Commenced in January 2007
Paper Count: 30172
Minimizing Examinee Collusion with a Latin- Square Treatment Structure
Authors: M. H. Omar
Abstract:Cheating on standardized tests has been a major concern as it potentially minimizes measurement precision. One major way to reduce cheating by collusion is to administer multiple forms of a test. Even with this approach, potential collusion is still quite large. A Latin-square treatment structure for distributing multiple forms is proposed to further reduce the colluding potential. An index to measure the extent of colluding potential is also proposed. Finally, with a simple algorithm, the various Latin-squares were explored to find the best structure to keep the colluding potential to a minimum.
Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1056474Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 845
 G. J. Cizek, Cheating on Tests, How to Do It, Detect It, and Prevent It. Mahwah, New Jersey: Lawrence Erlbaum Associates, 1999.
 M. Golub-Smith, "A Study of the Effects of Item Option Rearrangement on the Listening Comprehension Section of the Test of English as a Foreign Language," Educational Testing Service, Princeton, New Jersey, TOEFL Res. Rep. 24 (RR-87-17), Aug 1987.
 A. Ercole, K. D. Whittlestone, D. G. Melvin and J. Rashbass, "Collusion detection in multiple choice examinations," Medical Education, vol. 36, no. 2, pp. 166 - 172, Feb. 2002.
 L. R. Nelson, "Using Selected Indices to Monitor Cheating on Multiplechoice Exams,- Journal of Educational Research and Measurement, Volume 4, 2006.
 S. Martirosyan, "Perfect Hash Families, Identifiable Parent Property Codes and Covering Arrays," Ph.D. dissertation, Dept. Mathematics., Universit┬¿at Duisburg-Essen., Armenia, 2003.
 G. A. Milliken and D. A. Johnson, Analysis of Messy Data, Volume I: Designed Experiments. New York: Van Nostrand Reinhold, 1992.