現今的電腦化適性測驗(Computerized adaptive testing,簡稱CAT)或電腦化適性診斷測驗,目前大都以選擇題方式施測,而選擇題本身即有猜測度存在,因此本研究將以廣義隱藏式馬可夫模型結合核平滑化無參數試題反應理論模式,兼顧猜測度方式進行蒙地卡羅模擬研究,並藉由模擬結果推論至順序理論(OT)與試題關聯結構分析法(IRS)之適用性。
It used experience rule generally in setting value for Ordering theory and Item Relational Structure (IRS). And for the computerized adaptive testing (CAT) or computerized adaptive diagnosis testing, it almost be tested by multiple choice method which with guessing-parameter. In this study, it was processed in Monte Carlo Method which using general hidden Markov model with behavior model of smooth none parameter examination question which combining guessing-parameter, and study the suitable for using in OT and IRS.