This paper describes the use of grounded theory techniques to develop a performance model for fighter crews undergoing initial training in pairs combat techniques on the Tornado F3 Operational Conversion Unit (OCU). The work was undertaken as part of a study to evaluate the training effectiveness of the JOUST low-fidelity, multi-player simulator system used for pairs air combat training. The OCU instructors considered that students performed poorly in communications related aspects of pairs combat, and attributed this to the fact that the available synthetic training aids only took a single crew. They had identified the JOUST system as being a more appropriate training device.

The empirical evidence, in the form of archived narrative reports of past student performance in the pairs phase, was investigated in order to develop a student performance model, identifying the essential aspects of performance upon which the instructors were focused. A total of 46 reports completed by an equal mix of pilot an d navigator instructors (80%) of the instructor population) were analysed to provide 200 performance statements. The performance model was developed through successive application of open, axial and selective coding of these statements. Grounded theory technique was selected for evaluating this qualitative data as it provided a rigorous analytical procedure, and it facilitated the development of a model of student performance by analysis independent of input form the subject matter experts, namely the instructors whose opinions were to be validated. The model that was developed was found to support the anecdotal evidence from the instructors on students' communications problems, and was consistent with published models of situational awareness in air combat.