It has been an interesting month. I have received feedback from my last article [Internet Recruiting is Doomed!, June 15, 2000] telling me I was an idiot, clueless, negative, had “industrial-age ideas,” needed therapy, had little to offer recruiters, and wrote the most useless article ever written. OUCH! That hurt! Maybe I should have just advised readers how to do flip-searches. It would have been much safer and far less controversial. Nevertheless, I value feedback, and want to thank all the people who took time to write. I didn’t always agree with your opinions, but I read each one of them. Anyway, before you fire off more nasty-grams, be so kind as to follow my reasoning. To begin with, there are only a few recruiting questions that always yield trustworthy answers. They include:
- Are you now, or have you ever been, dead?
- Are you able to respond to the questions that I am asking you?
- Do you hold conversations with your Rice Crispies?
- Have you ever stopped breathing for longer than a week?
- Are you more than 137 years old?
- Do you believe your shoes are cleverly disguised aliens?
Unfortunately, measuring job skills is not always this clear cut. True, the Internet offers a great way to scan huge amounts of candidate data?but it is not being used efficiently. Doomed to die? No. Doomed to repeat the mistakes of yesterday? Yes. Will it get better? That remains to be seen. The Two-Minute Recruiter We have all been encouraged to believe that professional competency can be easily and quickly acquired. Even “how-to” instructions on the Internet are “parceled” into 1200 word chunks–longer text is boring and we click to another page. However, common sense would tell us that learning complicated skills takes time and practice. Remember, “You can have it either quick, cheap, or good. Pick one.” Sorry if I frustrated people by not solving their Internet recruiting problems in my last article. For those good folks who are willing to invest the necessary time and energy into escaping from “Internet Groundhog Day” dilemma, it all starts and ends with these five steps:
- Build a set of competencies (not job activities) that are capable of being accurately measured, can differentiate activities across jobs, can differentiate performance within jobs, and are time-insensitive. Easier said than done.
- Use this list of competencies to build profiles for each job by interviewing jobholders, managers, and visionary managers. Verify and prioritize your information through confirming surveys. Weight the competencies for job importance. Be sure the job analysis establishes the business need and job requirements. Use these competencies to advertise your job.
- Always use the most effective measurement tool to assess each candidate competency, i.e., behavioral interviewing, structured role-plays, case studies, tests, simulations, etc. Be sure each measurement tool has a standardized scoring guide to assure uniform administration and evaluation and has been evaluated for potential adverse impact. Casual interview may help build friendships, but they don’t predict job performance worth diddly.
- Run formal studies to prove that the tools and tests you use to measure competencies have significant correlations with accurately rated job skills. Document the studies with tables indicating adverse impact at each critical step in the selection process. Validation is an art that takes time, experience and skill.
- Be sure your competency list thoroughly measures all four critical job areas, i.e., problem solving, planning, interpersonal, and attitudes, interests and motivations. Don’t over measure or under measure any area.
There. That’s the two-minute answer. If you don’t have the wherewithal to do these five steps yourself, it’s ok. Not all of us are able to remove our own appendix, either. If it’s any consolation, even the “big dogs” tend to crash and burn in these areas. If you have time, budget, and foresight, you can hire a team of experienced I/O staff psychologists. But, if you don’t, you need to spend some bucks to hire a competent consultant to help you set up a system. Just be careful. There are many well-meaning people willing to help who are flat-out incompetent. A quick way to separate a true consultant from a wannabe is to check for an advanced degree in people measurement, see if they belong to the right professional associations, and, finally, ask him/her three questions:
- What is the difference is between a training competency, a management competency, and a selection competency? If he/she cannot explain the differences, keep looking; he/she does not have the skills to accurately measure job competencies.
- How is a selection test validated? If he/she says it is by finding “average” scores of high producers, keep looking; he/she does not have the skills to accurately measure job competencies.
- What tools are used for assessing different job competencies? If he/she only recommends pencil and paper tests or computer-based tools, keep looking; he/she does not have the skills to accurately measure job competencies.
Oops! A few people have boldly asked me asked me to check out their internet-based tests and “competency” systems. Some are better than others, but none are able to deliver what they promise. This is not unusual or unexpected. Unless you have considerable experience differentiating jobs and measuring human skills; it only seems logical to think that a list of tasks or technical skills will solve the problem. Task lists are useful for internal placement and training, but they are too detailed and burdensome for selection. Task lists are not competencies. They are sub-sets of competencies. I have similar comments regarding the so-called “selection” tests I have been asked to review. Not a single test would meet reputable standards for technical integrity – yet their promoters enthusiastically advertise these tests as “valid” for selection. I do not think these people are mean-spirited, I think they just don’t know what they don’t know. Buyer beware! Getting sued is the least of your worries here. Your greatest risk in using junk-science is divided between being perceived as unprofessional and trusting totally meaningless test results. A Quick Check If you want to take a quick inventory of the psychometric integrity of either your Internet site or recruiting efforts, take a few minutes to read through the recruiting practices spelled out by the Department of Labor over 30 years ago?they will show you how to get the best people for the job. You can download a copy of “Testing And Assessment: An Employers Guide to Good Practices” from my website at www.ScientificSelection.com. (Even if you are not subject to U.S. law, these practices still include the best-known recruiting technology). I want to warn you, following the Guide will not make your life easier. On the contrary, it will make your recruiting job much more difficult?you will just become a better recruiter and make significant contributions to your organization. If you don’t have the skills and experience to implement the Guide yourself, work with someone who does. You don’t have to be an installation expert, just an application expert. Investing in a good consultant will save you a fortune. Tales From The Mountain Top It is not always easy to see the results of your labor when using good selection tools, but here are a few examples of what can happen when you do it the right way:
- Bank reduces turnover 25% in competitive labor market.
- Electronics manufacturer profitable in one year instead of an expected two years.
- Turnover rate half that of a competitor in the same labor market.
- Foreign auto manufacturer out-produces Detroit Big-Three by 3 to 1 in U.S.-staffed plant.
These organizations are not miracles in the making. They just hired good consultants who studied job requirements, developed good measurement systems, and trained their staff members how to select qualified people. Think of it like this. You have the task of assembling the company softball team. You pick nine people who pass an interview and resume screen. Your competition picks nine people who pass an interview, resume screen and successfully complete exhaustive tryouts. Who do you think will win more games? Sorry. Got to go. I see a mob of angry villagers with torches ready to storm my castle. By the way, thanks to all the people who wrote nice things to me.