Visual Dashboard: Performing Statistical Analyses with Visual Tools
M x M/ 2 x 2 Table
Epidemiologists frequently use 2 x 2 tables to examine the relationship between two or more categorical values. In these tables, an exposure variable is usually considered as the risk factor. The outcome variable is considered the disease of consequence (e.g., the person had the disease or outcome of interest, or they did not). Values of the first variable will appear on the left margin of the table, and those of the second will be across the top of the table. Normally, cells contain counts of records matching the values in corresponding marginal labels. The M x N /2 x 2 Table analysis gadget in Visual Dashboard is demonstrated below:
- Select the Sample.PRJ Data Source. Open the Oswego form from the Data Source Explorer menu. Click OK.
- Right click on the canvas and select Add Analysis Gadget > M x N /2 x 2 table. The Crosstabulation Properties gadget configuration window opens to the Variables property panel.
- From the Exposure drop-down list, select VANILLA.
- From the Outcome drop-down list, select ILL.
Other options in the gadget allow you to add a weight variable or Stratify the data by another variable. This example does not use these options.
Figure 8.35: Crosstabulation gadget
- Click OK. Results appear on the Visual Dashboard canvas.
Figure 8.36: Crosstabulation results
The results display a 2 x 2 table and a single table analysis of all records crosstabulated by exposure to vanilla where the outcome was ill. Out of the 75 records in this dataset, 43 patients were exposed to vanilla and became ill. Eleven patients were exposed to vanilla and did not become ill. Of the 21 patients who were not exposed to vanilla, only three became ill.
The Single Table Analysis chart on the right displays Odds Ratios, Risk Ratios, and Statistical Tests results for this example.