pspp-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Output format for p-values


From: John Darrington
Subject: Re: Output format for p-values
Date: Sun, 27 Apr 2014 09:38:22 +0200
User-agent: Mutt/1.5.21 (2010-09-15)

Ouch!

Yes seems that oneway contrasts don't properly handle negative T-values.


I'll fix this asap, and check for any similar problems.

In the mean time, you can work around the issue, by choosing the sign of your
coefficients such that the contrast comes out positive.  In your example, 
instead of writing /CONTRAST= 1 -1 0 write /CONTRAST= -1 1 0

Thanks for reporting this.

J'

On Sat, Apr 26, 2014 at 07:42:08PM -0700, Douglas Bonett wrote:
     There is a bug in the computation of the p-value for the unequal variance
     option in Contrasts of One-way.  The output is shown below for a 3-group
     design with n = 3 per group and scores of 1,2,3,4,5,6,7,8,9.  The reported
     p-value is 1.98.
     
     
     
     ONEWAY /VARIABLES= score BY group
     
            /CONTRAST= 1 -1 0.
     
     
     
     ANOVA
     
     #====================#==============#==#===========#=====#====#
     
     #                    #Sum of Squares|df|Mean Square|  F  |Sig.#
     
     #=====#==============#==============#==#===========#=====#====#
     
     #score|Between Groups#         54.00| 2|      27.00|27.00| .00#
     
     #     |Within Groups #          6.00| 6|       1.00|     |    #
     
     #     |Total         #         60.00| 8|           |     |    #
     
     #=====#==============#==============#==#===========#=====#====#
     
     
     
     Contrast Coefficients
     
     #==========#==========================#
     
     #          #           group          #
     
     #          #--------+--------+--------#
     
     #          #    1.00|    2.00|    3.00#
     
     #========#=#========#========#========#
     
     #Contrast|1#       1|      -1|       0#
     
     #========#=#========#========#========#
     
     
     
     Contrast Tests
     
     
#===============================#=============#========#=====#====#===========#
     
     #                       Contrast#   Value of  |  Std.  |  t  | df |  Sig.
     (2- #
     
     #                               #   Contrast  |  Error |     |    |  
tailed)
     #
     
     
#=====#================#========#=============#========#=====#====#===========#
     
     #score|Assume equal    |    1   #        -3.00|     .82| 3.67|   6|
     .01#
     
     #     |variances       |        #             |        |     |    |
     #
     
     #     |Does not assume |    1   #        -3.00|     .82|-3.67|4.00|
     1.98#
     
     #     |equal           |        #             |        |     |    |
     #
     
#=====#================#========#=============#========#=====#====#===========#
     

-- 
PGP Public key ID: 1024D/2DE827B3 
fingerprint = 8797 A26D 0854 2EAB 0285  A290 8A67 719C 2DE8 27B3
See http://sks-keyservers.net or any PGP keyserver for public key.

Attachment: signature.asc
Description: Digital signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]