The General Linear Model
8 videos • 49 views • by Fourth Z As you consider all of the models we have discussed in this course, you might be struck by the similarities. All of the models include a quantitative response variable on the left side of the equal sign. They all have an intercept. The right side of the equation is always a linear combination of terms, and each of these terms involve a coefficient and an explanatory variable. The models are never perfect, so we always have residuals. There are some apparent differences, yet on closer inspection these differences merge into similarities. Categorical explanatory variables become multiple quantitative (dummy) variables. Differences in between- and within-subjects factors can be addressed through how we partition sums-of-squares, yet otherwise are similar in the model. In short, the more things change, the more they same the same. All of the designs we have studied can be accommodated in a model of one general form. This is referred to as the general linear model. In this unit we will relate designs to this model and see how the model can be written in terms of the effects of interest.