In: Statistics and Probability
A research firm was interested in knowing whether income level and location of service for procedures were independent. The following data were provided.
Clinic | Hospital | Private | |
<$30,000 | 85 | 16 | 6 |
$30,000 to $49,999 | 102 | 27 | 13 |
$50,000 to $99,999 | 36 | 22 | 15 |
> $99,999 | 15 | 23 | 25 |
Use a chi-square test of independence to determine whether income level and service location are independent. Test the null hypothesis the service location is independent of income at α = 0.01.
As we are trying to test here whether the 2 variables are independent, we first compute the grand total of each row and column as:
Now the expected value for each of the 12 cells is computed here
as:
Ei = (Rowi Total)(Columni Total) /
Grand Total
The expected value for each cell thus is computed here as:
The expected values are given in the circular bracket above.
Also now the chi square test statistic contribution for each cell here is computed as:
This is given in the square bracket for each cell.
The chi square test statistic now is computed here as:
Now, degrees of freedom here is computed as:
DF = (num of rows - 1)(num of columns - 1) = (4 - 1)(3 - 1) = 6
Now the p-value here is computed as:
As the p-value here is less than 0.01, therefore the test is significant and we can reject the null hypothesis here and conclude that we have sufficient evidence that the 2 variables are not independent here.