Few phenomena in psychology have been studied more than reaction time. However, while response latency now provides the basic data for research in many areas of cognitive psychology, little work has been done on the problem of incorporating response latency in the measurement of individual differences. A solution to this problem requires the construction of a mathematical model relating reaction time to individual and test item variation. The present research is concerned with the development and empirical validation of such a model. The model developed here provides a simultaneous analysis of test item responses and the latencies associated with those responses. The method is sufficiently general to be used in the analysis of a wide variety of data, ranging from responses to conventional ability tests to data collected in experimental problem-solving situations. The application of this model to a set of spatial visualizing ability test data is discussed. The results indicate that the model provides a parsimonious summary of the raw data, and estimates of individual differences useful in further analysis. (Author)