Author Topic: Saving Mantel Haenszel output  (Read 913 times)

DMarengo

  • Newbie
  • *
  • Posts: 5
    • View Profile
Saving Mantel Haenszel output
« on: July 13, 2012, 07:28:03 PM »
Hi!
I'm trying to obtain Mantel-Haenszel estimates but it seems I am stuck with some technical problem. When I run the Mantel-Haenszel procedure (by Conquest's GUI), results are shown in the output window as expected, but they get partly cleared out for some items (part of the results can't be accessed by scrolling the output window). This is the code I use:

datafile <<F:\directory\camp08.dat;
format pid 1-8, gender 9, areag 10, responses 11-32(a1);
group areag gender;
labels <<F:\directory\camp08.lab;
score (0,1) (0,1);
Model item;
Estimate !method=jml;
mh! estimates=mle, group=gender, reference=1;
Show >> F:\directory\camp08.shw;
Itanal >> F:\directory\camp08.itn;

As a tentative work-around I tried to save the results on file but I get no data written. This is the (probably aberrant) code I use:

mh! estimates=mle, group=gender, reference=1 >> F:\directory\camp08mh.dat ;

Is there any solution for this?

Best regards
Davide

Xiaoxun

  • Administrator
  • Newbie
  • *****
  • Posts: 2
    • View Profile
    • Email
Re: Saving Mantel Haenszel output
« Reply #1 on: July 15, 2012, 11:22:06 PM »
Hi, Davide:

In order to identify the issue you have experienced properly, could you please send your data and control files to us?

Please send the data and control file to xiaoxun.sun@acer.edu.au.

Regards
Xiaoxun

ConQuest

  • Global Moderator
  • Newbie
  • *****
  • Posts: 24
    • View Profile
    • Email
Re: Saving Mantel Haenszel output
« Reply #2 on: July 24, 2012, 01:29:40 AM »
David,

note that you can use

set buffersize=a big number;

to make sure you don't lose material from the output window.  Xiaoxun is looking into the export issue.

Ray

DMarengo

  • Newbie
  • *
  • Posts: 5
    • View Profile
Re: Saving Mantel Haenszel output
« Reply #3 on: July 25, 2012, 11:56:14 PM »
Hi Ray
this is exactly the solution I was looking for.
Thanks a lot.
Davide