The Selection Puzzle Pieces Revisited
In our prior two “Puzzling Scenario” blogs, we explored the concepts of how selection ratios and selection test validity affect selection results – that is finding candidates who have the ability and motivation to learn, adapt, advance, and perform well in their jobs.
We learned that the use of tests that lack job-related performance predictive power (e.g. validity) can significantly compromise selection efforts and that lower selection ratios (e.g. 2:1 or hiring 50 percent of the people screened) significantly increase the likelihood of hiring someone who will not be a good match for an organization.
The Missing Selection Puzzle Piece
The solution to the scenarios presented rests with a framework that combines the power of valid tests and rigorous selection ratios. The framework, known as the Taylor-Russell Table helps organizations realize the benefit of higher selection ratios. It’s worth looking at the origins of ability testing and the Taylor-Russell Table to better understand their important role in effective selection.
Tools That Helped Win WW II
In 1939, as American industry was preparing to meet war-driven material and equipment demands, economists and industrial psychologists were investigating ways to rapidly and accurately identify people who would have the abilities needed to learn quickly and perform well in production jobs. A number of ability tests were developed and systematically validated during this time (many of which are still in use today) to help identify people with needed abilities for given jobs. Subsequently, the U.S. military and industry rapidly adopted ability testing for job assignment, and such tests were widely used during and following World War II.
There also was a desire to establish selection systems with a high hit or match rate for the kind of candidates needed in given jobs.
The Taylor-Russell Contribution
Two researchers, with Purdue University roots, H.C. Taylor and J.T. Russell, provided a research paper and supporting tools that led to a significant employee selection process breakthrough. Taylor and Russell created a set of tables (now known as the Taylor-Russell Tables) that provided a consistent, repeatable way to reliably pair employee selection ratios with the predictive power of valid ability tests. Their paper, which appeared in the 1939 Journal of Applied Psychology was titled: “The Relationship of Validity Coefficients to the Practical Effectiveness of Tests in Selection: Discussion and Tables.”
One outcome from the Taylor-Russell work is the ability for table users to identify the percentages of people in an employee selection effort who will be successful from among those selected. (The caveat is that candidates need to be drawn from a normally distributed applicant pool, that a valid process or tests need to be used, and that the selection process needs to be executed in a uniform, rigorous fashion.)
Rigorous Selection Tools Help Grow the U.S. Economy
Here is an illustration of the power of the Taylor-Russell work. During World War II, companies such as Boeing needed to rapidly hire highly able, dependable employees. When Boeing selected people to wire the cockpit of an airplane, they needed a high degree of confidence that their selection efforts would consistently lead to identification of employees with the ability to learn and perform well as airplane cockpit electricians. They could not afford to waste time hiring people who could not learn to do the work. There was daily pressure to manufacture as many airworthy airplanes as possible in the least amount of time.
The Taylor-Russell research enabled organizations like Boeing to wed powerful predictors (e.g. valid ability tests or test batteries) with selection ratios aimed at the top 20 percent of job candidates. One of the reasons that the U.S. prevailed in World War II was because of industry’s ability to maximize productivity through the use of tools such as the Taylor-Russell Tables. Companies such as Boeing repeatedly and reliably found employees who were good job matches for Boeing’s work.
The good news is that the Taylor-Russell work is as relevant today as it was in 1939. In fact, the Taylor-Russell table shows that selection success is possible as much as 90 percent of the time.
Explaining the Taylor-Russell Table
Table I is a Taylor-Russell Table formatted for selection purposes.
The values labeled as “SR” in the top row represent given selection ratios. The ratio of .20 in column five corresponds to the 5:1 selection ratio we advocate as the target for rigorous selection. The values in column two labeled as “r” represent the correlation coefficients of the selection predictors used. In this case, an r value of 0 represents a tool (e.g. a test or an interview) that is completely random while a value of 1.0 represents a perfect tool or system meaning that the predictor would reliably predict the job performance of a person every time.
There is abundant empirical evidence that shows when an effective interview technique is paired with valid, job-related ability tests that the correlation coefficient for such a process is about .60. While this is far from perfect, it’s a respectable predictor value.
Winning 9 out of 10 Times with Selection Decisions
If you look at the intersection in gray between the .60 r value and the .20 SR value, you will see the number “90”. The Taylor-Russell Tables show that if you used a selection system that had validity of .60 and operated with a 5:1 selection ratio, you would be correct (select people who would perform above average) 90% of the time and 10% of the candidates would fall below average. That seems to be a “miss” rate that many of our clients can live with.
The exact same system with a 2:1 (.50 SR) selection ratio will be “right” 80 percent of the time, but the 10 percent difference in the hit rates will entice a lot of organization to go the 5:1 ratio rather than the 2:1. The reason for this is that there is greater risk of a bad hire and the associated costs of managing that bad hire with a 2:1 selection versus a 5:1 ratio. While you may need to look at more candidates with a 5:1 ratio to find the good ones, there is ample evidence to prove the worth of a good hire.
The Potential Bottom Line Impact
The key point is that it’s possible to consistently run an employee selection system at a yield that provides a nine in ten chance of success. If winning in the marketplace is tied to the talents and abilities of people on the team, who wouldn’t want such a guarantee?
No More Guessing™
When we talk about taking the guesswork out of selection, this is what we are talking about. Through a combination of a rigorous valid selection process, discipline to run the process as designed, and the use of a 5:1 selection ratio, employers have the opportunity to consistently and repeatedly hire people with the necessary abilities and motivations to delight customers, make a difference, and deliver winning results.
The measurable value of rigorous selection and effective performance is the topic for a future blog.