[Rasch] a simple software related question

Sedat Sen sedatsen06 at gmail.com
Thu Jan 17 03:28:28 EST 2013


Hi everyone,

I was replicating the analyses of the examples in R.J De Ayala's IRT book.
I tried to use different software packages including multilog, bilog mg,
Mplus, IRTPRO, and R. When I tried to do 2PL analysis using the same data,
I ended up with different results in Mplus. First I tried bilogmg as the
author did and I got the same results but when I tried Mplus I ended up
with the same results but with opposite signs. Negative estimations in
bilogmg turned out to be positive in Mplus and positives turned out to be
negative(see the output below). It is just reverse.

Does any of you guys know why this happens?

OUTPUT FROM BILOGMG

ITEM      INTERCEPT    SLOPE    THRESHOLD   LOADING   ASYMPTOTE    CHISQ
  DF
              S.E.       S.E.       S.E.       S.E.       S.E.      (PROB)
 -------------------------------------------------------------------------------
 ITEM0001 |   *2.584*  |   1.226  |  -2.107  |   0.775  |   0.000  |
302.4   4.0
                |   0.044* |   0.047* |   0.056* |   0.030* |   0.000* |
(0.0000)
                |             |             |              |             |
            |
 ITEM0002 |   *0.995*  |   1.992  |  -0.499  |   0.894  |   0.000  |
339.7   4.0
                |   0.032* |   0.061* |   0.013* |   0.027* |   0.000* |
(0.0000)
                |             |             |              |             |
            |
 ITEM0003 |   *0.394*  |   1.551  |  -0.254  |   0.840  |   0.000  |
387.4   4.0
                |   0.022* |   0.041* |   0.014* |   0.022* |   0.000* |
(0.0000)
                |             |             |              |          |
     |
 ITEM0004 |  *-0.416*  |   1.544  |   0.270  |   0.839  |   0.000  |
240.4   4.0
                |   0.021* |   0.039* |   0.014* |   0.021* |   0.000* |
(0.0000)
                |              |             |            |              |
            |
 ITEM0005 | * -0.551*  |   0.983  |   0.560  |   0.701  |   0.000  |
461.2   4.0
                |   0.018* |   0.026* |   0.021* |   0.019* |   0.000* |
(0.0000)
 -------------------------------------------------------------------------------

OUTPUT FROM MPLUS

Thresholds
    U1$1              *-2.585*      0.038    -67.795      0.000
    U2$1              *-0.996*      0.030    -33.719      0.000
    U3$1              *-0.394*      0.021    -18.501      0.000
    U4$1              * 0.416*      0.022     19.266      0.000
    U5$1               *0.550*      0.018     30.164      0.000

Thanks,

-- 
*Sedat
*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailinglist.acer.edu.au/pipermail/rasch/attachments/20130116/fa958c89/attachment.html 


More information about the Rasch mailing list