The role of testing realism on experimentally obtained stereotype strength

It has been shown (Hoffmann 2009) that pencil-and-paper tests, when used to determine stereotype strength, may yield data quite different to that obtained from hardware tests. The purpose of this research was to determine the level of realism required in order to have stereotypes that are consistent with real-world values. It is surprising that there does not appear to be any published research addressing this problem; all previous researches have considered either paper/pencil or hardware tests or sometimes both, but no level of simulation in between these techniques. In order to determine the effect of level of testing realism, a specific example was chosen to test, that of water tap operation stereotypes. Tests used verbal questioning, photographs of tap arrangements, a partial hardware arrangement and finally a full hardware setup. The results of these four levels of realism were very different and illustrated that, only with a full realistic simulation, did participants respond as might be expected in a real-world environment, illustrating the importance of testing realism when determining population stereotypes. This result has strong implications for product designers when selecting appropriate layouts of displays and controls and the linkages relating these.