St. John Health implemented pre-hire employee testing and interview systems about a year ago. The systems focused on patient-contact jobs ranging from porters to nurses and therapists. Follow-up studies have shown that these systems successfully identify employees who are empathetic, patient-oriented and effective. Significantly, these systems also reduce turnover among new hires by over 40 percent.
Given this impact, we suggest that other businesses may benefit by following these key developmental steps:
Step 1: Define your goals. Before you begin to even think about what your employment selection system should look like, think about what you’re after. Improved performance? Reduced training time? Reduced turnover/absenteeism? Enhanced ease of supervision? Different goals will lead to different decisions regarding the system you need.
Step 2: Develop a success profile. For each goal, analyze what employee characteristics relate to that goal. Personality, skill, mental ability and physical ability characteristics should all be considered. Our primary focus at St. John Health was on patient orientation and turnover, so we emphasized personality and attitudinal characteristics. It should go without saying that all employment “characteristics” should be job related.
Step 3: Decide on assessment methods and tools. Our advice is to keep things simple. We used an application and testing phase to screen out unsuitable candidates and then an interview phase to select the best. We also passed information obtained in the testing stage (e.g., areas where the candidate passed, but was marginal) on to the interviewers so they could probe those areas in further depth. We tailored the tests closely to the jobs, developing a service orientation test from scratch and used a “Polaris Item Library” to form the personality and attitudes test. Commercially available tests can also be used, but we suggest getting professional advice when selecting what to use. The key is to make sure the tests you use are very closely matched to the job responsibilities and activities that you are hiring for.
Step 4: Be prepared. If selection systems result in “disparate impact” (for example, minorities or females are screened out disproportionately), they need to be supported by validation (job-relatedness) documentation to pass legal muster. We conducted a validation study in which current incumbents took the tests on a confidential basis and then we correlated their test scores with their job performance. Commercially available tests can sometimes be supported by existing validation studies that the publishers have already conducted. Just make sure the previous studies focused on jobs that are similar to yours.
Step 5: Think through the administrative issues. Who will administer the tests? When and how? Who will conduct interviews? How will the interviewers be trained? Will you allow participants to retest? Will you use the system to evaluate returning employees? The design of test and interview systems is usually pretty straightforward. But, as you develop the systems, bring together a cross-section of users and other interested parties (such as from operations, HR, legal, training, etc.) and let them ask questions about how the systems will operate. Then, make sure you have thoroughly thought out the answers to their questions–before you implement.
John D. Arnold is president of Polaris Assessment Systems Inc. in Grosse Pointe Park, a Bronze-level member of the Detroit Regional Chamber. Todd Sperl is senior organizational effectiveness consultant for St. John Health, a Gold-level member.