Question

Suppose patients were recruited into a clinical trial from three different clinical sites. Use the following data to test for differences in the proportion of hypertensive patients across the three sites Site 2 Site 1 Site 3 Hypertensive 10 14 12 Not Hypertensive 68 56 Step 4 In the table below the upper number in each cell is the observed frequency (O) and the lower number (in parentheses) is the expected frequency (E). Compute the observed x2 using the following formula (O-E)2 E Observed freq. Site 1 Site 2 Site 3 Total |(Expected freq.) Hypertensive 10 14 12 36 |(12.6) (14.0) (???) Not Hypertensive 164 68 56 40 (64.0) (57.4) (42.6) Total 78 70 52 200 Step 5: Which one of the following conclusions is correct? O There is a statistically significant relationship between proportion of hypertensive patients and site. O There is a statistically significant difference between proportions of hypertensive patients across sites. O Proportions of hypertensive patients and site are dependent. O Proportions of hypertensive patients are not statistically different across sites. 40

Answer 1

The statistical software output for this problem is:

Contingency table results: Rows: var1 Columns: None Site 1 Site 2 Site 3 Total Hypertensive 10 14 12 36 Not Hypertensive 68 56 40 164 Total 78 70 52 200 Chi-Square test: Value P-value Statistic DF chi-square 2 2.5154611 0.2843

From above output:

We can see that the p - value is very high so we do not reject the null hypothesis. Hence,

Proportion of Hypertensive and site ate independent.

Option C is correct.