ANOVA and Post-hoc comparisons using group data

[Michael Kruger-2]

I have a researcher who has data that has been collected over an
extended period of time for about a dozen groups that he wants to
analyze in specific subsets of groups. The problem is that he has only
group data; means, std. errors nd n per group. Can SPSS 15 or 17 be used
to accurately analyze this type of data? If so, pleae point me in the
right direction. Tnanks in advance.

Michael Kruger

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Marta Garcia-Granero]

Michael Kruger wrote:
> I have a researcher who has data that has been collected over an
> extended period of time for about a dozen groups that he wants to
> analyze in specific subsets of groups. The problem is that he has only
> group data; means, std. errors nd n per group. Can SPSS 15 or 17 be used
> to accurately analyze this type of data? If so, pleae point me in the
> right direction. Tnanks in advance.

Hi Michael:

You can adapt the code wrote time ago for  that task. It uses a dataset
with means, SD & n

http://www.spsstools.net/Syntax/T-Test/ONEWAYwithSummaryData1.txt

HTH,
Marta García-Granero

--
For miscellaneous statistical stuff, visit:
http://gjyp.nl/marta/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Mirka]

Dear Michael,

why not use ONEWAY procedure with the matrix input? This procedure has a
lot of options for the post-hoc tests and has been in many previous
versions of SPSS.

Example:

matrix data variables= rowtype_   group  dep/ factor=group.
begin data.
MEAN        1       23.02
STDDEV     1       9.98
N                1       42
MEAN         2      22.18
STDDEV     2      14.38
N               2        540
Mean         3         23.01
STDDEV   3        13.00
N               3        62
end data.
execute.
ONEWAY
  dep BY group
   /MATRIX=IN(*)
  /STATISTICS DESCRIPTIVES HOMOGENEITY
  /MISSING ANALYSIS
  /POSTHOC = BONFERRONI ALPHA(.05).


Michael Kruger wrote:

> I have a researcher who has data that has been collected over an
> extended period of time for about a dozen groups that he wants to
> analyze in specific subsets of groups. The problem is that he has only
> group data; means, std. errors nd n per group. Can SPSS 15 or 17 be used
> to accurately analyze this type of data? If so, pleae point me in the
> right direction. Tnanks in advance.
>
> Michael Kruger
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD

--

Mirka Ondrack                     Tel:[phone][ex:33028]

Institute for Social Research     Fax:[phone]

York University, TEL 5064         email: [hidden email]
<mailto:[hidden email]>

Statistical Consulting Services:  http://www.isr.yorku.ca/scs/index.html

Online Appointment Scheduler:
    http://www.appointmentquest.com/provider/2000199121

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD