Implementation of Linear Congruent Methods and Multiplication Random Numbers for Academic Potential Tests

Abstract

APT (Academic Potential Test) is a test that aims to measure a person's ability in the academic field in general. In the implementation of the APT exam, it is carried out in the admission of new students and its application online, using a website-based application, each prospective new student will be given a login account to take the APT exam simultaneously and at a predetermined time. While the process can be accessed anywhere with an internet network. The implementation of the APT exam does not always run smoothly or well, in fact almost every time the APT exam is carried out there are problems, problems that arise because the questions given do not have differences in workmanship which causes the APT exam results to be impure and accurate. To overcome the problems that continue to occur in the implementation of the APT exam, an algorithm or method is needed that can randomize the questions in the APT exam. In this study, the Linear Congruent (LCM) and Multiplicative Random Number Generator (Multiplicative RNG) methods are random methods that are applied to randomize the APT exam questions so that the APT exam question packages can have different question positions and between question packages and the results of the application of this method will be compared to measures how complex the randomization is for each method. By using the LCM model the level of complexity of the questions increases to 100% while by using the MRNG method the level of complexity of the questions increases to 50%.