Situations and their influence on the measurement of latent traits

The overall aim of this research is to expand our understanding of the theoretical constructs that apply when people deal with situations. The specific context of this investigation is the use of simulation-based assessment techniques, primarily assessment centres and situation judgement tests. Simulation-based techniques in applied psychology are widely used, are perceived to be fair, are proven less biased, and predict future performance. However, these techniques suffer from construct validity problems and it is not clear what they actually measure. Furthermore, when combined with more efficient psychometric tests of individual differences they do not appear to offer substantial incremental predictive power. Four studies were completed to identify the nature of the constructs that can be applied to explain how people deal with situations. First I examined the fit of different theoretical models to explain performance in assessment centres and concluded that an interactionist model is most appropriate but existing constructs do not reveal its nature. I then developed and applied several methodological innovations using a series of low fidelity simulation paradigms. First I systematically varied content across situations and identified how individual differences affect performance. Then I extended both the range of measures used and situation complexity, and identified how cognitive situation models offer an alternative explanation as to how people deal with simulations. Finally, I measured the nature of the cognitive situation models that participants develop and make use of when transferring performance to new situations. I conclude by discussing how our theories in this area can be extended to incorporate cognitive situation models to help explain individual differences in conjunction with existing psychometric constructs. I argue that an improved understanding of the constructs explaining how people deal with situations offers a potential route to improve assessment practice and the prediction of future potential