The TWOSAMPLEFREQ statement performs power and sample size analyses for tests of two independent proportions. The Farrington-Manning score, Pearson’s chi-square, Fisher’s exact, and likelihood ratio chi-square tests are supported.
Table 77.19 summarizes the options available in the TWOSAMPLEFREQ statement.
Table 77.19: TWOSAMPLEFREQ Statement Options
Option |
Description |
---|---|
Define analysis |
|
Specifies the statistical analysis |
|
Specify analysis information |
|
Specifies the significance level |
|
Specifies the null odds ratio |
|
Specifies the null proportion difference |
|
Specifies the null relative risk |
|
Specifies the number of sides and the direction of the statistical test or confidence interval |
|
Specify effects |
|
Specifies the two independent proportions, and |
|
Specifies the odds ratio |
|
Specifies the proportion difference |
|
Specifies the reference proportion |
|
Specifies the relative risk |
|
Specify sample size and allocation |
|
Specifies the two group sample sizes |
|
Specifies the sample size allocation weights for the two groups |
|
Enables fractional input and output for sample sizes |
|
Specifies the common sample size per group |
|
Specifies the sample size |
|
Specify power |
|
Specifies the desired power of the test |
|
Control ordering in output |
|
Controls the output order of parameters |
Table 77.20 summarizes the valid result parameters for different analyses in the TWOSAMPLEFREQ statement.
Table 77.20: Summary of Result Parameters in the TWOSAMPLEFREQ Statement
To specify the proportions, choose one of the following parameterizations:
individual proportions (by using the GROUPPROPORTIONS= option)
difference between proportions and reference proportion (by using the PROPORTIONDIFF= and REFPROPORTION= options)
odds ratio and reference proportion (by using the ODDSRATIO= and REFPROPORTION= options)
relative risk and reference proportion (by using the RELATIVERISK= and REFPROPORTION= options)
To specify the sample size and allocation, choose one of the following parameterizations:
sample size per group in a balanced design (by using the NPERGROUP= option)
total sample size and allocation weights (by using the NTOTAL= and GROUPWEIGHTS= options)
individual group sample sizes (by using the GROUPNS= option)
This section summarizes the syntax for the common analyses supported in the TWOSAMPLEFREQ statement.
You can use the NPERGROUP= option in a balanced design and express effects in terms of the individual proportions, as in the following statements. Default values for the SIDES= and ALPHA= options specify a two-sided test with a significance level of 0.05.
proc power; twosamplefreq test=pchi groupproportions = (.15 .25) nullproportiondiff = .03 npergroup = 50 power = .; run;
You can also specify an unbalanced design by using the NTOTAL= and GROUPWEIGHTS= options and express effects in terms of the odds ratio. The default value of the NULLODDSRATIO= option specifies a test of no effect.
proc power; twosamplefreq test=pchi oddsratio = 2.5 refproportion = 0.3 groupweights = (1 2) ntotal = . power = 0.8; run;
You can also specify sample sizes with the GROUPNS= option and express effects in terms of relative risks. The default value of the NULLRELATIVERISK= option specifies a test of no effect.
proc power; twosamplefreq test=pchi relativerisk = 1.5 refproportion = 0.2 groupns = 40 | 60 power = .; run;
You can also express effects in terms of the proportion difference. The default value of the NULLPROPORTIONDIFF= option specifies a test of no effect, and the default value of the GROUPWEIGHTS= option specifies a balanced design.
proc power; twosamplefreq test=pchi proportiondiff = 0.15 refproportion = 0.4 ntotal = 100 power = .; run;
The following statements demonstrate a sample size computation for the Farrington-Manning score test for two proportions:
proc power; twosamplefreq test=fm proportiondiff = 0.06 refproportion = 0.32 nullproportiondiff = -0.02 sides = u ntotal = . power = 0.85; run;
The following statements demonstrate a power computation for Fisher’s exact conditional test for two proportions. Default values for the SIDES= and ALPHA= options specify a two-sided test with a significance level of 0.05.
proc power; twosamplefreq test=fisher groupproportions = (.35 .15) npergroup = 50 power = .; run;
The following statements demonstrate a sample size computation for the likelihood ratio chi-square test for two proportions. Default values for the SIDES= and ALPHA= options specify a two-sided test with a significance level of 0.05.
proc power; twosamplefreq test=lrchi oddsratio = 2 refproportion = 0.4 npergroup = . power = 0.9; run;