There’s a special situation in which you can measure dependence using either the correlation or Cramer’s V (Chapter 5). Suppose both variables indicate group membership. One variable might distinguish male from female customers, and the other whether the customer uses a credit card, as in this example. These variables define a contingency table for a sample of 150 customers at a department store.
(a) Compute the value of Cramer’s V for this table. You may need to look back at Chapter 5.
(b) Find the correlation between the numerical variables Sex (coded as 1 for male, 0 for female) and Cash (coded 1 for those paying with cash, 0 otherwise). Use the counts in the table to manufacture these two columns.
(c) What’s the relationship between Cramer’s V and the correlation?

  • CreatedJuly 14, 2015
  • Files Included
Post your question