How to test the hypothesis based on Chi-square test for independence of attributes.
At a large commercial bakery one of the measures utilized to evaluate employees is the number of broken cookies per 1000 produced. The quality control department wants to determine if there is a relationship between years of experience and the broken cookie rate.
Assume the job is repetitious, so after training, any improvement due to continued learning is likely offset by decreased motivation.
The broken cookie rates are calculated daily. The results for 100 workers is given below:
|
|
|
Years Since Training Period
|
|
|
|
|
<1 Year
|
1-4 Years
|
>4 Years
|
|
|
|
|
Group 1
|
Group 2
|
Group 3
|
Total
|
|
Broken Cookie Rate
|
High
|
6
|
9
|
9
|
24
|
|
|
Average
|
9
|
19
|
23
|
51
|
|
|
Low
|
7
|
8
|
10
|
25
|
|
|
Total
|
22
|
36
|
42
|
100
|
Formally test the null hypothesis using the chi-squared method with α=0.05.
What is your decision?
- Reject the null hypothesis of equal broken cookie rates.
- Fail to reject the null hypothesis of equal broken cookie rates.