hw 7
Hoja1
| HW 7 ANSWER SHEET | |||||
| Task 1: 2-way frequency table (contingency table) | |||||
| RISK: Hypertension | |||||
| yes | no | row marginals | |||
| RISK FACTOR: | yes | 14 | 7 | 21 | |
| Obesity | no | 12 | 27 | 39 | cross check |
| column marginals | 26 | 34 | 60 | 60 | |
| TASK 2: MINITAB output for chi-square test for association and Fisher's exact test | |||||
| Chi-Square Test | |||||
| Chi-Square | DF | P-Value | |||
| Pearson | 7.163 | 1 | 0.007 | ||
| Likelihood Ratio | 7.229 | 1 | 0.007 | ||
| Fisher’s Exact Test | |||||
| P-Value | 0.0131801 | ||||
| ANSWER: | |||||
| Task 3: Risk, relative risk, odds and odds ratio | |||||
| A: Risk (= row proportion) | |||||
| Risk of developing hypertension for obese subjects | 0.667 | ||||
| Risk of developing hypertension for non-obese subjects | 0.308 | ||||
| B: Relative risk (risk ratio or ratio of row proportions) | 2.17 | ||||
| C: Odds to develop hypertension (proportion of hypertension/no hypertension) | |||||
| For obese people | 2.000 | ||||
| for nonobese people | 0.444 | ||||
| D: Odds ratio | 4.5 | ||||
| Task 4: Confidence intervals for risk and odds ratios | |||||
| (output from https://statpages.info/ctab2x2.html) |
Both the Pearson chi-square test and Fisher's Exact test show that the two variables (obesity and hypertension) are associated, because both p-values are below the selected significance level of 0.05. The null hypothesis is rejected.
The risk of developing hypertension is 2.17 fold higher for obese people as compared to nonobese people (95% CI: 1.14 - 3.71).
The odds of developing hypertension are 4.5 times higher for obese people as compared to nonobese people (95% CI 1.4 - 14.0).