GENERATOR OF VARIANTS OF TESTS USING THE SAME QUESTIONS Cover Image

GENERATOR OF VARIANTS OF TESTS USING THE SAME QUESTIONS
GENERATOR OF VARIANTS OF TESTS USING THE SAME QUESTIONS

Author(s): Ovidiu DOMŞA, Nicolae Bold
Subject(s): Education, Educational Psychology, Sociology of Education
Published by: Carol I National Defence University Publishing House
Keywords: test; questions; evaluation; algorithm; random; permutations;

Summary/Abstract: To obtain the expected results in the academic environment, various mechanisms for presenting the information in a form as much as attractive were made. This thing implied the development of complex means which involves the usage of technical and software support as varied as possible. Another important component related to learning validation is the assessment component. In this moment, a variety of assessment mechanisms exists, from the classical ones which use detailed requirements and answers to those which use templates with simple or more complex questions and answers. In the assessment process it is important to use tests with the same content for all the students, in order to obtain a correct hierarchy of them. However, for a better security regarding the answers given by the student, it is useful to elaborate more test variants with the same questions, but with answers numbered in a different way. The generator presented in this paper uses a probabilistic algorithm for solving this request. The generator accomplishes two roles: firstly, it does the generation of a random permutation of the answers choices and, secondly, it generates a random permutation of the questions which forms the test. A specified number of tests will be generated using these permutations. The application which uses the generator contains three files: the first one with the question statements, the second one with the answers choices and the third with the right answer. Using the generator for the permutations of answers choices and of questions, variants of the same test are obtained.

  • Issue Year: 12/2016
  • Issue No: 02
  • Page Range: 586-592
  • Page Count: 7
  • Language: English