Chapter 3
More Principles Underlying the General Linear Model
There are two analyses presented in this chapter -- the formulae and computations are the same as the analysis in chapter 2. More data has been added to make it a more realistic problem. The second analysis is designed to show the similarity between analysis of regression and analysis of variance.
This chapter builds on chapter 2 and if there are points that you don't understand because of complexity of numbers it might be useful to refer to the more simplified set in chapter 2. A correlation between continuous variables is presented as the first example then a correlation between a continuous variable and a dichotomous variable will be presented. The similarities between this correlation and an analysis of variance will be shown.
The sample data was selected from a larger set that was administered to 5 different groups including psychiatric inpatients and professional staff that worked with psychiatric inpatients.
The sample data was selected from a larger set that was administered to 5 different groups including psychiatric inpatients and professional staff that worked with psychiatric inpatients.
The problems throughout this chapter use sample data from the preceding questionnaire. It should be recognized that is selected data -- that is incomplete and selected for the purpose of this example. At the same time is does represent results from a larger study. It is somewhat exaggerated here in that the two samples are: (1) patients at the time of admission to an inpatient hospital and (2) professional staff members. The data is randomly selected from those groups excluding subjects who had missing data. Only 20 cases were selected (10 from each group) so that the mathematical calculations can be followed.
