Table Analysis Task

About the Table Analysis Task

The Table Analysis task enables you to generate crosstabulation tables, also known as contingency tables, from your data.

Example:

To create this example:
  1. In the Tasks section, expand the Statistics folder and double-click Table Analysis. The user interface for the Table Analysis task opens.
  2. On the Data tab, select the SASHELP.CARS data set.
  3. Assign columns to these roles:
    Role
    Column
    Row variable
    Type
    Column variable
    DriveTrain
  4. To run the task, click Submit SAS Code.
Here is a sample of the results:
Example of Frequency Table for the Table Analysis Task
Example of Graph in the Table Analysis Task

Assigning Data to Roles

To run the Table Analysis task, you must first assign at least one column to the Role variables or Column variables roles. After you have assigned a row variable or a column variable, you can assign a column to the Strata variable role.
Roles
Description
Roles
Row variables
creates the rows for two-way to n-way frequency and crosstabulation tables.
Column variables
creates the columns for two-way to n-way frequency and crosstabulation tables.
Strata variables
creates the separate tables for n-way frequency and crosstabulation tables.
Additional Roles
Frequency count
specifies that each row in the table is assumed to represent n observations. In this example, n is the value of the frequency count for that observation.

Setting Options

Option Name
Description
Plots
By default, plots are included in the results. To suppress these plots, select the Suppress plots check box.
Frequency Table
Frequencies
Observed
displays the frequency count for each cell.
Expected
displays the expected cell frequency for each cell.
Deviation
displays the deviation of the cell frequency from the expected value for each cell.
Percentages
Cell
display of overall percentages in crosstabulation tables.
Row
display of row percentages in crosstabulation table cells.
Column
display of column percentages in crosstabulation table cells.
Cumulative
Column percentages
displays the cumulative column percentage in each cell.
Frequencies and percentages
displays the cumulative frequencies and percentages in one-way frequency tables.
Cell contributions to the chi-square statistics
displays each table cell’s contribution to the Pearson chi-square statistic in the crosstabulation table.
Statistics
Chi-square statistics
requests chi-square tests of homogeneity or independence and measures of association that are based on the chi-square statistic. The tests include the Pearson chi-square, likelihood-ratio chi-square, and Mantel-Haenszel chi-square. For 2×2 tables, this test includes Fisher's exact test and the continuity-adjusted chi-square.
Measures of association
computes several measures of association and their asymptotic standard errors (ASE). The measures include gamma, Kendall's tau-b, Stuart's tau-c, Somers' D (C|R), Somers' D (R|C), the Pearson and Spearman correlation coefficients, lambda (symmetric and asymmetric), and uncertainty coefficients (symmetric and asymmetric).
Cochran-Mantel-Haenszel statistics
requests Cochran-Mantel-Haenszel statistics, which test for association between the row and column variables after adjusting for the remaining variables in a multiway table. These statistics include the CMH correlation statistic, the row mean scores (ANOVA), and the adjusted relative risks and odds ratios.
Measures of agreement (for square tables)
computes tests and measures of classification agreement for square tables. This option provides McNemar's test for 2×2 tables and Bowker's test of symmetry for tables with more than two response categories. It also produces the simple kappa coefficient, the weighted kappa coefficient, the asymptotic standard errors for the simple and weighted kappas, and the corresponding confidence limits. When there are multiple strata and two response categories, this option also computes Cochran's Q test.
Odds ratio and relative risk (for 2x2 tables)
requests relative risk measures and their asymptotic Walk confidence limits for 2x2 tables.
Binomial proportions and risk differences (for 2x2 tables)
requests risks (binomial proportions) and risk differences for 2x2 tables.
Methods
Missing value treatment
specifies how to treat missing values:
Exclude missing values
specifies that an observation is excluded from a table if the observation has a missing value for any of the variables.
Display missing value frequencies
displays the frequencies of the missing values in the frequency and crosstabulation tables. These frequencies are not included in any computations of percentages, tests, or measures.
Include missing values in calculations
treats the missing values as valid for all variables.
Exact Test
Fisher’s exact test
requests Fisher’s exact test for tables that are larger than 2x2.