[Rasch] Simulated eigenvalues greater than empirical data analysis?

Mike Linacre rmt at rasch.org
Tue Mar 13 15:15:09 EST 2012


Everyone:

Imogene and I have discovered that the earlier implausible numbers were due 
to incompatibilities in data file formats.

The variance results for a simulated dataset analyzed by Winsteps are now 
reasonable:

      Table of STANDARDIZED RESIDUAL variance (in Eigenvalue units)
                                                  -- Empirical --    Modeled
Total raw variance in observations     =        151.9 100.0%         100.0%
   Raw variance explained by measures   =         65.9  43.4%          44.0%
     Raw variance explained by persons  =         54.6  35.9%          36.4%
     Raw Variance explained by items    =         11.3   7.5%           7.6%
   Raw unexplained variance (total)     =         86.0  56.6% 100.0%   56.0%
     Unexplned variance in 1st contrast =          1.7   1.1%   2.0%
     Unexplned variance in 2nd contrast =          1.6   1.1%   1.9%

Mike L.

At 3/11/2012, Imogene wrote:
>As you can see there is now very little variance explained by measures and 
>the strength of the 1st and 2nd contrasts is high! Is this a plausible result?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailinglist.acer.edu.au/pipermail/rasch/attachments/20120313/e9feec23/attachment.html 


More information about the Rasch mailing list