Search for most updated materials ↑

 
prev next front |1 |2 |3 |4 |5 |6 |7 |8 |9 |10 |11 |12 |13 |14 |15 |16 |17 |18 |19 |20 |21 |22 |23 |24 |25 |26 |27 |28 |29 |30 |31 |32 |33 |34 |35 |36 |37 |review



Are the overall rates the same in the two groups?

Possible Approaches:

Pearson Chi-Square Test for Comparing Two Independent Proportions 

The value of Pearson’s chi-square statistic, with one degree of freedom, is given by

Chi_2^P = {M_1(P_1 - P)^2 / P (1 - P)} + {M_2 (P_2 - P)^2 / P (1 - P)}= {1479 (.062 - .053)^2 / {(.053) (1 - .053)}} + {1341 (.043 - .053)^2} / {(.053) (1 - .053)}} = 4.69 (p =.03)

But this test is invalid, since it ignores the within-in cluster correlation.