Advertisement
They Just Don’t Get It: Part 2
Get articles like this
in your inbox
The longest running and most trusted source of information serving talent acquisition professionals.
Advertisement
Related Articles
In Part 1, we discussed the relative strengths and weaknesses of techie tests. In this part we’ll examine the personality test vendors and the automated selection systems?and make a few recommendations about how you can better design a recruiting website. Accurately Measuring Applicants – Personality Tests This area seems to be growing with leaps and bounds. Too bad. Personality tests are far less reliable than techie tests and filled with more empty promises. Look around you. Do all the people who work with you have the same personality? No? Do all the people with the same personality have the same performance level? No? Then why believe the hogwash that personality tests predict job performance? To date, there are still only about ten solid factors associated with job performance. The rest are either junk science, or someone’s idea of a good training program. Training tests won’t help much with selecting good people, or dodging the inevitable lawsuit (Target Stores lost a lawsuit that cost them millions?all because someone used the wrong kind of selection test). Empty Promise Number 1: We will give our test to people in the job, calculate an average, and then compare each applicant’s scores to the average. Give me a break! Think about it. Averages hide individual differences. The average shoe size of 1000 college students was a size 10. But 900 students in the study wore either larger or smaller sizes – 90% of people in the study did not match the group average! Average scores tell you nothing about individual performance differences. Empty Promise Number 2: Our personality test predicts job success. We know this because we have developed a personality type test that includes 360-degree ratings and performance appraisals. It also predicts managerial style. Give me a break! This claim has two answers. First, the seller does not know the difference between causative and correlational job factors. Correlational data means that blond hair and blue eyes are “correlated” with being Scandinavian; but having blond hair and blue eyes does not “cause” you to be Scandinavian. Similarly, managerial or personality style does not “cause” a person to be successful or unsuccessful. Do all the managers with the same management style perform the same? Do all the good managers have the same management style? Second, survey ratings (i.e., 360-degree ratings) are notoriously subject to rater bias and halo?politically savvy and charismatic people tend to get higher scores than their equally skilled, but grouchy, co-workers. Furthermore, jobs tend to be 50% hard skills and 50% “schmoozing.” So, where is the “skill” piece in the personality test? And, how truthful is performance appraisal information where you work? Personality has about a 2% relationship with hard skills such as problem solving. You want to make 100K decisions on those kinds of odds? Accurately Measuring Applicants – Automated Application Blanks This area addresses the automated and non-automated recruiting websites. These sites tend to stress the paper flow, standardized application form, and applicant management process. Empty Promise Number 1: We ask applicants to “sell me the pen.” Give me a break! If I sell the pen to Marty on Monday and Jill on Tuesday, will I be selling under the same conditions? Who sets the scoring standards to assure that every applicant gets the same situation? Is Marty a sadist? Is Jill a pushover? Are the selling techniques the right ones for the job? Role-plays are tests. Good tests use standardized conditions and standardized scoring guides. Empty Promise Number 2: We eliminate the resume and standardize the hiring process by weighting and rating application blank data. Give me a break! Do you know people with different backgrounds who tend to perform the same? Do you know people with less experience who are high performers? How about people with more experience who are low performers? Standardized application blanks will improve processing efficiency, but is a mistake to think they will minimize hiring errors. Self-reported skills are still self-reported skills. Whether it is on a resume or a standardized application form, self-reports tend to exaggerate skills and hide deficiencies. If you are a seeker of truth, you need to be a better investigator. Empty Promise Number 3: We use interviews and reference checks to assess applicant skills. Give me a break! You don’t have to believe the research on this one. Just do a little homework. Call up the hiring manager in nine months and ask about the quality of the applicant’s: 1) cognitive ability, 2) planning ability, 3) interpersonal skills, and 4) attitudes, interests and motivations. It won’t take long to realize that your interview data is about 50% accurate. True, background checks have their place: screening out “bottom feeders,” but they seldom “screen in” people with good job skills. Summary – Automated Hiring Effectiveness It is not surprising that a website designed by recruiters looks like an automated application blank, a website designed by techies looks like a technical test, and a website designed by trainers tends to look like prework for a communication workshop. We all have a tendency to build on what we know. However, the truly efficient recruiting website of tomorrow will go beyond those limitations by having a roundtable of experts contributing to product development. These people will include recruiters, HR staffers, measurement experts, professional test developers, technical experts, and web programmers. Combining these talents will produce websites that include a full compliment of tools. These include: