4 This chapter documents the statistical procedures that PSPP supports so
8 * DESCRIPTIVES:: Descriptive statistics.
9 * FREQUENCIES:: Frequency tables.
10 * EXAMINE:: Testing data for normality.
11 * CORRELATIONS:: Correlation tables.
12 * CROSSTABS:: Crosstabulation tables.
13 * FACTOR:: Factor analysis and Principal Components analysis
14 * NPAR TESTS:: Nonparametric tests.
15 * T-TEST:: Test hypotheses about means.
16 * ONEWAY:: One way analysis of variance.
17 * RANK:: Compute rank scores.
18 * REGRESSION:: Linear regression.
19 * RELIABILITY:: Reliability analysis.
20 * ROC:: Receiver Operating Characteristic.
30 /MISSING=@{VARIABLE,LISTWISE@} @{INCLUDE,NOINCLUDE@}
31 /FORMAT=@{LABELS,NOLABELS@} @{NOINDEX,INDEX@} @{LINE,SERIAL@}
33 /STATISTICS=@{ALL,MEAN,SEMEAN,STDDEV,VARIANCE,KURTOSIS,
34 SKEWNESS,RANGE,MINIMUM,MAXIMUM,SUM,DEFAULT,
35 SESKEWNESS,SEKURTOSIS@}
36 /SORT=@{NONE,MEAN,SEMEAN,STDDEV,VARIANCE,KURTOSIS,SKEWNESS,
37 RANGE,MINIMUM,MAXIMUM,SUM,SESKEWNESS,SEKURTOSIS,NAME@}
41 The @cmd{DESCRIPTIVES} procedure reads the active file and outputs
43 statistics requested by the user. In addition, it can optionally
46 The VARIABLES subcommand, which is required, specifies the list of
47 variables to be analyzed. Keyword VARIABLES is optional.
49 All other subcommands are optional:
51 The MISSING subcommand determines the handling of missing variables. If
52 INCLUDE is set, then user-missing values are included in the
53 calculations. If NOINCLUDE is set, which is the default, user-missing
54 values are excluded. If VARIABLE is set, then missing values are
55 excluded on a variable by variable basis; if LISTWISE is set, then
56 the entire case is excluded whenever any value in that case has a
57 system-missing or, if INCLUDE is set, user-missing value.
59 The FORMAT subcommand affects the output format. Currently the
60 LABELS/NOLABELS and NOINDEX/INDEX settings are not used. When SERIAL is
61 set, both valid and missing number of cases are listed in the output;
62 when NOSERIAL is set, only valid cases are listed.
64 The SAVE subcommand causes @cmd{DESCRIPTIVES} to calculate Z scores for all
65 the specified variables. The Z scores are saved to new variables.
66 Variable names are generated by trying first the original variable name
67 with Z prepended and truncated to a maximum of 8 characters, then the
68 names ZSC000 through ZSC999, STDZ00 through STDZ09, ZZZZ00 through
69 ZZZZ09, ZQZQ00 through ZQZQ09, in that sequence. In addition, Z score
70 variable names can be specified explicitly on VARIABLES in the variable
71 list by enclosing them in parentheses after each variable.
73 The STATISTICS subcommand specifies the statistics to be displayed:
77 All of the statistics below.
81 Standard error of the mean.
87 Kurtosis and standard error of the kurtosis.
89 Skewness and standard error of the skewness.
99 Mean, standard deviation of the mean, minimum, maximum.
101 Standard error of the kurtosis.
103 Standard error of the skewness.
106 The SORT subcommand specifies how the statistics should be sorted. Most
107 of the possible values should be self-explanatory. NAME causes the
108 statistics to be sorted by name. By default, the statistics are listed
109 in the order that they are specified on the VARIABLES subcommand. The A
110 and D settings request an ascending or descending sort order,
120 /FORMAT=@{TABLE,NOTABLE,LIMIT(limit)@}
121 @{AVALUE,DVALUE,AFREQ,DFREQ@}
122 /MISSING=@{EXCLUDE,INCLUDE@}
123 /STATISTICS=@{DEFAULT,MEAN,SEMEAN,MEDIAN,MODE,STDDEV,VARIANCE,
124 KURTOSIS,SKEWNESS,RANGE,MINIMUM,MAXIMUM,SUM,
125 SESKEWNESS,SEKURTOSIS,ALL,NONE@}
127 /PERCENTILES=percent@dots{}
128 /HISTOGRAM=[MINIMUM(x_min)] [MAXIMUM(x_max)]
129 [@{FREQ[(y_max)],PERCENT[(y_max)]@}] [@{NONORMAL,NORMAL@}]
130 /PIECHART=[MINIMUM(x_min)] [MAXIMUM(x_max)]
131 [@{FREQ,PERCENT@}] [@{NOMISSING,MISSING@}]
133 (These options are not currently implemented.)
139 The @cmd{FREQUENCIES} procedure outputs frequency tables for specified
141 @cmd{FREQUENCIES} can also calculate and display descriptive statistics
142 (including median and mode) and percentiles,
143 @cmd{FREQUENCIES} can also output
144 histograms and pie charts.
146 The VARIABLES subcommand is the only required subcommand. Specify the
147 variables to be analyzed.
149 The FORMAT subcommand controls the output format. It has several
154 TABLE, the default, causes a frequency table to be output for every
155 variable specified. NOTABLE prevents them from being output. LIMIT
156 with a numeric argument causes them to be output except when there are
157 more than the specified number of values in the table.
160 Normally frequency tables are sorted in ascending order by value. This
161 is AVALUE. DVALUE tables are sorted in descending order by value.
162 AFREQ and DFREQ tables are sorted in ascending and descending order,
163 respectively, by frequency count.
166 The MISSING subcommand controls the handling of user-missing values.
167 When EXCLUDE, the default, is set, user-missing values are not included
168 in frequency tables or statistics. When INCLUDE is set, user-missing
169 are included. System-missing values are never included in statistics,
170 but are listed in frequency tables.
172 The available STATISTICS are the same as available in @cmd{DESCRIPTIVES}
173 (@pxref{DESCRIPTIVES}), with the addition of MEDIAN, the data's median
174 value, and MODE, the mode. (If there are multiple modes, the smallest
175 value is reported.) By default, the mean, standard deviation of the
176 mean, minimum, and maximum are reported for each variable.
179 PERCENTILES causes the specified percentiles to be reported.
180 The percentiles should be presented at a list of numbers between 0
182 The NTILES subcommand causes the percentiles to be reported at the
183 boundaries of the data set divided into the specified number of ranges.
184 For instance, @code{/NTILES=4} would cause quartiles to be reported.
186 The HISTOGRAM subcommand causes the output to include a histogram for
187 each specified numeric variable. The X axis by default ranges from
188 the minimum to the maximum value observed in the data, but the MINIMUM
189 and MAXIMUM keywords can set an explicit range. Specify NORMAL to
190 superimpose a normal curve on the histogram. Histograms are not
191 created for string variables.
193 The PIECHART adds a pie chart for each variable to the data. Each
194 slice represents one value, with the size of the slice proportional to
195 the value's frequency. By default, all non-missing values are given
196 slices. The MINIMUM and MAXIMUM keywords can be used to limit the
197 displayed slices to a given range of values. The MISSING keyword adds
198 slices for missing values.
200 The FREQ and PERCENT options on HISTOGRAM and PIECHART are accepted
201 but not currently honored.
204 @comment node-name, next, previous, up
208 @cindex Normality, testing for
212 VARIABLES=var_list [BY factor_list ]
213 /STATISTICS=@{DESCRIPTIVES, EXTREME[(n)], ALL, NONE@}
214 /PLOT=@{BOXPLOT, NPPLOT, HISTOGRAM, ALL, NONE@}
216 /COMPARE=@{GROUPS,VARIABLES@}
219 /PERCENTILE=[value_list]=@{HAVERAGE, WAVERAGE, ROUND, AEMPIRICAL, EMPIRICAL @}
220 /MISSING=@{LISTWISE, PAIRWISE@} [@{EXCLUDE, INCLUDE@}]
221 [@{NOREPORT,REPORT@}]
225 The @cmd{EXAMINE} command is used to test how closely a distribution is to a
226 normal distribution. It also shows you outliers and extreme values.
228 The VARIABLES subcommand specifies the dependent variables and the
229 independent variable to use as factors for the analysis. Variables
230 listed before the first BY keyword are the dependent variables.
231 The dependent variables may optionally be followed by a list of
232 factors which tell PSPP how to break down the analysis for each
233 dependent variable. The format for each factor is
239 The STATISTICS subcommand specifies the analysis to be done.
240 DESCRIPTIVES will produce a table showing some parametric and
241 non-parametrics statistics. EXTREME produces a table showing extreme
242 values of the dependent variable. A number in parentheses determines
243 how many upper and lower extremes to show. The default number is 5.
246 The PLOT subcommand specifies which plots are to be produced if any.
248 The COMPARE subcommand is only relevant if producing boxplots, and it is only
249 useful there is more than one dependent variable and at least one factor. If
250 /COMPARE=GROUPS is specified, then one plot per dependent variable is produced,
251 containing boxplots for all the factors.
252 If /COMPARE=VARIABLES is specified, then one plot per factor is produced, each
253 each containing one boxplot per dependent variable.
254 If the /COMPARE subcommand is ommitted, then PSPP uses the default value of
257 The ID subcommand also pertains to boxplots. If given, it must
258 specify a variable name. Outliers and extreme cases plotted in
259 boxplots will be labelled with the case from that variable. Numeric or
260 string variables are permissible. If the ID subcommand is not given,
261 then the casenumber will be used for labelling.
263 The CINTERVAL subcommand specifies the confidence interval to use in
264 calculation of the descriptives command. The default it 95%.
267 The PERCENTILES subcommand specifies which percentiles are to be calculated,
268 and which algorithm to use for calculating them. The default is to
269 calculate the 5, 10, 25, 50, 75, 90, 95 percentiles using the
272 The TOTAL and NOTOTAL subcommands are mutually exclusive. If NOTOTAL
273 is given and factors have been specified in the VARIABLES subcommand,
274 then then statistics for the unfactored dependent variables are
275 produced in addition to the factored variables. If there are no
276 factors specified then TOTAL and NOTOTAL have no effect.
279 If many dependent variable are given, or factors are given for which
280 there are many distinct values, then @cmd{EXAMINE} will produce a very
281 large quantity of output.
284 @section CORRELATIONS
289 /VARIABLES = varlist [ WITH varlist ]
294 /VARIABLES = varlist [ WITH varlist ]
295 /VARIABLES = varlist [ WITH varlist ]
298 [ /PRINT=@{TWOTAIL, ONETAIL@} @{SIG, NOSIG@} ]
299 [ /STATISTICS=DESCRIPTIVES XPROD ALL]
300 [ /MISSING=@{PAIRWISE, LISTWISE@} @{INCLUDE, EXCLUDE@} ]
304 The @cmd{CORRELATIONS} procedure produces tables of the Pearson correlation coefficient
305 for a set of variables. The significance of the coefficients are also given.
307 At least one VARIABLES subcommand is required. If the WITH keyword is used, then a non-square
308 correlation table will be produced.
309 The variables preceding WITH, will be used as the rows of the table, and the variables following
310 will be the columns of the table.
311 If no WITH subcommand is given, then a square, symmetrical table using all variables is produced.
314 The @cmd{MISSING} subcommand determines the handling of missing variables.
315 If INCLUDE is set, then user-missing values are included in the
316 calculations, but system-missing values are not.
317 If EXCLUDE is set, which is the default, user-missing
318 values are excluded as well as system-missing values.
321 If LISTWISE is set, then the entire case is excluded from analysis
322 whenever any variable specified in any @cmd{/VARIABLES} subcommand
323 contains a missing value.
324 If PAIRWISE is set, then a case is considered missing only if either of the
325 values for the particular coefficient are missing.
326 The default is PAIRWISE.
328 The PRINT subcommand is used to control how the reported significance values are printed.
329 If the TWOTAIL option is used, then a two-tailed test of significance is
330 printed. If the ONETAIL option is given, then a one-tailed test is used.
331 The default is TWOTAIL.
333 If the NOSIG option is specified, then correlation coefficients with significance less than
334 0.05 are highlighted.
335 If SIG is specified, then no highlighting is performed. This is the default.
338 The STATISTICS subcommand requests additional statistics to be displayed. The keyword
339 DESCRIPTIVES requests that the mean, number of non-missing cases, and the non-biased
340 estimator of the standard deviation are displayed.
341 These statistics will be displayed in a separated table, for all the variables listed
342 in any /VARIABLES subcommand.
343 The XPROD keyword requests cross-product deviations and covariance estimators to
344 be displayed for each pair of variables.
345 The keyword ALL is the union of DESCRIPTIVES and XPROD.
353 /TABLES=var_list BY var_list [BY var_list]@dots{}
354 /MISSING=@{TABLE,INCLUDE,REPORT@}
355 /WRITE=@{NONE,CELLS,ALL@}
356 /FORMAT=@{TABLES,NOTABLES@}
357 @{LABELS,NOLABELS,NOVALLABS@}
362 /CELLS=@{COUNT,ROW,COLUMN,TOTAL,EXPECTED,RESIDUAL,SRESIDUAL,
363 ASRESIDUAL,ALL,NONE@}
364 /STATISTICS=@{CHISQ,PHI,CC,LAMBDA,UC,BTAU,CTAU,RISK,GAMMA,D,
365 KAPPA,ETA,CORR,ALL,NONE@}
368 /VARIABLES=var_list (low,high)@dots{}
371 The @cmd{CROSSTABS} procedure displays crosstabulation
372 tables requested by the user. It can calculate several statistics for
373 each cell in the crosstabulation tables. In addition, a number of
374 statistics can be calculated for each table itself.
376 The TABLES subcommand is used to specify the tables to be reported. Any
377 number of dimensions is permitted, and any number of variables per
378 dimension is allowed. The TABLES subcommand may be repeated as many
379 times as needed. This is the only required subcommand in @dfn{general
382 Occasionally, one may want to invoke a special mode called @dfn{integer
383 mode}. Normally, in general mode, PSPP automatically determines
384 what values occur in the data. In integer mode, the user specifies the
385 range of values that the data assumes. To invoke this mode, specify the
386 VARIABLES subcommand, giving a range of data values in parentheses for
387 each variable to be used on the TABLES subcommand. Data values inside
388 the range are truncated to the nearest integer, then assigned to that
389 value. If values occur outside this range, they are discarded. When it
390 is present, the VARIABLES subcommand must precede the TABLES
393 In general mode, numeric and string variables may be specified on
394 TABLES. In integer mode, only numeric variables are allowed.
396 The MISSING subcommand determines the handling of user-missing values.
397 When set to TABLE, the default, missing values are dropped on a table by
398 table basis. When set to INCLUDE, user-missing values are included in
399 tables and statistics. When set to REPORT, which is allowed only in
400 integer mode, user-missing values are included in tables but marked with
401 an @samp{M} (for ``missing'') and excluded from statistical
404 Currently the WRITE subcommand is ignored.
406 The FORMAT subcommand controls the characteristics of the
407 crosstabulation tables to be displayed. It has a number of possible
412 TABLES, the default, causes crosstabulation tables to be output.
413 NOTABLES suppresses them.
416 LABELS, the default, allows variable labels and value labels to appear
417 in the output. NOLABELS suppresses them. NOVALLABS displays variable
418 labels but suppresses value labels.
421 PIVOT, the default, causes each TABLES subcommand to be displayed in a
422 pivot table format. NOPIVOT causes the old-style crosstabulation format
426 AVALUE, the default, causes values to be sorted in ascending order.
427 DVALUE asserts a descending sort order.
430 INDEX/NOINDEX is currently ignored.
433 BOX/NOBOX is currently ignored.
436 The CELLS subcommand controls the contents of each cell in the displayed
437 crosstabulation table. The possible settings are:
453 Standardized residual.
455 Adjusted standardized residual.
459 Suppress cells entirely.
462 @samp{/CELLS} without any settings specified requests COUNT, ROW,
463 COLUMN, and TOTAL. If CELLS is not specified at all then only COUNT
466 The STATISTICS subcommand selects statistics for computation:
473 Pearson chi-square, likelihood ratio, Fisher's exact test, continuity
474 correction, linear-by-linear association.
478 Contingency coefficient.
482 Uncertainty coefficient.
498 Spearman correlation, Pearson's r.
505 Selected statistics are only calculated when appropriate for the
506 statistic. Certain statistics require tables of a particular size, and
507 some statistics are calculated only in integer mode.
509 @samp{/STATISTICS} without any settings selects CHISQ. If the
510 STATISTICS subcommand is not given, no statistics are calculated.
512 @strong{Please note:} Currently the implementation of CROSSTABS has the
517 Pearson's R (but not Spearman) is off a little.
519 T values for Spearman's R and Pearson's R are wrong.
521 Significance of symmetric and directional measures is not calculated.
523 Asymmetric ASEs and T values for lambda are wrong.
525 ASE of Goodman and Kruskal's tau is not calculated.
527 ASE of symmetric somers' d is wrong.
529 Approximate T of uncertainty coefficient is wrong.
532 Fixes for any of these deficiencies would be welcomed.
538 @cindex factor analysis
539 @cindex principal components analysis
540 @cindex principal axis factoring
541 @cindex data reduction
544 FACTOR VARIABLES=var_list
546 [ /METHOD = @{CORRELATION, COVARIANCE@} ]
548 [ /EXTRACTION=@{PC, PAF@}]
550 [ /PRINT=[INITIAL] [EXTRACTION] [UNIVARIATE] [CORRELATION] [COVARIANCE] [DET] [SIG] [ALL] [DEFAULT] ]
554 [ /FORMAT=[SORT] [BLANK(@var{n})] [DEFAULT] ]
556 [ /CRITERIA=[FACTORS(@var{n})] [MINEIGEN(@var{l})] [ITERATE(@var{m})] [ECONVERGE (@var{delta})] [DEFAULT] ]
558 [ /MISSING=[@{LISTWISE, PAIRWISE@}] [@{INCLUDE, EXCLUDE@}] ]
561 The FACTOR command performs Factor Analysis or Principal Axis Factoring on a dataset. It may be used to find
562 common factors in the data or for data reduction purposes.
564 The VARIABLES subcommand is required. It lists the variables which are to partake in the analysis.
566 The /EXTRACTION subcommand is used to specify the way in which factors (components) are extracted from the data.
567 If PC is specified, then Principal Components Analysis is used. If PAF is specified, then Principal Axis Factoring is
568 used. By default Principal Components Analysis will be used.
570 The /METHOD subcommand should be used to determine whether the covariance matrix or the correlation matrix of the data is
571 to be analysed. By default, the correlation matrix is analysed.
573 The /PRINT subcommand may be used to select which features of the analysis are reported:
577 A table of mean values, standard deviations and total weights are printed.
579 Initial communalities and eigenvalues are printed.
581 Extracted communalities and eigenvalues are printed.
583 The correlation matrix is printed.
585 The covariance matrix is printed.
587 The determinant of the correlation or covariance matrix is printed.
589 The significance of the elements of correlation matrix is printed.
591 All of the above are printed.
593 Identical to INITIAL and EXTRACTION.
596 If /PLOT=EIGEN is given, then a ``Scree'' plot of the eigenvalues will be printed. This can be useful for visualising
597 which factors (components) should be retained.
599 The /FORMAT subcommand determined how data are to be displayed in loading matrices. If SORT is specified, then the variables
600 are sorted in descending order of significance. If BLANK(@var{n}) is specified, then coefficients whose absolute value is less
601 than @var{n} will not be printed. If the keyword DEFAULT is given, or if no /FORMAT subcommand is given, then no sorting is
602 performed, and all coefficients will be printed.
604 The /CRITERIA subcommand is used to specify how the number of extracted factors (components) are chosen. If FACTORS(@var{n}) is
605 specified, where @var{n} is an integer, then @var{n} factors will be extracted. Otherwise, the MINEIGEN setting will
606 be used. MINEIGEN(@var{l}) requests that all factors whose eigenvalues are greater than or equal to @var{l} are extracted.
607 The default value of @var{l} is 1. The ECONVERGE and ITERATE settings have effect only when iterative algorithms for factor
608 extraction (such as Principal Axis Factoring) are used. ECONVERGE(@var{delta}) specifies that iteration should cease when
609 the maximum absolute value of the communality estimate between one iteration and the previous is less than @var{delta}. The
610 default value of @var{delta} is 0.001.
611 The ITERATE(@var{m}) setting sets the maximum number of iterations to @var{m}. The default value of @var{m} is 25.
613 The @cmd{MISSING} subcommand determines the handling of missing variables.
614 If INCLUDE is set, then user-missing values are included in the
615 calculations, but system-missing values are not.
616 If EXCLUDE is set, which is the default, user-missing
617 values are excluded as well as system-missing values.
619 If LISTWISE is set, then the entire case is excluded from analysis
620 whenever any variable specified in the @cmd{VARIABLES} subcommand
621 contains a missing value.
622 If PAIRWISE is set, then a case is considered missing only if either of the
623 values for the particular coefficient are missing.
624 The default is LISTWISE.
631 @cindex nonparametric tests
636 nonparametric test subcommands
641 [ /STATISTICS=@{DESCRIPTIVES@} ]
643 [ /MISSING=@{ANALYSIS, LISTWISE@} @{INCLUDE, EXCLUDE@} ]
645 [ /METHOD=EXACT [ TIMER [(n)] ] ]
648 NPAR TESTS performs nonparametric tests.
649 Non parametric tests make very few assumptions about the distribution of the
651 One or more tests may be specified by using the corresponding subcommand.
652 If the /STATISTICS subcommand is also specified, then summary statistics are
653 produces for each variable that is the subject of any test.
655 Certain tests may take a long time to execute, if an exact figure is required.
656 Therefore, by default asymptotic approximations are used unless the
657 subcommand /METHOD=EXACT is specified.
658 Exact tests give more accurate results, but may take an unacceptably long
659 time to perform. If the TIMER keyword is used, it sets a maximum time,
660 after which the test will be abandoned, and a warning message printed.
661 The time, in minutes, should be specified in parentheses after the TIMER keyword.
662 If the TIMER keyword is given without this figure, then a default value of 5 minutes
667 * BINOMIAL:: Binomial Test
668 * CHISQUARE:: Chisquare Test
669 * WILCOXON:: Wilcoxon Signed Ranks Test
670 * SIGN:: The Sign Test
675 @subsection Binomial test
677 @cindex binomial test
680 [ /BINOMIAL[(p)]=var_list[(value1[, value2)] ] ]
683 The /BINOMIAL subcommand compares the observed distribution of a dichotomous
684 variable with that of a binomial distribution.
685 The variable @var{p} specifies the test proportion of the binomial
687 The default value of 0.5 is assumed if @var{p} is omitted.
689 If a single value appears after the variable list, then that value is
690 used as the threshold to partition the observed values. Values less
691 than or equal to the threshold value form the first category. Values
692 greater than the threshold form the second category.
694 If two values appear after the variable list, then they will be used
695 as the values which a variable must take to be in the respective
697 Cases for which a variable takes a value equal to neither of the specified
698 values, take no part in the test for that variable.
700 If no values appear, then the variable must assume dichotomous
702 If more than two distinct, non-missing values for a variable
703 under test are encountered then an error occurs.
705 If the test proportion is equal to 0.5, then a two tailed test is
706 reported. For any other test proportion, a one tailed test is
708 For one tailed tests, if the test proportion is less than
709 or equal to the observed proportion, then the significance of
710 observing the observed proportion or more is reported.
711 If the test proportion is more than the observed proportion, then the
712 significance of observing the observed proportion or less is reported.
713 That is to say, the test is always performed in the observed
716 PSPP uses a very precise approximation to the gamma function to
717 compute the binomial significance. Thus, exact results are reported
718 even for very large sample sizes.
723 @subsection Chisquare Test
725 @cindex chisquare test
729 [ /CHISQUARE=var_list[(lo,hi)] [/EXPECTED=@{EQUAL|f1, f2 @dots{} fn@}] ]
733 The /CHISQUARE subcommand produces a chi-square statistic for the differences
734 between the expected and observed frequencies of the categories of a variable.
735 Optionally, a range of values may appear after the variable list.
736 If a range is given, then non integer values are truncated, and values
737 outside the specified range are excluded from the analysis.
739 The /EXPECTED subcommand specifies the expected values of each
741 There must be exactly one non-zero expected value, for each observed
742 category, or the EQUAL keywork must be specified.
743 You may use the notation @var{n}*@var{f} to specify @var{n}
744 consecutive expected categories all taking a frequency of @var{f}.
745 The frequencies given are proportions, not absolute frequencies. The
746 sum of the frequencies need not be 1.
747 If no /EXPECTED subcommand is given, then then equal frequencies
751 @subsection Wilcoxon Matched Pairs Signed Ranks Test
752 @comment node-name, next, previous, up
754 @cindex wilcoxon matched pairs signed ranks test
757 [ /WILCOXON varlist [ WITH varlist [ (PAIRED) ]]]
760 The /WILCOXON subcommand tests for differences between medians of the
762 The test does not make any assumptions about the variances of the samples.
763 It does however assume that the distribution is symetrical.
765 If the @code{WITH} keyword is omitted, then tests for all
766 combinations of the listed variables are performed.
767 If the @code{WITH} keyword is given, and the @code{(PAIRED)} keyword
768 is also given, then the number of variables preceding @code{WITH}
769 must be the same as the number following it.
770 In this case, tests for each respective pair of variables are
772 If the @code{WITH} keyword is given, but the
773 @code{(PAIRED)} keyword is omitted, then tests for each combination
774 of variable preceding @code{WITH} against variable following
775 @code{WITH} are performed.
779 @subsection Sign Test
784 [ /SIGN varlist [ WITH varlist [ (PAIRED) ]]]
787 The /SIGN subcommand tests for differences between medians of the
789 The test does not make any assumptions about the
790 distribution of the data.
792 If the @code{WITH} keyword is omitted, then tests for all
793 combinations of the listed variables are performed.
794 If the @code{WITH} keyword is given, and the @code{(PAIRED)} keyword
795 is also given, then the number of variables preceding @code{WITH}
796 must be the same as the number following it.
797 In this case, tests for each respective pair of variables are
799 If the @code{WITH} keyword is given, but the
800 @code{(PAIRED)} keyword is omitted, then tests for each combination
801 of variable preceding @code{WITH} against variable following
802 @code{WITH} are performed.
805 @comment node-name, next, previous, up
812 /MISSING=@{ANALYSIS,LISTWISE@} @{EXCLUDE,INCLUDE@}
813 /CRITERIA=CIN(confidence)
821 (Independent Samples mode.)
822 GROUPS=var(value1 [, value2])
826 (Paired Samples mode.)
827 PAIRS=var_list [WITH var_list [(PAIRED)] ]
832 The @cmd{T-TEST} procedure outputs tables used in testing hypotheses about
834 It operates in one of three modes:
836 @item One Sample mode.
837 @item Independent Groups mode.
842 Each of these modes are described in more detail below.
843 There are two optional subcommands which are common to all modes.
845 The @cmd{/CRITERIA} subcommand tells PSPP the confidence interval used
846 in the tests. The default value is 0.95.
849 The @cmd{MISSING} subcommand determines the handling of missing
851 If INCLUDE is set, then user-missing values are included in the
852 calculations, but system-missing values are not.
853 If EXCLUDE is set, which is the default, user-missing
854 values are excluded as well as system-missing values.
857 If LISTWISE is set, then the entire case is excluded from analysis
858 whenever any variable specified in the @cmd{/VARIABLES}, @cmd{/PAIRS} or
859 @cmd{/GROUPS} subcommands contains a missing value.
860 If ANALYSIS is set, then missing values are excluded only in the analysis for
861 which they would be needed. This is the default.
865 * One Sample Mode:: Testing against a hypothesised mean
866 * Independent Samples Mode:: Testing two independent groups for equal mean
867 * Paired Samples Mode:: Testing two interdependent groups for equal mean
870 @node One Sample Mode
871 @subsection One Sample Mode
873 The @cmd{TESTVAL} subcommand invokes the One Sample mode.
874 This mode is used to test a population mean against a hypothesised
876 The value given to the @cmd{TESTVAL} subcommand is the value against
877 which you wish to test.
878 In this mode, you must also use the @cmd{/VARIABLES} subcommand to
879 tell PSPP which variables you wish to test.
881 @node Independent Samples Mode
882 @comment node-name, next, previous, up
883 @subsection Independent Samples Mode
885 The @cmd{GROUPS} subcommand invokes Independent Samples mode or
887 This mode is used to test whether two groups of values have the
888 same population mean.
889 In this mode, you must also use the @cmd{/VARIABLES} subcommand to
890 tell PSPP the dependent variables you wish to test.
892 The variable given in the @cmd{GROUPS} subcommand is the independent
893 variable which determines to which group the samples belong.
894 The values in parentheses are the specific values of the independent
895 variable for each group.
896 If the parentheses are omitted and no values are given, the default values
897 of 1.0 and 2.0 are assumed.
899 If the independent variable is numeric,
900 it is acceptable to specify only one value inside the parentheses.
901 If you do this, cases where the independent variable is
902 greater than or equal to this value belong to the first group, and cases
903 less than this value belong to the second group.
904 When using this form of the @cmd{GROUPS} subcommand, missing values in
905 the independent variable are excluded on a listwise basis, regardless
906 of whether @cmd{/MISSING=LISTWISE} was specified.
909 @node Paired Samples Mode
910 @comment node-name, next, previous, up
911 @subsection Paired Samples Mode
913 The @cmd{PAIRS} subcommand introduces Paired Samples mode.
914 Use this mode when repeated measures have been taken from the same
916 If the @code{WITH} keyword is omitted, then tables for all
917 combinations of variables given in the @cmd{PAIRS} subcommand are
919 If the @code{WITH} keyword is given, and the @code{(PAIRED)} keyword
920 is also given, then the number of variables preceding @code{WITH}
921 must be the same as the number following it.
922 In this case, tables for each respective pair of variables are
924 In the event that the @code{WITH} keyword is given, but the
925 @code{(PAIRED)} keyword is omitted, then tables for each combination
926 of variable preceding @code{WITH} against variable following
927 @code{WITH} are generated.
931 @comment node-name, next, previous, up
935 @cindex analysis of variance
940 [/VARIABLES = ] var_list BY var
941 /MISSING=@{ANALYSIS,LISTWISE@} @{EXCLUDE,INCLUDE@}
942 /CONTRAST= value1 [, value2] ... [,valueN]
943 /STATISTICS=@{DESCRIPTIVES,HOMOGENEITY@}
947 The @cmd{ONEWAY} procedure performs a one-way analysis of variance of
948 variables factored by a single independent variable.
949 It is used to compare the means of a population
950 divided into more than two groups.
952 The variables to be analysed should be given in the @code{VARIABLES}
954 The list of variables must be followed by the @code{BY} keyword and
955 the name of the independent (or factor) variable.
957 You can use the @code{STATISTICS} subcommand to tell PSPP to display
958 ancilliary information. The options accepted are:
961 Displays descriptive statistics about the groups factored by the independent
964 Displays the Levene test of Homogeneity of Variance for the
965 variables and their groups.
968 The @code{CONTRAST} subcommand is used when you anticipate certain
969 differences between the groups.
970 The subcommand must be followed by a list of numerals which are the
971 coefficients of the groups to be tested.
972 The number of coefficients must correspond to the number of distinct
973 groups (or values of the independent variable).
974 If the total sum of the coefficients are not zero, then PSPP will
975 display a warning, but will proceed with the analysis.
976 The @code{CONTRAST} subcommand may be given up to 10 times in order
977 to specify different contrast tests.
980 @comment node-name, next, previous, up
986 [VARIABLES=] var_list [@{A,D@}] [BY var_list]
987 /TIES=@{MEAN,LOW,HIGH,CONDENSE@}
988 /FRACTION=@{BLOM,TUKEY,VW,RANKIT@}
990 /MISSING=@{EXCLUDE,INCLUDE@}
992 /RANK [INTO var_list]
993 /NTILES(k) [INTO var_list]
994 /NORMAL [INTO var_list]
995 /PERCENT [INTO var_list]
996 /RFRACTION [INTO var_list]
997 /PROPORTION [INTO var_list]
999 /SAVAGE [INTO var_list]
1002 The @cmd{RANK} command ranks variables and stores the results into new
1005 The VARIABLES subcommand, which is mandatory, specifies one or
1006 more variables whose values are to be ranked.
1007 After each variable, @samp{A} or @samp{D} may appear, indicating that
1008 the variable is to be ranked in ascending or descending order.
1009 Ascending is the default.
1010 If a BY keyword appears, it should be followed by a list of variables
1011 which are to serve as group variables.
1012 In this case, the cases are gathered into groups, and ranks calculated
1015 The TIES subcommand specifies how tied values are to be treated. The
1016 default is to take the mean value of all the tied cases.
1018 The FRACTION subcommand specifies how proportional ranks are to be
1019 calculated. This only has any effect if NORMAL or PROPORTIONAL rank
1020 functions are requested.
1022 The PRINT subcommand may be used to specify that a summary of the rank
1023 variables created should appear in the output.
1025 The function subcommands are RANK, NTILES, NORMAL, PERCENT, RFRACTION,
1026 PROPORTION and SAVAGE. Any number of function subcommands may appear.
1027 If none are given, then the default is RANK.
1028 The NTILES subcommand must take an integer specifying the number of
1029 partitions into which values should be ranked.
1030 Each subcommand may be followed by the INTO keyword and a list of
1031 variables which are the variables to be created and receive the rank
1032 scores. There may be as many variables specified as there are
1033 variables named on the VARIABLES subcommand. If fewer are specified,
1034 then the variable names are automatically created.
1036 The MISSING subcommand determines how user missing values are to be
1037 treated. A setting of EXCLUDE means that variables whose values are
1038 user-missing are to be excluded from the rank scores. A setting of
1039 INCLUDE means they are to be included. The default is EXCLUDE.
1041 @include regression.texi
1045 @section RELIABILITY
1051 /SCALE (@var{name}) = @{var_list, ALL@}
1052 /MODEL=@{ALPHA, SPLIT[(N)]@}
1053 /SUMMARY=@{TOTAL,ALL@}
1054 /MISSING=@{EXCLUDE,INCLUDE@}
1057 @cindex Cronbach's Alpha
1058 The @cmd{RELIABILTY} command performs reliablity analysis on the data.
1060 The VARIABLES subcommand is required. It determines the set of variables
1061 upon which analysis is to be performed.
1063 The SCALE subcommand determines which variables reliability is to be
1064 calculated for. If it is omitted, then analysis for all variables named
1065 in the VARIABLES subcommand will be used.
1066 Optionally, the @var{name} parameter may be specified to set a string name
1069 The MODEL subcommand determines the type of analysis. If ALPHA is specified,
1070 then Cronbach's Alpha is calculated for the scale. If the model is SPLIT,
1071 then the variables are divided into 2 subsets. An optional parameter
1072 @var{N} may be given, to specify how many variables to be in the first subset.
1073 If @var{N} is omitted, then it defaults to one half of the variables in the
1074 scale, or one half minus one if there are an odd number of variables.
1075 The default model is ALPHA.
1077 By default, any cases with user missing, or system missing values for
1079 in the VARIABLES subcommand will be omitted from analysis.
1080 The MISSING subcommand determines whether user missing values are to
1081 be included or excluded in the analysis.
1083 The SUMMARY subcommand determines the type of summary analysis to be performed.
1084 Currently there is only one type: SUMMARY=TOTAL, which displays per-item
1085 analysis tested against the totals.
1093 @cindex Receiver Operating Characterstic
1094 @cindex Area under curve
1097 ROC @var{var_list} BY @var{state_var} (@var{state_value})
1098 /PLOT = @{ CURVE [(REFERENCE)], NONE @}
1099 /PRINT = [ SE ] [ COORDINATES ]
1100 /CRITERIA = [ CUTOFF(@{INCLUDE,EXCLUDE@}) ]
1101 [ TESTPOS (@{LARGE,SMALL@}) ]
1102 [ CI (@var{confidence}) ]
1103 [ DISTRIBUTION (@{FREE, NEGEXPO @}) ]
1104 /MISSING=@{EXCLUDE,INCLUDE@}
1108 The @cmd{ROC} command is used to plot the receiver operating characteristic curve
1109 of a dataset, and to estimate the area under the curve.
1110 This is useful for analysing the efficacy of a variable as a predictor of a state of nature.
1112 The mandatory @var{var_list} is the list of predictor variables.
1113 The variable @var{state_var} is the variable whose values represent the actual states,
1114 and @var{state_value} is the value of this variable which represents the positive state.
1116 The optional subcommand PLOT is used to determine if and how the ROC curve is drawn.
1117 The keyword CURVE means that the ROC curve should be drawn, and the optional keyword REFERENCE,
1118 which should be enclosed in parentheses, says that the diagonal reference line should be drawn.
1119 If the keyword NONE is given, then no ROC curve is drawn.
1120 By default, the curve is drawn with no reference line.
1122 The optional subcommand PRINT determines which additional tables should be printed.
1123 Two additional tables are available.
1124 The SE keyword says that standard error of the area under the curve should be printed as well as
1126 In addition, a p-value under the null hypothesis that the area under the curve equals 0.5 will be
1128 The COORDINATES keyword says that a table of coordinates of the ROC curve should be printed.
1130 The CRITERIA subcommand has four optional parameters:
1132 @item The TESTPOS parameter may be LARGE or SMALL.
1133 LARGE is the default, and says that larger values in the predictor variables are to be
1134 considered positive. SMALL indicates that smaller values should be considered positive.
1136 @item The CI parameter specifies the confidence interval that should be printed.
1137 It has no effect if the SE keyword in the PRINT subcommand has not been given.
1139 @item The DISTRIBUTION parameter determines the method to be used when estimating the area
1141 There are two possibilities, @i{viz}: FREE and NEGEXPO.
1142 The FREE method uses a non-parametric estimate, and the NEGEXPO method a bi-negative
1143 exponential distribution estimate.
1144 The NEGEXPO method should only be used when the number of positive actual states is
1145 equal to the number of negative actual states.
1146 The default is FREE.
1148 @item The CUTOFF parameter is for compatibility and is ignored.
1151 The MISSING subcommand determines whether user missing values are to
1152 be included or excluded in the analysis. The default behaviour is to
1154 Cases are excluded on a listwise basis; if any of the variables in @var{var_list}
1155 or if the variable @var{state_var} is missing, then the entire case will be