Key Questions of Bivariate Tabular Analysis

Is there a relationship between the independent variable and the dependent variable?

Changes in the distributions of the dependent variable across the categories of the independent variable (i.e. the column percentages change across the rows)

How strong is the relationship?

Magnitude of the differences in the percentages across the rows

Magnitude of the measures of association

What direction is the relationship?

Patterns in the column percentages across the rows

Sign of the measures of association

NOTE: You should only describe a relationship as positive or negative if both your independent variable and dependent variable are ordinal or higher.

If your data are from a sample, can we conclude that the relationship we find the in the table is the product of a real relationship in the population from which the sample was drawn (ie. Can we reject the null hypothesis for the population)?

probability values associated with Chi-Square and/or measures of association.

What do these findings mean in plain English?

Describe what the pattern in the data mean in terms of your concepts? (e.g. "Women are more likely to vote Democratic than men are.")

Are there other variables that could account for the relationship you have found?

introduce a control variable in multivariate tabular analysis