Re: PSPP-BUG: different output for K-S one-sample test

[John Darrington]

You have not run the same test.

On PSPP you ran
 NPAR TEST /KOLMOGOROV-SMIRNOV (NORMAL) =  age.

wheareas on SPSS you ran

 DATASET ACTIVATE DataSet1.
 EXAMINE VARIABLES=age
  /PLOT BOXPLOT STEMLEAF NPPLOT
  /COMPARE GROUPS
  /STATISTICS DESCRIPTIVES
  /CINTERVAL 95
  /MISSING LISTWISE
  /NOTOTAL.

The KS test performed by NPAR is not the same as that performed by EXAMINE,
One important difference is, like the text says, the EXAMINE one includes
the Lillefors correction.

J'

On Fri, Feb 05, 2021 at 10:50:21PM +0100, Andreas Hammer wrote:
     Hi,
     
     here the SPSS results.
     
     Regards,
     
     Andreas
     
     
     Am 05.02.2021 um 21:08 schrieb John Darrington:
     > It seems that somehow this report got overlooked.
     > 
     > You say that the results are different under SPSS.    Are you able to 
run the
     > same test under SPSS and post the results?   Then perhaps we can 
investigate
     > what might be causing the difference.
     > 
     > Regards,
     > 
     > John
     > 
     > On Fri, Feb 05, 2021 at 07:51:31PM +0100, Andreas Hammer wrote:
     >       Hello,
     >       is this problem solved?
     >       
https://lists.gnu.org/archive/html/bug-gnu-pspp/2020-03/msg00000.html
     >       I think the k-s results are still different from SPSS.
     >       I run pspp 1.4.2-g3990e1 and pspp-1.4.1-g79ad47 under win10.
     >       Thanks.for attention.
     >       Best regards
     >       Andreas Hammer
     > 
     > 
     >       NPAR TEST
     >          /KOLMOGOROV-SMIRNOV (NORMAL) =  age.
     >       Kolmogorov-Smirnov Test für eine Stichprobe
     >       ╭────────────────────────────────┬─────╮
     >       │                                │ age │
     >       ├────────────────────────────────┼─────┤
     >       │N                               │  234│
     >       │Normal Parameter Mittelwert     │43,90│
     >       │                 Std. Abweichung│10,40│
     >       │Größte Differenz Absolut        │  ,09│
     >       │                 Positiv        │  ,08│
     >       │                 Negativ        │ -,09│
     >       │Kolmogorov-Smirnov Z            │ 1,43│
     >       │Asymp. Sig. (2-seitig)          │ ,023│
     >       ╰────────────────────────────────┴─────╯
     > 
     >       _______________________________________________
     >       Bug-gnu-pspp mailing list
     >       Bug-gnu-pspp@gnu.org
     >       https://lists.gnu.org/mailman/listinfo/bug-gnu-pspp
     > 
     -- 
     Andreas Hammer M. A.
     Beratung  & Evaluation
     Eppinger Str. 62
     D-76684 Östringen
     
     Telefon  +49 72 59 - 92 57 29
     Mobil    +49 170 - 2 82 68 14
     Signal   +49 170 - 2 82 68 14<br>
     WhatsAPP +49 151 - 59 82 01 95<br>
     Telegram +49 151 - 59 82 01 95<br>
     
     http://www.andreas-hammer.eu/
     Twitter: www.twitter.com/AndreasHammer
     Xing: https://www.xing.com/profile/Andreas_Hammer28
     Linked.In: www.linkedin.com/in/andreas-hammer
     Terminanfragen auch über http://doodle.com/AndreasHammer
     
     Bankverbindung: IBAN DE87 6635 0036 0017 6394 54
     Steuer-Nr. 30361/32021 - Gerichtsstand Bruchsal