The essence of developing test is to construct atest that have a desired quality by selecting theappropriate items, not minding the type of toolthat was used. There are two main test theories,such as classical test theory (CTT) and itemresponse theory (IRT). According to Greg(2009), classical test theory is a body of theoryand research regarding psychological testingthat predicts/explains the difficulty of questions,provides insight into reliability of assessmentscores, and helps in representing whatexaminees know and can do. The essential basisof classical test theory (CTT) is that manyquestions combine to produce a measurement(assessment score) representing what a test takerknows and can do. CTT works well for mostassessment applications for reason such as itsability to work with smaller sample size(example, 100 or less), and that it is relativelysimple to compute and understand the statistics(Greg, 2009). In classical test theory, statisticssuch as item difficulty index, itemdiscrimination index, indices of reliability, andvalidity which are used for interpreting testscores are sample dependent. This means thatclassical test theory has the limitation of sampledependency for estimating the test itemparameters of item difficulty and itemdiscrimination. Adedoyin and Adedoyin (2013)opined that classical test theory of itemparameter estimates using heterogeneoussamples generally result in higher estimates ofitem discrimination indices as measured bypoint-biserial correlation coefficients, whereasitem difficulty estimates rise and fall with highand low ability groups of examinees.