True or false?: To analyze categorical data, you must convert the counts into percentages.
False. When analyzing categorical data, the counts or percents of individuals that fall into each category may be used.
What is a marginal distribution?
The distribution of one categorical variable among all individuals described by the table.
What does the marginal distribution tell us about the variables’ relationship?
Nothing.
True or false?: In describing the relationship between two categorical variables, it is customary to use percents rather than counts.
True. Using counts in this circumstance can be harder to compare and misleading.
What is a conditional distribution?
The distribution of values for a variable among only individuals who have a given value of the other variable.
True or false?: Lurking variables only influence quantitative variables.
False. Lurking variables can influence the relationships between categorical variables too.
What is Simpson’s paradox?
Simpson’s paradox describes the situation in which an association or comparison that holds for several groups can reverse direction when the data are combined to form a single group.
