History and Future of Employment Personality Testing

The REP – Pre-Employment Survey

You Already Have A Powerful Team

HR professionals and hiring managers combine to make a powerful team when evaluating the knowledge and work history of candidates. The true effectiveness of this team is proven in an unusual way – failure analysis – the analysis of hires that turn out poorly.

When companies analyze employees fired for poor performance, the reasons are rarely related to inadequate assessment of the candidate’s industry knowledge or work history.

Reasons for dismissal:

Typical answers include:

  • Not getting along with other workers
  • Dishonesty
  • Laziness
  • Indifference to the quality of their work
  • Not following the rules
  • Doing personal business at work
  • Insubordination
  • Tardiness or absenteeism
  • Bad Attitude
  • Poor work habits
  • Wasting time

“No Ask – No Tell”

These are behaviors directly related to the employees’ personality and value systems. Their previous employers know the answers to questions about these behaviors – but we cannot legally ask those questions and they cannot legally answer them. That’s where the field of psychology has typically “jumped in” to help.

Psychological Testing Tries to fill the Gap

Psychologists have been actively involved in the hiring process since World War II, when the US Army asked them to help in the rapid selection of workers in defense factories. Methods were primitive, but the best available at the time.

When Raymond Cattell developed the 16 PF in 1949, he was looking for a way to describe “personality types” - common characteristics that matched clinical profiles. He observed that commonalities existed between people and believed that he could discover various constellations of traits that were common to different personality types. His goal was descriptive in nature. He wanted to more usefully categorize people in order to help them.

The Next Six Decades

The world of personality testing has not moved significantly beyond Cattell in almost six decades. We’ve gotten better at describing traits. We’ve developed more subtle names and definitions of traits. We’ve gotten better at devising questions that “pickup on” characteristics. We’ve gotten more sophisticated at “tweaking” profiles in order to be able to apply them to different situations. Nevertheless, we are still using personality profiles as an intermediate step in the prediction of performance. This extra step introduces significant error.
(See the PowerPoint Presentation: The Problems with Profiles)

The Next Generation: Performance Prediction not Personality Description

Dr. Phillip E. Rosner took a totally different approach to the process of selection. He was always uncomfortable with the prediction error introduced by profiling. It seemed logical that removing that intermediate step would accomplish it neatly.

Predict performance directly from the personality questions. The concept made sense, but was ruled “impossible” by authorities in test construction, statistics and programming. The state of statistics, he was told, did not have the tools to accomplish the task with the needed accuracy. For almost three decades that situation continued to exist.

Dr. Jack L. Edwards, a world recognized expert in Artificial Intelligence (AI), had developed the statistical tools which could make Dr. Rosner’s concept feasible. Working together, they developed the Rosner-Edwards Protocol (REP) (patent pending).

The Rosner Edwards Protocol:

Predicting the Probability of Successful Performance

A Self-Assessment Survey (SAS) is taken by a sample of your highest and lowest performers. A specially designed AI engine determines the 80-100 questions that best differentiate between these two groups. The resulting Unique Selection Survey (USS) is then used as part of the normal selection procedure. The USS predicts the probability that the candidate will be a high-level performer in that unique setting.

The Survey “Gets Smarter”

Over time, the performance of everyone who has taken the survey is updated (quarterly or semi-annually) and a separate AI engine recalculates and adjusts the survey to more accurately predict successful performance.

Advantages

The REP makes a direct connection between the personality questions and the specific performance being predicted. No profile is generated – therefore, no profiling error.

Your people are the normative group, -- therefore that error is eliminated. The result is a prediction of performance in your organization, not a profile

The survey is dynamic not static. This is an on-going, re-adjustment based on additional performance data from everyone who has ever taken the survey. This instrument changes as your products or markets change.

Your survey is unique to your organization. The survey we develop for you is not some kind of average personality profile appropriate to your kind of organization. It is a uniques tool designed to differentiate the potential high-performers (using the specific performance measure you choose) in your organization. No other company has the same survey.

In summary…

We do not use descriptive language subject to multiple interpretations. We answer the single question you want answered:

How likely is this person to be a high-performer in our organization?

We answer it simply – in numeric form – easy to understand – clear.

It becomes a part of the accept/reject process along with the other data you collect, such as skills, intelligence and past performance. Additionally, other benefits to the organization are present:

  • Cost of survey development is greatly reduced
  • The surveys are administered on-line, 24 hours per day/7 days per week.
  • No traveling to corporate office or other site is required.
  • No personnel is required to proctor the instrument.
  • Results are instantaneous and reported to the most convenient location(s).
  • Cost per administration is equal to or less than other non-dynamic tests.