Categorical Data

1
Categorical Data

• Categorical IV Categorical DV

2
Overview

• Defined Frequencies or proportions
  amongst levels of variables.
• Variables IV is categorical, DV is
  categorical
• Relationship Relationship between two or more
  variables.
• Example How do males/females vote guilty
  or not guilty?
• Assumptions Typically you want greater than 5
  per cell.

3
Final main type of analysis
4
Theory behind Chi-Square

• How can you test the relationship between two (or
  more) categorical variables?
• Compare the frequency you observe in cells to the
  frequency you would expect by chance. So compare
  count (sample) to expected count (population)

5
Theory behind Chi-Square

• How can you test the relationship between two (or
  more) categorical variables?
• Compare the frequency you observe in cells to the
  frequency you would expect by chance. So compare
  count (sample) to expected count (population)

6
Theory behind Chi-Square

• How can you test the relationship between two (or
  more) categorical variables?

7
Pearson Chi-Square

• Significance
• Effect Size
• Graph

8
Pearson Chi-Square

• If you have more than a 2 x 2
• such as 2 x 3, 3 x 2, 3 x 3, etc.
• then the output is treated as an Omnibus
  test
• and you conduct the follow-up tests by
  conducting multiple 2 x 2 tests.

9
Example of Omnibus and follow-up tests
10
Example of Omnibus and follow-up tests