In: Statistics and Probability
Chi- Squared Test of Independence
6) Business Non-response questionnaires were mailed to 300 randomly selected businesses in each of three categorical sizes. The following data shows the number of responses.
Size of Company |
Response |
No Response |
Small |
175 |
125 |
Medium |
145 |
155 |
Large |
120 |
180 |
Test the claim that the size of the company is independent of the company`s response
a. State the null and alternative hypothesis |
|
b. Give the p-value |
|
c. Give a conclusion for the hypothesis test |
Please don't hesitate to give a "thumbs up" for the
answer in case the answer has helped you
a. Hypothesis design:
Null hypothesis, Ho: size of the company is independent of the company`s response
Alternate hypothesis, Ha: size of the company is NOT independent of the company`s response
b. p-value.
The contingency table below provides the following information: the observed cell totals, (the expected cell totals) and [the chi-square statistic for each cell].
For example for 1st cell:
We calculate Expected value, E = RowSum*ColumnSum/Total
= 300*440/900 = 146.67
Chi-square = (O-E)^2/E = (146.67-175)^2 / 146.67 = 5.47
Results | ||||||
response | non response | Row Totals | ||||
small | 175 (146.67) [5.47] | 125 (153.33) [5.24] | 300 | |||
medium | 145 (146.67) [0.02] | 155 (153.33) [0.02] | 300 | |||
large | 120 (146.67) [4.85] | 180 (153.33) [4.64] | 300 | |||
Column Totals | 440 | 460 | 900 (Grand Total) |
The chi-square statistic is the sum of individual Chi-Squares in
square brackets in the table above = 5.47 + 5.24 + .02 + .02 +
4.85+ 4.64 = 20.2322.
df = (rows-1)*(column -1) = (3-1)*(2-1) = 2
p-value = 1 - CHISQ.DIST(20.2322, 2, TRUE) = .00004
The p-value is .00004.
Answer: p-value = .00004
c.
The result is significant at p < .05. We reject null hypothesis.
Answer: Conclusion: Size of the company is NOT independent of the company`s response