Tagged: Contingency Table

From the ANALYSIS menu of SPSS, the crosstabs procedure in descriptive statistics is used to create contingency tables also known as two-way frequency table, cross tabulation, which describe the association between two categories variables.

In a crosstab, the categories of one variable determine the rows of the contingency table, and the categories of the other variable determine the columns. The contingency table dimensions can be reported as $R\times C$, where $R$ is the number of categories for the row variables, and $C$ is the number of categories for the column variable. Additionally, a “square” crosstab is one in which the row and column variables have the same number of categories. Tables of dimensions $2 \times 2$, $3\times 3$, $4\times 4$, etc., are all square crosstab.

To perform Chi-Square test on cross-tabulation in SPSS, first click Analysis from main menu, then Descriptive Statistics and then crosstabs, as shown in figure below

As an example, we are using “satisf.sav” data file that is already available in SPSS installation folder. Suppose, we are interested in finding the relationship between “Shopping Frequency” and “Made Purchase” variable. For this purpose, shift any one of the variable from left pan to the right pan as row(s) and the other in right pan as column(s). Here, we are taking “Shopping Frequency” as row(s) and “Made Purchase” as column(s) variable. Pressing OK will give the contingency table only.

The ROW(S) box is used to enter one or more variables to be used in the cross-table and Chi-Square statistics. Similarly, the COLUMNS(S) box is used to enter one or more variables to be used in the cross-table and Chi-Square statistics. Note At least one row and one column variable should be used.

When you need to find the association between three or more variables the layer box is used. When the layer variable is specified, the crosstab between the row and the column variables will be created at each level of the layer variable. You can have multiple layers of variables by specifying the first layer variable and then clicking next to specify the second layer variable. Alternatively, you can try out multiple variables as single layers at a time by putting them all in layer 1 of 1 box.

The STATISTICS button will lead to a dialog box which contains different inferential statistics for finding the association between categorical variables.

The CELL button will lead to a dialog box which controls which output is displayed in each cell of the crosstab, such as observed frequency, expected frequency, percentages, and residuals, etc., as shown below.

To perform the Chi-Square test on the selected variables, click on “Statistics” button and choose (tick) the option of “Chi-Square” from the top-left side of the dialog box shown below. Note the Chi-square check box must have tick in it, otherwise only cross-table will be displayed.

Press “Continue” button and then OK button. We will get output windows containing the cross-tabulation results in Chi-Square statistics as shown below

The Chi-Square results indicate that there is association between categories of “Sopping Frequency” variable and “Made Purchase” variable, since, p-value is smaller than say 0.01 level of significance.

For video lecture on Contingency Table, Chi-Square statistics, See the video lectures