Investigating the Predictive Validity of the COMPASS Pilot Selection Test

ABSTRACT

Objective: The present study examined the predictive validity of a widely used pilot selection test known as COMPASS, in terms of its ability to distinguish between candidates and predict individuals’ aptitude for flying.

Background: The Computerized Pilot Aptitude Screening System, more commonly known as COMPASS is a widely used selection test for ab-initio pilots. However, a literature search failed to uncover any research examining its predictive validity.

Method: Thirty-five trainee pilots completed COMPASS, along with a simulated flight on a computer-based flight simulator to obtain an objective flight performance measure. Hours to flying milestones, including first solo and Recreational Pilot License, along with more subjective measures such as performance ratings by flying instructors were also compared to COMPASS. Twenty non-pilots also completed COMPASS.

Results: The results revealed COMPASS to be effective in distinguishing between trainee pilots and non-pilots. However, there was no relationship between COMPASS scores and the various flight performance measures.

Conclusion: These results should aid decision-makers when considering the viability of pilot selection tests, and reiterate the importance of using multiple methods/tools when attempting to select appropriate candidates for a specific role.

Acknowledgments

We would like to thank all the students who volunteered their time to complete the research. Without their support, such research would not be possible. We would also like to thank Mr De Manincor, Mr Harte, Mr Hyde, and Mr Arnold from the Flying Operations Unit at UNSW for their assistance with COMPASS and the FSTD. Finally, we would like to acknowledge the relationship between UNSW and TU Munich. It is through this relationship and specifically the Practicum programme that this research was possible.

Disclosure Statement

In accordance with Taylor & Francis policy and my ethical obligation as a researcher, I am reporting that no external funding was received to carry out the research performed in this manuscript, and hence no potential conflict of interest exists.