The NPAR1WAY Procedure

OUTPUT Statement

  • OUTPUT < OUT=SAS-data-set > < output-options > ;

The OUTPUT statement creates a SAS data set that contains statistics that PROC NPAR1WAY computes. Table 83.4 lists the statistics that can be stored in the output data set. You identify which statistics to include by specifying output-options.

The output data set contains one observation for each analysis variable that you name in the VAR statement. If you use a BY statement, the output data set contains an observation or set of observations for each BY group. For more information, see the section Contents of the Output Data Set.

As an alternative to the OUTPUT statement, you can use the Output Delivery System (ODS) to store statistics that PROC NPAR1WAY computes. ODS can create a SAS data set from any table that PROC NPAR1WAY produces. For more information, see the section ODS Table Names and Chapter 20: Using the Output Delivery System.

The OUTPUT statement data set does not include Monte Carlo estimates of exact p-values, multiple comparison statistics, or stratified analysis statistics. You can use ODS to store these statistics in output data sets.

You can specify the following options in the OUTPUT statement:

OUT=SAS-data-set

names the output data set. When you use an OUTPUT statement but do not specify the OUT= option, PROC NPAR1WAY creates a data set and names it by using the DATAn convention.

output-options

specifies the statistics to include in the output data set. Table 83.4 lists the output-options that are available in the OUTPUT statement. Descriptions of the output-options follow the table in alphabetical order.

When you specify an output-option, the output data set includes the test statistic and associated values from the analysis that you specify. The associated values might include standardized statistics, one- and two-sided p-values, exact p-values, degrees of freedom, and confidence limits. See the section Contents of the Output Data Set for a list of output data set variables that each output-option produces.

If you do not specify any output-options in the OUTPUT statement, the output data set includes all available statistics from the analyses that you request in the PROC NPAR1WAY statement. If you request a statistic in the OUTPUT statement but do not request the corresponding analysis in the PROC NPAR1WAY statement, PROC NPAR1WAY performs the corresponding analysis.

Table 83.4: OUTPUT Statement Output Options

Output Option

Output Data Set Statistics

AB

Ansari-Bradley test

ANOVA

Analysis of variance

CONOVER

Conover test

EDF | KS

Kolmogorov-Smirnov test,

 

Cramér–von Mises test, and

 

Kuiper test for two-sample data

FP

Fligner-Policello test

HL

Hodges-Lehmann estimates

KLOTZ

Klotz test

MEDIAN

Median test

MOOD

Mood test

SAVAGE

Savage test

SCORES=DATA

Test with input data as scores

ST

Siegel-Tukey test

VW | NORMAL

Van der Waerden (normal scores) test

WILCOXON

Wilcoxon test for two-sample data and

 

Kruskal-Wallis test


You can specify the following output-options:

AB

includes statistics from the Ansari-Bradley analysis in the output data set. The AB option in the PROC NPAR1WAY statement requests analysis of Ansari-Bradley scores. For more information, see the section Ansari-Bradley Scores.

ANOVA

includes analysis of variance statistics in the output data set. The ANOVA option in the PROC NPAR1WAY statement requests a standard analysis of variance for the raw data.

CONOVER

includes statistics from the Conover analysis in the output data set. The CONOVER option in the PROC NPAR1WAY statement requests analysis of Conover scores. For more information, see the section Conover Scores.

EDF
KS

includes empirical distribution function statistics in the output data set. The EDF option in the PROC NPAR1WAY statement requests computation of these statistics, which include the Kolmogorov-Smirnov test, the Cramér–von Mises test, and the Kuiper test for two sample-data. For more information, see the section Tests Based on the Empirical Distribution Function.

FP

includes the Fligner-Policello test in the output data set. The FP option in the PROC NPAR1WAY statement requests the Fligner-Policello test, which is available for two-sample data. For more information, see the section Fligner-Policello Test.

HL

includes the Hodges-Lehmann estimate and confidence limits in the output data set. The HL option in the PROC NPAR1WAY statement requests Hodges-Lehmann estimation, which is available for two-sample data. For more information, see the section Hodges-Lehmann Estimation of Location Shift.

KLOTZ

includes statistics from the Klotz analysis in the output data set. The KLOTZ option in the PROC NPAR1WAY statement requests analysis of Klotz scores. For more information, see the section Klotz Scores.

MEDIAN

includes statistics from the median analysis in the output data set. The MEDIAN option in the PROC NPAR1WAY statement requests analysis of median scores. For more information, see the section Median Scores.

MOOD

includes statistics from the Mood analysis in the output data set. The MOOD option in the PROC NPAR1WAY statement requests analysis of Mood scores. For more information, see the section Mood Scores.

SAVAGE

includes statistics from the Savage analysis in the output data set. The SAVAGE option in the PROC NPAR1WAY statement requests analysis of SAVAGE scores. For more information, see the section Savage Scores.

SCORES=DATA
PERM

includes statistics from the analysis of data scores in the output data set. The SCORES=DATA option in the PROC NPAR1WAY statement requests an analysis that uses the raw input data as scores. For more information, see the section Scores for Linear Rank and One-Way ANOVA Tests.

ST

includes statistics from the Siegel-Tukey analysis in the output data set. The ST option in the PROC NPAR1WAY statement requests analysis of Siegel-Tukey scores. For more information, see the section Siegel-Tukey Scores.

VW
NORMAL

includes statistics from the Van der Waerden (normal scores) analysis in the output data set. The VW (NORMAL) option in the PROC NPAR1WAY statement requests analysis of Van der Waerden scores. For more information, see the section Van der Waerden (Normal) Scores.

WILCOXON

includes statistics from the Wilcoxon analysis in the output data set. The WILCOXON option in the PROC NPAR1WAY statement requests analysis of Wilcoxon scores. For two-sample data, this analysis includes the Wilcoxon rank-sum test. For all data (two-sample and multisample), the analysis includes the Kruskal-Wallis test. For more information, see the section Wilcoxon Scores.