The Use of Serial Correlation in the Analysis of Data from Interrupted Time Series Trials with Single Subjects in Educational Research

There is need in educational psychology and elsewhere for a simple but reliable technique for analysing data from baseline/intervention type studies with single subjects. The serial-correlation approach is suitable for this purpose and provides not just a test of statistical significance but also an index of the strength of relationship between intervention and any change, as well as an indication of the proportion of the total variance accounted for by the relationship.