1.2. The Paired-Comparisons t-Test as a Fixed Effects Method
Perhaps the simplest design that meets these two requirements is a before-after study. Suppose, for example, that 100 people volunteer to participate in a weight loss program. They all get weighed when they enter the study, producing the variable W1. All 100 people are then given a new medication believed to facilitate weight loss. After two months on this medication, they are weighed again, producing the variable W2. So we have measurements of weight on two occasions for each participant. The participants are off the medication for the first measurement and are on the medication for the second.
How should such data be analyzed? Before answering that question, let's first concede that ...
Get Fixed Effects Regression Methods for Longitudinal Data Using SAS now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
