Several weeks ago I had a meeting with a group of hiring managers for a large manufacturing organization. Their company’s leadership had made the decision to implement a screening tool into their entry-level hiring process. This particular selection instrument was a carefully designed employee assessment by Select International’s Research and Development team. Our team of Industrial/Organizational Psychologists developed an assessment that measures several job relevant risk factors commonly associated with a broad range of industrial jobs. These risk factors are typically found in manufacturing, construction, mining, and similar fields that require physical activity in which employee safety is an important factor.
One of my goals during this hiring manager meeting was to educate the participants on the assessment and gain their buy-in. Although their leadership had already green lit the implementation of this selection process improvement, it was important to get the commitment of the hiring managers. After all, they would be the ones tasked with implementing this change at their facilities. I shared the results of the job analysis that was completed within their organization that established the job relevancy of the factors the assessment measures. I shared with them some examples of successful outcomes from other similar companies that implemented the assessment. We reviewed some example items from the assessment. There was some healthy skepticism from some of the managers about what a person’s responses to specific items on the assessment has to do with being a good employee.
As I was trying to formulate a response related to their skepticism, one of the other hiring managers interjected his thought on the matter. He said, “It’s like a life insurance policy. The adjuster asks you a lot of questions, and based upon your responses they establish a risk level and subsequent policy and payment level.”
Great response! He hit the nail on the head. Insurance companies collect data from policy applicants via questionnaires. They compare those responses to their actuarial databases/tables (i.e. “big data”) to determine the estimated life span of the applicant. Essentially, insurance companies compare their assessment data (the applicant’s questionnaire responses) to the outcome data (average lifespan of persons that had similar responses on the questionnaire).
This is what we do when we empirically validate our assessments. We collect performance data (e.g. safety incidents, recordable injuries, turnover, attendance records, supervisor performance ratings, etc.) and compare it to the assessment data. Based upon that analysis we can determine the relative risk levels of job applicants. Does this mean that the assessment is 100% accurate? Of course not, and neither is a life insurance company’s actuarial table. Just because I have an elevated cholesterol level does not mean that I will die on the year as indicated on an insurer’s actuarial table. However, the response to the cholesterol question in conjunction with the other questions on the life insurance application provide the insurance company with an accurate estimate of my risk level as a policy holder. Knowing the probability of my lifespan (based upon empirical data) provides them with the information they need to make an informed decision and set a policy that effectively manages their risk exposure.
Hiring (especially in high volumes) is no different. Employee selection criteria should be job relevant and informed by empirical validation data to effectively manage the risk associated with hiring a new employee.