Much of the literature on the analysis of longitudinal data assumes that the data is complete (no observations are missing in the repeated measurements of the same individual). Yet longitudinal studies typically have missing data. Recent research has focused on the development of statistical methods that will permit the complexities of typical longitudinal data set. Wei and Johnson (l985) have proposed a procedure to allow the investigator to draw an overall conclusion whether the new treatment group constitutes an improvement over the control group for the entire study period. The test procedure allows different patterns of missing observations for the two groups to be compared. This new statistical method will be investigated and illustrated by the analysis of incomplete longitudinal dental data.
