests before you administer them to candidates.

Can integrity tests be faked?

Faking is a problem for all kinds of testing. Overt employee integrity tests make it easy for candidates to tell employers what they want to hear. For example, candidates may have to rate statements like “I have lied to my boss to get out of trouble” or “I would steal from work if I could get away with it.” Most candidates will instantly know which answers are acceptable. So, people who score high on integrity tests could either be ethical paragons or accomplished liars.

What to do:

Ask test providers how their tests deal with candidates faking their answers. Many tests contain mechanisms to “catch” lies (e.g. lie scales). Also, some research suggests that faking doesn’t affect rankings of candidates, although it might affect overall score. So, a well designed test can be effective to decide which candidates will proceed in the hiring process.

Of course, deceptive people can still ‘pass.’ And sometimes faking can have an effect on hiring decisions. Be prepared to take results with a grain of salt.

Can integrity testing screen out good candidates?

False positives are always a concern. Past research found that employee integrity tests result in honest people being labeled dishonest. Some studies even show that overt integrity tests can sometimes misclassify almost half of honest candidates.

Many employers are tempted to use them to shrink their applicant pool. So, they reject everyone who scores below standard. But, if honest, talented employees are among those rejected, companies could be missing out.

What to do:

It’s best to avoid allowing employee integrity tests to make decisions for you. Take some time to look at answers and interpret results. You can also use integrity tests in conjunction with other assessment methods. For example, integrity tests have high incremental validity when they’re paired with cognitive ability tests. This means that using integrity tests can enhance the predictive validity of cognitive tests. In fact, research suggests that those two tests together have the highest predictive power for job performance.

Are employee integrity tests ethical?

Overt integrity tests often measure past dishonest behavior and attitudes towards dishonesty. Both of those measures can create ethical dilemmas.

On one hand, we could wonder whether past offenders should be penalized forever. People can, and often do, repent. Past behavior doesn’t always predict future actions. It’s difficult to be sure about the dishonesty levels of candidates who report stealing. Do they feel free to reveal it because they don’t think it’s bad? Or is it because they’re generally honest and regret their actions? It’s possible that situational factors caused their delinquency.

Attitudes towards dishonesty can also be misleading. Consider questions like “what should happen to an employee caught stealing?” Candidates who answer that perpetrators shouldn’t get arrested or fired aren’t necessarily dishonest. They could just believe in corrective action and second chances. And, candidates who bestow severe punishments to minor offenses don’t necessarily have high integrity. Instead, their answers may show lack of flexibility and close-mindedness.

What to do:

Again, not rushing to reject candidates is probably a good idea. Some test providers give you a detailed analysis of answers. Taking time to interpret them and look for patterns can be a good approach.

It’s true that sometimes employers don’t have time to analyze sophisticated results. For this reason, you could use integrity tests in late stages of your hiring process. Then, you can administer the test to a few good candidates.

Are there objections about research on integrity tests?

Past research by the US Office of Technology Assessment (OTA) raised some concerns on the matter. It suggested that test providers controlled studies concluding that employee integrity tests are valid. This doesn’t necessarily mean that those studies are wrong. But, more independent research could be useful. Most tests providers are also reluctant to let their tests be validated externally. This raises some doubts about their validity.

The OTA research also identified methodology problems. For example, some employee integrity tests were validated through polygraph test results. But the validity of the polygraph itself is doubtful. Same goes for other methods of comparing future theft to test results. Unless employees are proven to have stolen, there’s no way to find the thief. And unavoidably, there’s no guarantee that employees who pass integrity tests won’t steal. They just mightn’t get caught. Research can’t include them in assessing test validity.

What to do:

Ask test providers to prove their test can predict what it was designed to measure. They should be able to explain how they validated it and who participated in the process. You can also ask them to explain to you how they checked the test’s reliability. Lack of official documentation is a red flag.

Is ‘integrity’ always desirable?

It all depends on how you define it. For example, some integrity tests ask candidates to state whether they like taking risks. The logic behind this is people who like risks are more likely to deviate from acceptable behavior. So integrity tests can classify them as ‘low integrity’ individuals. Yet, in many instances, risk takers can be positive for organizations. For example, startup environments or large-scale organizational changes may need people who take chances. Rejecting them might be harmful for some companies.

Also, disputing established rules can lead to innovation and positive change. If employees think a company policy is unfair or ineffective, it might be a good thing to challenge it. Always following the rules can often be counterproductive.

What to do:

It might be best to keep an open mind when interpreting results. Look at which questions count towards an overall ‘integrity index.’ Decide whether it’s important that a candidate once wrote a bad check or cheated on tests at school.

It might be good practice to use test results to drive candidates’ interviews, instead of screening them out. For example, you could use their answers on the test to talk about what risks they like taking and how it has worked for them so far.

The bottomline

Employee integrity tests can be useful. But, they shouldn’t be the sole means of sorting through a candidate pool. Companies shouldn’t use integrity tests to substitute structured interviews or work samples (which are both better predictors of job performance). Instead, companies should use well-designed integrity tests to shed more light on how suitable candidates are. It’s important to be aware of integrity tests’ limitations and interpret their results with caution.