Test Development and Validation
Olson Consulting Group has developed and validated many tests for use in selection. Creating a test is one thing, but making sure the test is valid is a job that requires skill and expertise. With a statistical dissertation, Dr. Richard D. Olson is an expert in test research and development.
We begin by conducting a job analysis. Next, we create a test for research. We ask job incumbents to complete the test and we ask those incumbents' supervisors to rate their performance on the job. At the same time, we carefully train all raters in how to rate performance, as this can be a subjective task full of errors. Finally, we compare the incumbents' test scores with their performance ratings and finalize the test.
When we are able to show a significant relationship between ratings and test scores, our client gains a validated test for use in selection.
Contact us for more information about test validation or to discuss your specific need.
A national clothing retailer increased sales by
27% after implementing our custom designed test in their selection process.
What is a valid test?
First, a test must be reliable to be valid. A reliable test consistently produces the same results.
Next, a test must show statistically significant correlations between test results and actual performance. That is, low scores on the test are made by people who receive low performance ratings, and people who receive high performance ratings score highly on the test.
Finally, a test is only considered valid based on the context in which it is being used. That is, a reliable typing test is valid for a job that requires skill in typing. It is not valid for a job in which typing is not a requirement to be successful.
Contact us at +1.952.830.1190 or firstname.lastname@example.org if you need more information about how testing is used in your organization.