Preferred Name

Nikole Gregg

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Date of Graduation

Spring 2018

Document Type


Degree Name

Master of Arts (MA)


Department of Graduate Psychology


John Hathcoat


Written communication is a skill necessary for not only the success of undergraduate students, but for post-graduates in the workplace. Furthermore, according to employers the writing skills of post-graduates tend to be below expectations. Therefore, the assessment of such skills within higher education is in high demand. Written communication assessments tend to be administered in one of two conditions: 1) course embedded and 2) a low-stakes, non-embedded condition. The current study investigated possible construct-irrelevant variance in writing assessment scores by using data from a mid-sized public university in the Mid-Atlantic region of the United States. Specifically, 157 student products were scored using the Association of American Colleges and Universities’ Written Communication rubric by Multi-State Collaborative trained raters. A final sample size of 57 student products were in the non-embedded assessment condition and 107 student products were in the embedded assessment condition. Differential item functioning analyses were conducted using a Rasch Rating Scale model and an Ordinal Regression wherein Verbal SAT was used an external criterion of ability. Said differently, this study investigated whether students of the same proficiency had different probabilities of receiving particular written communication scores. After controlling for motivation, the results provide evidence of possible differential item functioning for Content Development as well as Genre and Disciplinary Conventions. Students of the same ability tend to obtain higher written communication scores in the non-embedded assessment condition. These results raise concerns about the presence of construct-irrelevant variance aside from motivation. Future research should investigate faculty feedback, allotted time, and task structure as possible sources of construct-irrelevant variance when using low-stakes, non-embedded assessments of written communication.



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.