Analysis of Covariance Task :: SAS(R) Studio 3.3: User's Guide

About the Analysis of Covariance Task

The Analysis of Covariance task fits a linear model that combines the continuous and categorical predictors of a continuous dependent variable. This task also produces graphical output to interpret the results.

Example: Analyzing the Sashelp.Class Data Set

To create this example:

In the Tasks section, expand the Statistics folder and double-click Analysis of Covariance. The user interface for the Analysis of Covariance task opens.
On the Data tab, select the SASHELP.CLASS data set.

Assign variables to these roles:

Role	Column Name
Dependent variable	Height
Categorical variable	Sex
Continuous variables	Weight

To run the task, click .

Here is a subset of the results:

Graph of Analysis of Covariance for Height

Assigning Data to Roles

To run the Analysis of Covariance task, you must assign columns to the Dependent variable, Categorical variable, and Continuous covariate roles.

Role	Description
Dependent variable	specifies a continuous numeric variable.
Categorical variable	specifies a character or numeric variable that specifies the levels of the groups.
Continuous covariate	specifies a continuous numeric variable that is related to the dependent variable. You can specify whether to center the covariate variable.

Setting Options

Option	Description
Model
Intercepts	specifies whether to use the equal or unequal intercepts for each level of the categorical variable.
Slopes	specifies whether to use the equal or unequal slopes for each level of the categorical variable.
Show parameter estimates	produces a solution to the normal equations (parameter estimates). By default, the task displays a solution if your model does not include any classification variables. Select this option only if you want to see the solution for models with classification effects.
Multiple Comparisons
Perform multiple comparisons	performs the least squares means for the categorical variable.
Covariate value	specifies the value to use in multiple comparisons. The covariate value can be either the mean value or a specified value.
Method	requests a multiple comparison adjustment for the p-values and confidence limits for the differences of LS-means. Here are the available methods: Bonferroni Dunnett Nelson Scheffe Sidak Tukey
Significance level	specifies the significance level for the comparisons. The default is 0.05.
Plots
You can choose to display only the default plots in your output, select the plots to display in the output, or display no plots in the output. The list of available plots depends on the method that you selected for multiple comparisons.

Setting the Output Options

You can specify whether to create an output data set. You can also specify the values to include in the output data set. You can include predicted values, residuals, standard errors, and influence statistics.