One thing about being a consultant is that you have the opportunity to see a lot of different approaches to solving the same problem. Eventually, if you are lucky, you learn that both problems and solutions tend to fall into patterns. I have been romping through the “Best of the Best” sites and “Wannabe” sites, and have observed that they still “don’t get it yet.” The only two patterns now on the Internet are “Band-Aid” solutions or high-tech solutions that use yesterday’s technology. The goal of a good recruiting website is to fit people into jobs-similar to fitting pieces into a puzzle. In other words, an organization has a problem to solve, they create a job to solve this problem, the job requires certain unique skills, a person with matching skills takes the job, and the problem is solved. So far, so good. But, wait! What is currently being done to assure this happens? Automation? Nice, but not good enough. Techie tests? Also nice, but not good enough. It takes more… No web-based recruiting process can be better than the underlying data it automates. This involves the completeness of job requirements (i.e., does the puzzle have any hidden shape requirements), the ability to accurately measure each requirement (i.e., what shape really is the piece), and the system it uses to assess the degree of fit (i.e., just how well does the piece fit). By the way, if you’re in this business for the money, you don’t need to read any farther. It will be a waste of your time. If, on the other hand, you want to be the best site on the web or do the best recruiting job possible, please continue. Complete and Accurate Job Requirements Hiring Manager: “These applicants are worthless! You’re wasting my time! Bring me QUALIFIED people to interview!” Recruiter: “What are you looking for? I used the data you gave me on the requisition.” Hiring Manager: “No! No! No! Those people were no good. I’ll know it when I see it. Just don’t waste my time!” Recruiter (under his/her breath): “Jerk!” Hiring Manager (under his/her breath): “Jerk!” Problem: When job requirements are unclear, hiring managers tend to use a gut-level standard that: 1) they are unable to communicate, 2) changes with the job, 3) changes with each interview, or 4) changes when one applicant is compared with another. (Does the term “moving target” sound familiar?) Solution: Comprehensive organizational information is needed about the job. This does not always come from the hiring manager. It requires a documented job analysis (i.e., detailed questions of jobholders and managers that are converted into weighted, measurable competencies confirmed by questionnaires). It should cover all four performance areas: 1) cognitive ability, 2) planning ability, 3) interpersonal skills, and 4) attitudes, interests and motivations. An incomplete job requisition will inevitably lead to the recruiter’s version of “pin the tail on the donkey.” Even if an applicant is hired, incomplete job requirements will contribute to high turnover, low productivity, and poor organizational performance, not to mention potential legal problems in the U.S. Accurately Measuring Applicants – Technical Tests This area is almost as elusive as incomplete job requirements, with one exception-tests offer a false sense of security. You know, like signing up for a dating service and expecting to meet your ideal mate. Enter, stage left, the techie test vendors. Techie tests are a nice addition. But scores may not deliver everything you expect. Specifically, it is wrong to assume that any single test score, techie or otherwise, is strongly associated with good job performance. For example:
- Look around you. Do all the people on your staff with the same test scores or certifications perform equally well?
- How do you know the applicant did not become “test smart” by practicing first with other tests?
- Is a high score really the “best” score for your job or would a lesser score do?
- Do your score requirements screen out qualified candidates who could otherwise learn on the job?
- Is the techie test “technically” sound? A good test takes more than a panel of content experts to generate questions. This is only the first step. Good tests have documented studies that show scores predict performance levels; they contain only items that have been statistically examined for level of difficulty; the answers for each question contain items that are equally difficult; and individual tests contain items that are drawn from a large item bank. In other words, does the test really measure what it pretends to measure?
- Finally, does the techie test recognize that there are THREE different levels of techie knowledge and that a multiple choice test only measures the first one?
Web-based test vendors: If you don’t understand these requirements, your test is probably not as strong as you believe it is. Web-based test users: If your current vendor cannot provide answers to these questions, you need to find one who can. Even if a techie test was technically sound, did you do your homework setting valid cut-off points for your organization? The test vendor only needs to show the test is good. Organizations need to show the test is good for their positions. There are cases in the courts now against organizations that skipped this step-and they are losing. We live in a litigious society-especially in the US. Are you ready to explain to your company president why you ignored a fundamental principle of testing and selection? And how do you address the fact that when line managers are asked for reasons why people succeed or fail, their answers sound like this:
Article Continues Below
- Can’t manage time
- Can’t get along with people
- Doesn’t seem to care
- Can’t sell
- Not a “people person”
- Ignores deadlines
- Poor communication with customers
- Doesn’t like the work
- Not a team player
- Doesn’t support organizational goals
- Can’t/wouldn’t learn new information
- Poor attitude
- Bad “fit”
Did you notice that techie skills didn’t make the list? This is not an oversight. Techie skills are seldom mentioned. Techie managers just assume that techies will have techie knowledge. It’s the other things that lead to job failure. These are not measured using techie tests. Any effective Web-based hiring system must have a way to include additional skill definitions beyond techie data-otherwise it will have major flaws that will be eventually exploited by competitors. But it is not enough to just list a series of competencies. In many cases competencies that are developed by people who are not experts turn out to be “mind candy” that are basically immeasurable. Enter, stage right, the personality test vendors and the automated selection system vendors. To be continued in my next article…