eduzhai > Physical Sciences > Physics Sciences >

Computational Precision of the Power Function for Conditional Tests of Assumptions of the Rasch Model

  • Save

... pages left unread,continue reading

Document pages: 12 pages

Abstract: Draxlerand Zessin [1] derived the power function for a class of conditional tests ofassumptions of a psychometric model known as the Rasch model and suggested anMCMC approach developed by Verhelst [2] for thenumerical approximation of the power of the tests. In this contribution, theprecision of the Verhelst approach is investigated and compared with an exactsampling procedure proposed by Miller and Harrison [3] for which the discrete probability distribution to be sampled from isexactly known. Results show no substantial differences between the twonumerical procedures and quite accurate power computations. Regarding thequestion of computing time the Verhelst approach will have to be consideredmuch more efficient.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×