ABSTRACT
Indicator, or dummy variable regression, as it is often known, employs
qualitative or categorical variables as all or some of its predictor variables,
xis. In the regression models discussed in the previous chapters, the xi predictor variables were quantitative measurements, such as time, tempera-
ture, chemical level, or days of exposure. Indicator regression uses categorical
variables, such as sex, machine, process, anatomical site (e.g., forearm,
abdomen, inguinal region), or geographical location, and these categories
are coded, which allows them to be ranked. For example, female may be
represented as ‘‘0,’’ and male as ‘‘1.’’ Neither sex is rankable or distinguishable,
except by the code ‘‘0’’ or ‘‘1.’’