MOSES Subcommand (NPAR TESTS command) 

NPAR TESTS MOSES[(n)]=varlist BY variable(value1,value2)

The MOSES test of extreme reactions tests whether the range of an ordinal variable is the same in a control group and a comparison group. The control and comparison groups are defined by a grouping variable. The output includes a Frequency table, showing, for each variable before BY, the total number of cases and the number of cases in each group. The output also includes a Test Statistics table, showing the number of removed outliers, span of the control group before and after outliers are removed, and one-tailed probability of the span with and without outliers.

Syntax

  • The minimum specification is a test variable, the keyword BY, a grouping variable, and two values in parentheses.
  • The test variable must be at least at the ordinal level of measurement.
  • The first value of the grouping variable defines the control group, and the second value defines the comparison group.
  • By default, 5% of the cases are trimmed from each end of the range of the control group to remove outliers. You can override the default by specifying a value in parentheses following the MOSES subcommand keyword. This value represents an actual number of cases, not a percentage.

Operations

  • Values from the groups are arranged in a single ascending sequence. The span of the control group is computed as the number of cases in the sequence containing the lowest and highest control values.
  • No adjustments are made for tied cases.
  • Cases with values other than values that are specified for the grouping variable are excluded.
  • Test statistics are calculated for each variable that is named before BY.

Example

NPAR TESTS MOSES=V1 BY V3(0,1) /MOSES=V1 BY V3(1,0).
  • The first MOSES subcommand tests V1 by using value 0 of V3 to define the control group and value 1 for the comparison group. The second MOSES subcommand reverses the comparison and control groups.