Creating Contingency Tables using StatCrunch

Overview

• Instruction on creating contingency tables using StatCrunch.
• Using a toy data set named "two categorical variables" with columns var1 and var2.

Data Set Details

• var1: Contains 10 values; 4 rows with value 'b' and 6 rows with value 'a'.
• var2: Contains 10 values; 6 rows with value 'c' and 4 rows with value 'd'.

Creating Contingency Table

1. Procedure:

   • Under the stat menu, select: Tables > Contingency > With Data.
   • Choose var1 as the row variable and var2 as the column variable.
   • Click on 'Compute'.

2. Results:

   • Table displays unique values for each column in the first row and column.
   • Remaining cells show frequency for each variable pairing.
   • Includes row totals, column totals, and total number of pairs.
   • Displays results of a standard chi-square test for independence.
   • Warning for low cell counts in chi-square test results.

Customizing Contingency Table

• Edit Options:

  • Under the options menu, choose 'Edit' to customize.
  • Under display options, select additional information to display.
  • Example: Select 'Row Percent' and click 'Compute'.

• Resulting Table:

  • Shows that 66.6% of 'a' values are paired with 'c', and 33.3% with 'd'.
  • Shows a 50/50 split between 'c' and 'd' for 'b' values.

Hypothesis Testing

• Chi-square Test:

  • Not suitable for small cell counts in this data.
  • Options available under 'Hypothesis Test' in the options menu.

• Alternative Test:

  • De-select chi-squared and select Fisher's exact test for independence.
  • Additional option: StatCrunch can calculate confidence intervals (not demonstrated).

• Fisher's Exact Test:

  • Click 'Compute' for results.
  • Displays new results for Fisher's exact test.