differences between version 14 and 15 ?

[Søren Bech]

Dear Users

I have just compared the exact same analysis of a data set using version
14
and 15 (see below) - the computing time using version 14 was 7,33 s and
with version 15 it was 3 min 27,75 s - AND the analysis using version 15
could not compute the degrees of freedom - thus the ANOVA was not
calculated using version 15 (could this be the reason for the long
computation time ?)

Could anyone help me explain what is going on here ?

Regards

Soren Bech

Notes for the analysis using version 14:

Notes
Output Created          05-JAN-2007 12:49:58
Comments
Input   Data    P:\Perceptionforsoeg\Visuel perception\Forsøg\BS3
perception\Kiggepanel_desriptiv_analyse\Results\results_total.sav
        Active Dataset  DataSet1
        Filter  attribute = 1 (FILTER)
        Weight  <none>
        Split File      <none>
        N of Rows in Working Data File  210
Missing Value Handling  Definition of Missing   User-defined missing
values are treated as missing.
        Cases Used      Statistics are based on all cases with valid data
for all variables in the model.
Syntax          UNIANOVA
  rating  BY Picture TV SubjID repeat
  /RANDOM = SubjID repeat
  /METHOD = SSTYPE(3)
  /INTERCEPT = INCLUDE
  /EMMEANS = TABLES(Picture*TV)
  /CRITERIA = ALPHA(.05)
  /DESIGN = TV Picture SubjID repeat Picture*TV SubjID*TV TV*repeat
Picture
 *SubjID Picture*repeat SubjID*repeat Picture*SubjID*TV Picture*TV*repeat
  SubjID*TV*repeat Picture*SubjID*repeat .

Resources       Elapsed Time    0:00:07,33



Notes for the analysis using version 15:

Notes
Output Created          26-JAN-2007 23:22:55
Comments
Input   Data    P:\Perceptionforsoeg\Visuel perception\Forsøg\BS3
perception\Kiggepanel_desriptiv_analyse\Results\results_total.sav
        Active Dataset  DataSet1
        Filter  attribute = 1 (FILTER)
        Weight  <none>
        Split File      <none>
        N of Rows in Working Data File  210
Missing Value Handling  Definition of Missing   User-defined missing
values are treated as missing.
        Cases Used      Statistics are based on all cases with valid data
for all variables in the model.
Syntax          UNIANOVA
  rating  BY TV Picture SubjID repeat
  /RANDOM = SubjID repeat
  /METHOD = SSTYPE(3)
  /INTERCEPT = INCLUDE
  /EMMEANS = TABLES(Picture*TV)
  /PRINT = HOMOGENEITY
  /PLOT = SPREADLEVEL
  /CRITERIA = ALPHA(.05)
  /DESIGN = TV Picture SubjID repeat Picture*TV SubjID*TV TV*repeat
Picture
 *SubjID Picture*repeat SubjID*repeat Picture*SubjID*TV Picture*TV*repeat
  SubjID*TV*repeat Picture*SubjID*repeat .

Resources       Processor Time  0:03:24,76
        Elapsed Time    0:03:27,75