Fig. 12From: Evaluation of statistical methods used in the analysis of interrupted time series studies: a simulation studyDurbin-Watson tests for autocorrelation. For each combination of length of data series and true magnitude of autocorrelation the Durbin Watson test results from 10,000 simulated data sets are summarised. The horizontal axis is the length of the data series, the vertical axis is the proportion of results indicating: ρ > 0 (blue), ρ < 0, (orange) ρ = 0 (black) and an inconclusive test (grey). Each graph shows results for a different magnitude of autocorrelation. The simulation combination presented is for a level change of 2 and slope change of 0.1; however, other combinations give similar resultsBack to article page