In: Statistics and Probability
An insurance company wants to know what whether perceptions of personal health are independent of gender. The company has hired you as a statistical consultant. You then collected a random sample of people and asked about their perceived health status. Based on the results of this survey, is the sex of the respondent independent from their perceived health status? Use 95% confidence.
The hypothesis being tested is:
H0: Sex of the respondent and perceived health status are independent
Ha: Sex of the respondent and perceived health status are not independent
Women | Men | Total | ||
Very Good | Observed | 89 | 121 | 210 |
Expected | 104.27 | 105.73 | 210.00 | |
O - E | -15.27 | 15.27 | 0.00 | |
(O - E)² / E | 2.24 | 2.20 | 4.44 | |
Good | Observed | 82 | 65 | 147 |
Expected | 72.99 | 74.01 | 147.00 | |
O - E | 9.01 | -9.01 | 0.00 | |
(O - E)² / E | 1.11 | 1.10 | 2.21 | |
Not so good | Observed | 42 | 30 | 72 |
Expected | 35.75 | 36.25 | 72.00 | |
O - E | 6.25 | -6.25 | 0.00 | |
(O - E)² / E | 1.09 | 1.08 | 2.17 | |
Total | Observed | 213 | 216 | 429 |
Expected | 213.00 | 216.00 | 429.00 | |
O - E | 0.00 | 0.00 | 0.00 | |
(O - E)² / E | 4.44 | 4.38 | 8.82 | |
8.82 | chi-square | |||
2 | df | |||
.0121 | p-value |
The p-value is 0.0121.
Since the p-value (0.0121) is less than the significance level (0.05), we can reject the null hypothesis.
Therefore, we can conclude that sex of the respondent and perceived health status are not independent.