25.11.2014 Views

Biostatistics

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

6.6 CONFIDENCE INTERVAL FOR THE DIFFERENCE BETWEEN TWO POPULATION PROPORTIONS 187<br />

6.6 CONFIDENCE INTERVAL FOR<br />

THE DIFFERENCE BETWEEN TWO<br />

POPULATION PROPORTIONS<br />

The magnitude of the difference between two population proportions is often of interest. We<br />

may want to compare, for example, men and women, two age groups, two socioeconomic<br />

groups, or two diagnostic groups with respect to the proportion possessing some characteristic<br />

of interest. An unbiased point estimator of the difference between two population<br />

proportions is provided by the difference between sample proportions, ^p 1 ^p 2 .Aswe<br />

have seen, when n 1 and n 2 are large and the population proportions are not too close to 0 or 1,<br />

the central limit theorem applies and normal distribution theory may be employed to obtain<br />

confidence intervals. The standard error of the estimate usually must be estimated by<br />

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi<br />

^p<br />

^s^p1 ^p 2<br />

¼ 1 ð1 ^p 1 Þ<br />

þ ^p 2ð1 ^p 2 Þ<br />

n 1<br />

n 2<br />

because, as a rule, the population proportions are unknown. A 100ð1<br />

confidence interval for p 1 p 2 is given by<br />

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi<br />

^p<br />

ð^p 1 ^p 2 Þz 1 ð1 ^p 1 Þ<br />

1 a=2 þ ^p 2ð1 ^p 2 Þ<br />

n 1<br />

n 2<br />

aÞ percent<br />

(6.6.1)<br />

We may interpret this interval from both the probabilistic and practical points of view.<br />

EXAMPLE 6.6.1<br />

Connor et al. (A-17) investigated gender differences in proactive and reactive aggression in<br />

a sample of 323 children and adolescents (68 females and 255 males). The subjects were<br />

from unsolicited consecutive referrals to a residential treatment center and a pediatric<br />

psychopharmacology clinic serving a tertiary hospital and medical school. In the sample,<br />

31 of the females and 53 of the males reported sexual abuse. We wish to construct a 99<br />

percent confidence interval for the difference between the proportions of sexual abuse in<br />

the two sampled populations.<br />

Solution:<br />

The sample proportions for the females and males are, respectively, ^p F ¼<br />

31=68 ¼ :4559 and ^p M ¼ 53=255 ¼ :2078. The difference between sample<br />

proportions is ^p F ^p M ¼ :4559 :2078 ¼ :2481. The estimated standard<br />

error of the difference between sample proportions is<br />

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi<br />

ð:4559Þ ð:5441Þ<br />

^s^pF ^p M<br />

¼<br />

þ ð:2078Þ<br />

ð:7922Þ<br />

68<br />

255<br />

¼ :0655<br />

The reliability factor from Appendix Table D is 2.58, so that our confidence<br />

interval, by Expression 6.6.1, is<br />

:2481 2:58 ð:0655Þ<br />

:0791;:4171

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!