Fixed behaviour of oneway when presented with missing values

diff --git a/src/ChangeLog b/src/ChangeLog
index 3c7527c7c967bd16b14b6ebe34dc757a7d02c19b..11de18a8a6428a991559b46afb5fa9620f1cebd0 100644 (file)
--- a/src/ChangeLog
+++ b/src/ChangeLog
@@ -1,5 +1,7 @@
 Sat Oct 30 09:16:29 WST 2004 John Darrington <john@darrington.wattle.id.au>
 
+       * oneway.q   Fixed up the behaviour when given missing values
+
        * levene.c oneway.q Fixed a buglet with the levene statistic and
        incorporated the levene test into the oneway command.
 

diff --git a/src/oneway.q b/src/oneway.q
index 185a96303ab3f98c0973d188a95444d09e28f00c..075ce0a1f00b04513481e84530ff22c71ae73244 100644 (file)
--- a/src/oneway.q
+++ b/src/oneway.q
@@ -95,7 +95,7 @@ static void show_contrast_coeffs(void);
 static void show_contrast_tests(void);
 
 
-enum stat_table_t {STAT_DESC, STAT_HOMO};
+enum stat_table_t {STAT_DESC = 1, STAT_HOMO = 2};
 
 static enum stat_table_t stat_tables ;
 
@@ -903,12 +903,8 @@ calculate(const struct casefile *cf, void *cmd_)
                                 (hsh_hash_func *) hash_value,
                                 0,
                                 (void *) indep_var->width );
-
-
-
   precalc(cmd);
 
-
   for(r = casefile_get_reader (cf);
       casereader_read (r, &c) ;
       case_destroy (&c)) 
@@ -919,10 +915,30 @@ calculate(const struct casefile *cf, void *cmd_)
        dict_get_case_weight(default_dict,&c,&bad_weight_warn);
       
       const union value *indep_val = case_data (&c, indep_var->fv);
+
+      /* Deal with missing values */
+      if ( value_is_missing(indep_val,indep_var) )
+          continue;
+
+      /* Skip the entire case if /MISSING=LISTWISE is set */
+      if ( cmd->miss == ONEWAY_LISTWISE ) 
+       {
+         for(i = 0; i < n_vars ; ++i) 
+           {
+             const struct variable *v = vars[i];
+             const union value *val = case_data (&c, v->fv);
+
+             if (value_is_missing(val,v) )
+                 break;
+           }
+         if ( i != n_vars ) 
+             continue;
+
+       }
+      
          
       hsh_insert ( global_group_hash, (void *) indep_val );
 
-
       for ( i = 0 ; i < n_vars ; ++i ) 
        {
          const struct variable *v = vars[i];
@@ -986,7 +1002,9 @@ calculate(const struct casefile *cf, void *cmd_)
 
   
   if ( stat_tables & STAT_HOMO ) 
-    levene(cf, indep_var, n_vars, vars, LEV_LISTWISE, value_is_missing);
+    levene(cf, indep_var, n_vars, vars, 
+          (cmd->miss == ONEWAY_LISTWISE) ? LEV_LISTWISE : LEV_ANALYSIS ,
+          value_is_missing);
 
   ostensible_number_of_groups = hsh_count (global_group_hash);
 
@@ -1037,8 +1055,5 @@ postcalc (  struct cmd_oneway *cmd UNUSED )
 
       totals->se_mean = totals->std_dev / sqrt(totals->n);
        
-
-
-      
     }
 }

diff --git a/tests/Makefile.am b/tests/Makefile.am
index a988206717434bed2c433c1152fe986f4aea39ee..11904772f7aec94aab5c95d0d807c0039f1238fb 100644 (file)
--- a/tests/Makefile.am
+++ b/tests/Makefile.am
@@ -16,6 +16,7 @@ TESTS = \
        command/list.sh \
        command/loop.sh \
        command/oneway.sh \
+       command/oneway-missing.sh \
        command/print.sh \
        command/sample.sh \
        command/sort.sh \

diff --git a/tests/command/oneway-missing.sh b/tests/command/oneway-missing.sh
new file mode 100755 (executable)
index 0000000..76ab99e
--- /dev/null
+++ b/tests/command/oneway-missing.sh
@@ -0,0 +1,166 @@
+#!/bin/sh
+
+# This program tests that the ONEWAY anova command works OK when there is missing data
+
+TEMPDIR=/tmp/pspp-tst-$$
+
+here=`pwd`;
+
+# ensure that top_srcdir is absolute
+cd $top_srcdir; top_srcdir=`pwd`
+
+export STAT_CONFIG_PATH=$top_srcdir/config
+
+
+cleanup()
+{
+     rm -rf $TEMPDIR
+}
+
+
+fail()
+{
+    echo $activity
+    echo FAILED
+    cleanup;
+    exit 1;
+}
+
+
+no_result()
+{
+    echo $activity
+    echo NO RESULT;
+    cleanup;
+    exit 2;
+}
+
+pass()
+{
+    cleanup;
+    exit 0;
+}
+
+mkdir -p $TEMPDIR
+
+cd $TEMPDIR
+
+activity="create program"
+cat > $TEMPDIR/out.stat <<EOF
+DATA LIST LIST /v1 * v2 * dep * vn *.
+BEGIN DATA
+. .  1  4
+3 3  1  2
+2 2  1  2
+1 1  1  2
+1 1  1  2
+4 4  1  2
+5 5  2  2
+2 2  2  2
+4 4  2  2
+2 2  2  2
+3 3  2  2
+7 7  3  2
+4 4  3  2
+5 5  3  2
+3 3  3  2
+6 6  3  2
+END DATA
+
+ONEWAY
+       v1 v2 BY dep
+       /STATISTICS descriptives homogeneity
+       /MISSING ANALYSIS 
+       .
+EOF
+if [ $? -ne 0 ] ; then no_result ; fi
+
+
+activity="run program"
+$SUPERVISOR $here/../src/pspp -o raw-ascii $TEMPDIR/out.stat
+if [ $? -ne 0 ] ; then no_result ; fi
+
+activity="copy output"
+cp $TEMPDIR/pspp.list $TEMPDIR/pspp.list1
+if [ $? -ne 0 ] ; then no_result ; fi
+
+activity="create program 2"
+cat > $TEMPDIR/out.stat <<EOF
+DATA LIST LIST /v1 * v2 * dep * vn * .
+BEGIN DATA
+4 .  1  2 
+3 3  1  2
+2 2  1  2
+1 1  1  2
+1 1  1  2
+4 4  1  2
+5 5  2  2
+2 2  2  2
+4 4  2  2
+2 2  2  2
+3 3  2  2
+7 7  3  2
+4 4  3  2
+5 5  3  2
+3 3  3  2
+6 6  3  2
+END DATA
+
+ONEWAY
+       v1 v2 BY dep
+       /STATISTICS descriptives homogeneity
+       /MISSING LISTWISE
+       .
+EOF
+if [ $? -ne 0 ] ; then no_result ; fi
+
+activity="run program 2"
+$SUPERVISOR $here/../src/pspp -o raw-ascii $TEMPDIR/out.stat
+if [ $? -ne 0 ] ; then no_result ; fi
+
+activity="compare outputs"
+diff $TEMPDIR/pspp.list $TEMPDIR/pspp.list1
+if [ $? -ne 0 ] ; then fail ; fi
+
+# Now try a missing dependent variable
+# Everything depends upon it, so it should behave as if LISTWISE were set
+activity="create program 3"
+cat > $TEMPDIR/out.stat <<EOF
+DATA LIST LIST /v1 * v2 * dep * vn * .
+BEGIN DATA
+4 2  .  2 
+3 3  1  2
+2 2  1  2
+1 1  1  2
+1 1  1  2
+4 4  1  2
+5 5  2  2
+2 2  2  2
+4 4  2  2
+2 2  2  2
+3 3  2  2
+7 7  3  2
+4 4  3  2
+5 5  3  2
+3 3  3  2
+6 6  3  2
+END DATA
+
+ONEWAY
+       v1 v2 BY dep
+       /STATISTICS descriptives homogeneity
+       /MISSING ANALYSIS
+       .
+EOF
+if [ $? -ne 0 ] ; then no_result ; fi
+
+activity="run program 3"
+$SUPERVISOR $here/../src/pspp -o raw-ascii $TEMPDIR/out.stat
+if [ $? -ne 0 ] ; then no_result ; fi
+
+activity="compare outputs"
+diff $TEMPDIR/pspp.list $TEMPDIR/pspp.list1
+if [ $? -ne 0 ] ; then fail ; fi
+
+
+pass

diff --git a/tests/command/oneway.sh b/tests/command/oneway.sh
index f647e4e0bbbb8fd6b6269723ed99e91a2cc21a0c..e81d8032910a25b7223853e10561b97e4dda2be6 100755 (executable)
--- a/tests/command/oneway.sh
+++ b/tests/command/oneway.sh
@@ -94,14 +94,26 @@ diff $TEMPDIR/pspp.list - << EOF
 |BRAND   |F8.0  |
 +--------+------+
 
-2.1 ONEWAY.  Test of Homogeneity of Variances
+2.1 ONEWAY.  Descriptives
+#===============#========#==#====#==============#==========#=======================#=======#=======#
+#               |        #  |    |              |          |    95% Confidence     |       |       #
+#               |        #  |    |              |          +-----------+-----------+       |       #
+#               |        # N|Mean|Std. Deviation|Std. Error|Lower Bound|Upper Bound|Minimum|Maximum#
+#===============#========#==#====#==============#==========#===========#===========#=======#=======#
+#Breaking Strain|Aspeger # 5|2.20|          1.30|       .58|        .58|       3.82|   1.00|   4.00#
+#               |Bloggs  # 5|3.20|          1.30|       .58|       1.58|       4.82|   2.00|   5.00#
+#               |Charlies# 5|5.00|          1.58|       .71|       3.04|       6.96|   3.00|   7.00#
+#               |Total   #15|3.47|          1.77|       .46|       2.49|       4.45|   1.00|   7.00#
+#===============#========#==#====#==============#==========#===========#===========#=======#=======#
+
+2.2 ONEWAY.  Test of Homogeneity of Variances
 #===============#================#===#===#============#
 #               #Levene Statistic|df1|df2|Significance#
 #===============#================#===#===#============#
 #Breaking Strain#            .092|  2| 12|        .913#
 #===============#================#===#===#============#
 
-2.2 ONEWAY.  ANOVA
+2.3 ONEWAY.  ANOVA
 #==============================#==============#==#===========#=====#============#
 #                              #Sum of Squares|df|Mean Square|  F  |Significance#
 #===============#==============#==============#==#===========#=====#============#
@@ -110,7 +122,7 @@ diff $TEMPDIR/pspp.list - << EOF
 #               |Total         #         43.73|14|           |     |            #
 #===============#==============#==============#==#===========#=====#============#
 
-2.3 ONEWAY.  Contrast Coefficients
+2.4 ONEWAY.  Contrast Coefficients
 #==========#=======================#
 #          #      Manufacturer     #
 #          #-------+------+--------#
@@ -120,7 +132,7 @@ diff $TEMPDIR/pspp.list - << EOF
 #        |2#      0|    -1|       1#
 #========#=#=======#======#========#
 
-2.4 ONEWAY.  Contrast Tests
+2.5 ONEWAY.  Contrast Tests
 #===============================================#=================#==========#=====#=====#===============#
 #                                       Contrast#Value of Contrast|Std. Error|  t  |  df |Sig. (2-tailed)#
 #===============#======================#========#=================#==========#=====#=====#===============#