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.

244 CHAPTER 7 HYPOTHESIS TESTING<br />

Dialog box:<br />

Session command:<br />

Stat Basic Statistics 2-Sample t MTB > TwoSample 95.0 C1 C2<br />

SUBC> Alternative 1,<br />

Choose Samples in different columns. Type C1 SUBC> Pooled.<br />

in First and C2 in Second. Click the Options box<br />

and select “less than” in the Alternatives box.<br />

Check Assume equal variances. Click OK.<br />

Output:<br />

Two-Sample T-Test and CI: C, SCI<br />

Two-sample T for C vs SCI<br />

N Mean StDev<br />

C 10 126.1 21.8<br />

SCI 10 133.1 32.2<br />

SE Mean<br />

6.9<br />

10<br />

Difference mu C mu SCI<br />

Estimate for difference: 7.0<br />

95% upper bound for difference: 14.3<br />

T-Test of difference 0 (vs jtj. The default output is a<br />

p value for a two-sided test. The researcher using SAS ® must divide this quantity in half<br />

when the hypothesis test is one-sided. The SAS ® package also tests for equality of<br />

population variances as described in Section 7.8. Figure 7.3.3 shows the SAS ® output<br />

for Example 7.3.2.<br />

Alternatives to z and t Sometimes neither the z statistic nor the t statistic is<br />

an appropriate test statistic for use with the available data. When such is the case, one<br />

may wish to use a nonparametric technique for testing a hypothesis about the difference<br />

between two population measures of central tendency. The Mann-Whitney test statistic<br />

and the median test, discussed in Chapter 13, are frequently used alternatives to the z and<br />

t statistics.

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

Saved successfully!

Ooh no, something went wrong!