ABSTRACT
In this chapter we show a detailed analysis of a three-way treatment structure when many treatment combinations are missing.
A home economist conducted a sample survey experiment to study how much lowersocioeconomic-level mothers knew about nutrition and to judge the effect of a training program designed to increase their knowledge of nutrition. A test was administered to the mothers both before and after the training program, and the changes in their test scores were measured. These changes are reported in Table 17.1. The mothers tested were classi- ed according to three factors: age, race, and whether they were receiving food stamps.
