One way that examinees can gain an unfair advantage on a test is by having prior access to the test questions and their answers, known as preknowledge. Determining which examinees had preknowledge can be a difficult task. Sometimes, the compromised test content that examinees use to get preknowledge has mistakes in the answer key. Examinees who had preknowledge can be identified by determining whether they used this flawed answer key. This research consisted of three papers aimed at helping testing programs detect examinees who used a flawed answer key. The first paper developed three methods for detecting examinees who used a flawed answer key. These methods were applied to a real data set with a flawed answer key for which 37 of the 65 answers were incorrect. One requirement for these three methods was that the flawed answer key had to be known. The second paper studied the problem of estimating an unknown flawed answer key. Four methods of estimating the unknown flawed key were developed and applied to real and simulated data. Two of the methods had promising results. The methods of estimating an unknown flawed answer key required comparing examinees’ response patterns, which was a time-consuming process. In the third paper, OpenMP and OpenACC were used to parallelize this process, which allowed for larger data sets to be analyzed in less time.
| Identifer | oai:union.ndltd.org:UTAHS/oai:digitalcommons.usu.edu:etd-8027 |
| Date | 01 May 2018 |
| Creators | Scott, Marcus W. |
| Publisher | DigitalCommons@USU |
| Source Sets | Utah State University |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | All Graduate Theses and Dissertations |
| Rights | Copyright for this work is held by the author. Transmission or reproduction of materials protected by copyright beyond that allowed by fair use requires the written permission of the copyright owners. Works not in the public domain cannot be commercially exploited without permission of the copyright owner. Responsibility for any use rests exclusively with the user. For more information contact digitalcommons@usu.edu. |
Page generated in 0.002 seconds