Date:         Mon, 28 Jun 1999 11:01:40 -0500
Reply-To:     Tim Dunsworth <Tim.Dunsworth@METROSTATE.EDU>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Tim Dunsworth <Tim.Dunsworth@METROSTATE.EDU>
Subject:      Re: K-Wallace
Content-Type: text/plain; charset=US-ASCII

You need to restructure your data in something like the following format:

sample mesh catch
1          2.5     2.1
1          3.0     4.8
2          2.5     0.6
2          3.0     6.1
3          2.5     3.3
...

Then the catch column will be the dependent variable and the mesh column will be the independent variable (hopefully the sample variable just represents random replicates and not another iv that needs to be taken in to account, as that would give you an incomplete experimental design with no replication that I don't think could be modelled successfully). If the data set is not too large, this restructuring can be done manually in the data editor, but otherwise someone else on the list could perhaps offer a syntax fragment that will do the job - ? HTH.

>>> Teresa Thorpe <thorpet@WILMINGTON.NET> 06/28 7:53 AM >>>
I have just purchased SPSS (grad version) and would like to run a
Kruskall-Wallace test.  I have 7 samples which are catch rates for
different gillnet mesh sizes and want to determine the significance of
changes in catch rates for each gillnet mesh size.  Unfortunately, the ref.
book which accompanied the software doesn't have detailed information on
data file set-up.

I therefore have 7 columns, each is a different mesh size with a range of
catch rates for each (n=unequal see below).  I have no idea how to set up
the data file to run a K-Wallace (or ANOVA etc.) as it asks for the column
with the independent variable ? I clearly don't have it set up like that.
It is the same for ANOVA, although I can run ANOVA's in excel no problem
due to the set-up of the package.  I need to run K-Wallace on a few data
sets, however, as the variances are unlikely to be homogenous.

2.5" mesh      3" mesh     etc....
2.1                4.8                       (data in rows are catch rates)
0.6                6.1
3.3                etc.
etc.

Thankyou for your time, I hate to bother the list with such a basic
question.

Sincerely,
Teresa Thorpe
-----------------------
Center for Marine Science Research
7205 Wrightsville Avenue
Wilmington, NC. 28403.
Tel: 910-7902292
Fax: 910-2568856
thorpet@wilmington.net