The Effect of Multiple True-False Test Scoring Method on the Accuracy of Estimating Ability and Item Difficulty Parameters Using Partial Credit Model

Abstract

The purpose of the study was to examine the effect of
the multiple true-false test scoring method on the accuracy of
estimating ability and difficulty parameters using Partial
Credit Model. To achieve this goal, a multiple true- false
mathematics test composed of (35) multiple true-false items
was developed and administered to (873) male and female 11th
graders of the literary stream and the information technology
stream in Irbid First Directorate of Education. The test was
scored by using four different methods that varied in the
degree of considering partial knowledge . By using BIGSTEPS
, estimates of individuals' abilities, items difficulty values,
standard errors ,and person and item fit statistics were
obtained for each scoring method. Data was re-analyzed after
excluding misfitting subjects and items. The findings of the
study indicated that the fourth method in which partial
knowledge was most considered is the most accurate method
for estimating ability and item difficulty parameters