Conspiracy Theory: An Empirical Study of Cheating in a Continuous Testing Environment

The effect of item characteristics on examinee recall of test items is examined in the paper. Study One evaluated item length, difficulty, guessing, discrimination, serial position, misfit, format, memoriability, clarity, familiarity, interest, and examinee ability as predictors of item recall. Item recall was coded either using a full item memory criteria or partial item memory criteria. Results suggest that item guessing, item difficulty, item length, item type, and examinee ability are related to full item memory, whereas item guessing, serial position, item type, misfit, and examinee ability are related to partial item memory. Various memory mechanisms are discussed as possible explanations for these findings. Study Two was a simulation of a multi-stage test to provide insight as to the number of conspirators required to compromise a test, and to assess the differences between a simulation where item theft is modeled after the parameter estimated in Study One and a simulation where each item has an equal probability of being stolen. Results suggest that a small group of conspirators can significantly inflate test scores for lower ability examinees, and that there is little differences in estimated score inflation between the two simulations. Implications of the findings are discussed.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Psychology

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2015

Author dc:creator

• Chuah, Siang Chee

Contributors dc:contributor

• Drasgow, Fritz

• Psychology, Industrial

Language dc:language

eng

Identifier

(MiAaPQ)AAI3198953

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/82084