Experimental! Not recommended. This implements Linacre's (1989) XCON algorithm as XMLE "Exclusory Maximum Likelihood Estimation".

 

The reason that JMLE is statistically inconsistent under some conditions, and noticeably estimation-biased for short tests or small samples, is that it includes the possibility of extreme scores in the estimation space, but cannot actually estimate them. The XMLE algorithm essentially removes the possibility of extreme response vectors from the estimation space. This makes XMLE consistent, and much less biased than JMLE. In fact it is even less biased than CMLE for small samples, this is because CMLE only eliminates the possibility of extreme person response vectors, not the possibility of extreme item response vectors.

 

Considerations with XMLE=YES include:

(1) Anchoring values changes the XMLE probabilities. Consequently, measures from, say, a Table 20 score table do not match measures from the estimation run. Consequently, it may be necessary to estimate item calibrations with XMLE=YES. Then anchor the items and perform XMLE=NO.

(2) Items and persons with extreme (zero, minimum possible and perfect, maximum possible) scores are deleted from the analysis.

(3) For particular data structures, measures for finite scores may not be calculable.

 

Selecting XMLE=YES, automatically makes STBIAS=NO and PAIRED=NO, because XMLE is a more powerful bias correction technique..

 

Example: Produce XMLE estimates, to compare with JMLE estimates, and so investigate the size of the JMLE estimation bias.

XMLE=YES