Chapter 10--Hypothesis Testing--Categorical Data

Thus we wish to test h0 p 12 vs ha p 12 let nd

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: es between treatments. Let p be the probability that a discordant pair is of type A. Chapter 10: Hypothesis Testing: Categorical Data Stat 491: Biostatistics Introduction Two-Sample Test for Binomial Proportions McNemar’s Test Estimation of Sample Size and Power R × C Contingency Tables Chi-Square Goodness-of-Fit Test The Kappa Statistic McNemar’s Test: Normal-Theory and Exact Tests If the treatments are equally effective then about an equal number of discordant pairs would be expected. Thus we wish to test H0 : p = 1/2 vs Ha : p = 1/2. Let nD and nA be the number of discordant and type A discordant pairs, respectively. For large sample (nD ≥ 20), we can apply the normal-method prop.test(x=n_A,n=n_D, p=0.5,correct=TRUE, alternative="two.sided") For small samples (nD < 20), we can use binom.test(x=n_A,n=n_D,p=0.5, alternative="two.sided") Example: Assess the statistical significance of the chemo data. Chapter 10: Hypothesis Testing: Categorical Data Stat 491: Biostatistics Introduction Two-Sample Test for Binomial Proportions McNemar’s Test Estimation of Sample Size and Power R × C Contingency Tables Chi-Square Goodness-of-Fit Test The Kappa Statistic McNemar’s Test: Example A study is conducted to computer a computer devise for measuring blood pressure with standard methods. Twenty patients are recruited, and their hypertensive status is assessed by the computer device a trained observer. Hypertensive status us defined either hypertensive (+) if systolic blood pressure is ≥ 160 or if diastolic blood pressure is ≥ 95; the patient is considered normotensive (-) otherwise. The following data was obtained. computer<-c(F,F,T,T,F,T,F,T,T,F,T,T,F,T,F,T,T,F,F,F) trained<-c(F,F,F,T,F,F,F,T,T,F,F,F,F,F,T,F,F,F,F,F) The contingency table may be constructed by table(computer,trained) Chapter 10: Hypothesis Testing: Categorical Data Stat 491: Biostatistics Introduction Two-Sample Test for Binomial Proportions McNemar’s Test Estim...
View Full Document

This note was uploaded on 02/03/2014 for the course STAT 491 taught by Professor Solomonharrar during the Fall '12 term at Montana.

Ask a homework question - tutors are online