Overview (NPAR TESTS command)

NPAR TESTS is a collection of nonparametric tests. These tests make minimal assumptions about the underlying distribution of the data. 1 In addition to the nonparametric tests that are available in NPAR TESTS, the k-sample chi-square and Fisher’s exact test are available in procedure CROSSTABS.

The tests that are available in NPAR TESTS can be grouped into three broad categories based on how the data are organized: one-sample tests, related-samples tests, and independent-samples tests. A one-sample test analyzes one variable. A test for related samples compares two or more variables for the same set of cases. An independent-samples test analyzes one variable that is grouped by categories of another variable.

The one-sample tests that are available in procedure NPAR TESTS are:

  • BINOMIAL
  • CHISQUARE
  • K-S (Kolmogorov-Smirnov)
  • RUNS

Tests for two related samples are:

  • MCNEMAR
  • SIGN
  • WILCOXON

Tests for k related samples are:

  • COCHRAN
  • FRIEDMAN
  • KENDALL

Tests for two independent samples are:

  • M-W (Mann-Whitney)
  • K-S (Kolmogorov-Smirnov)
  • W-W (Wald-Wolfowitz)
  • MOSES

Tests for k independent samples are:

  • K-W (Kruskal-Wallis)
  • MEDIAN

Options

Statistical Display
In addition to the tests, you can request univariate statistics, quartiles, and counts for all variables that are specified on the command. You can also control the pairing of variables in tests for two related samples.
Random Sampling
NPAR TESTS must store cases in memory when computing tests that use ranks. You can use random sampling when there is not enough space to store all cases.

Basic specification

The basic specification is a single test subcommand and a list of variables to be tested. Some tests require additional specifications. CHISQUARE has an optional subcommand.

Subcommand order

Subcommands can be used in any order.

Syntax rules

  • The KS_SIM, STATISTICS, SAMPLE, and MISSING subcommands are optional. Each subcommand can be specified only once per NPAR TESTS command.
  • You can request any or all tests, and you can specify a test subcommand more than once on a single NPAR TESTS command.
  • If you specify a variable more than once on a test subcommand, only the first variable is used.
  • Keyword ALL in any variable list refers to all user-defined variables in the active dataset.
  • Keyword WITH controls pairing of variables in two-related-samples tests.
  • Keyword BY introduces the grouping variable in two- and k-independent-samples tests.
  • Keyword PAIRED can be used with keyword WITH on the MCNEMAR, SIGN, and WILCOXON subcommands to obtain sequential pairing of variables for two related samples.
  • POISSON = varlist is not supported in the K-S subcommand.

Operations

  • If a string variable is specified on any subcommand, NPAR TESTS will stop executing.
  • When ALL is used, requests for tests of variables with themselves are ignored and a warning is displayed.

Limitations

  • A maximum of 100 subcommands is allowed.
  • A maximum of 48,000 variables total per NPAR TESTS command is allowed.
  • A maximum of 200 values for subcommand CHISQUARE is allowed.
1 Siegel, S., and N. J. Castellan. 1988. Nonparametric statistics for the behavioral sciences. New York: McGraw-Hill, Inc..