204 110.3 two prpoprtion z test

204 110.3 two prpoprtion z test - similar patients: The...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
11.3 Tests Comparing Two Proportions Our next statistical test will compare two sample proportions p 1 ' = p 2 ' = This test assumes large sample sizes, thus the standard normal z distribution is used. z = σ p1-p2 is the standard deviation for the difference between the two proportions and is calculated as follows: σ p1-p2 = Standard error of the difference between two proportions. If the null hypothesis is true , Ho: p 1 = p 2 , the formula for the above standard error can be rewritten without the subscripts and we obtain the simpler formula: Since p and q are not known we can use an estimate of p obtained by pooling the data. z = ) 1 1 ( ' ' 2 1 2 2 1 1 n n q p n x n x + - p' = q' = 1 – p’ Example: Assume a new medical procedure is being proposed which is claimed to have a smaller chance of a serious side effect. A researcher decides to test this claim obtaining information on 1000
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: similar patients: The first group of 500 had the new procedure while the second group of 500 had the second procedure. The following results were observed Procedure n number with the serious side effect New (1) 500 48 Current(2) 500 85 Is the above data statistically significant that the new procedure reduces the proportion of patients who have the Serious side effect? ∧ 1 p = = 500 48 = .096 ∧ 2 p = = = .170 1) Ho: p 1 = p 2 Ha: 2) α = .05 3) reject Ho if Z Z ≥ α = 1.645 4) p' = = .133 q' = .867 z = 5) TI PHStat Stats 2 Prop ZTest 2 sample tests x 1 48 z = -3.446 Z test differences in proportions n 1 500 pval = 2.849 x10-4 Hypothesized difference x 2 85 sample 1 nbr successes 48 n 2 500 sample size 500 p 1 <p 2 sample 2 nbr successes 85 sample size 500 lower tailed test...
View Full Document

This note was uploaded on 12/05/2011 for the course MATH 2040 taught by Professor Raysievers during the Fall '10 term at Utah Valley University.

Page1 / 2

204 110.3 two prpoprtion z test - similar patients: The...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online