Estimating the Students&rsquo; Mathematical Ability Using the Newton-Raphson Algorithm Under the Item Response Theory and the Classical Response Theory: A Comparative Study

Sisteta  Kamdon,; Nurijam Hanna  Mohammad,; Ummar  Sallil,

Estimating the Students’ Mathematical Ability Using the Newton-Raphson Algorithm Under the Item Response Theory and the Classical Response Theory: A Comparative Study

Sisteta Kamdon | Nurijam Hanna Mohammad | Ummar Sallil

Discipline: Education

Abstract:

This paper utilized the Newton-Raphson Algorithm in calibrating the Item-Response Theory (IRT) to estimate the mathematical abilities of the senior high school students of the Mindanao State University - Tawi-Tawi Campus. For comparison, we also employed the Classical Test Theory (CTT) for the same purpose. It was found out that more of the ability estimates of the concerned students were at the average or above average level in the IRT than in the CTT. Nevertheless, more passers were noted under the IRT than those under the CTT. Moreover, there was a significant difference between the transmuted grades under the IRT and the CTT among the STEM groups while no significant difference was noted among the HUMMS groups. Furthermore, ability estimates of all groups under the IRT and CTT showed significant relationship. The results appeared to support the strength of the IRT, particularly the Newton-Raphson Algorithm, over the CTT in estimating students’ mathematical abilities. Hence, the Newton-Raphson method of the IRT can be a better estimation method that teachers/educators can use in establishing the abilities/performances of their students.