Date of Graduation


Document Type


Degree Name

Doctor of Education (Ed.D.)


School of Education


Learning and Instruction


Learning & Instruction EdD

First Advisor

Patricia Busk

Second Advisor

Robert Burns

Third Advisor

Brad Washington


The purpose of this study was to investigate how training and professional development effected university-level instructors’ perceived usefulness, perceived ease of use, behavioral intent to use, perception of self-efficacy, and frequency of use of audio-, video-, and speech-to-text-recognition-based technologies associated with the feedback and assessment process in college-level teaching. Except for usefulness, each dependent variable was divided into two based on whether the item was multimedia or not: (a) use of technology with multimedia and (b) use of technology without multimedia. The convenience sample included 52 university-level instructors who had enrolled in either the Canvas® Essentials (a basics course) or Canvas® Feedback and Assessment (an advanced course) training. The advanced training focused on how to use audio, video, and speech-to-text-recognition features of the learning management system to provide feedback to students.

The study commenced in August of 2015 and concluded in April of 2016. A pretest questionnaire was administered prior to each Canvas® training class, and instruction began immediately thereafter lasting 2 hours per class session. The posttest was administered 4 weeks after the training class. Twenty-six instructors represented the treatment group, and 26 represented the comparison group.

Means measuring intent, self-efficacy, and usefulness indicated either agreement or strong agreement for both treatment and comparison groups; however, the variables of intent and usefulness resulted in little-to-no change in means from pretest to posttest. For the variable of self-efficacy, both groups’ means increased from pretest to posttest. Higher means indicated stronger agreement with the construct.

The variable of self-efficacy also resulted in a statistically significant change from pretest to posttest for both groups. Treatment-group participants’ mean went up .41 of a point from pretest to posttest and had a strong effect (ES = .86), indicating that they were somewhat skilled at posttest. The comparison-group means also reflected increased agreement in self-efficacy, participants on average reported that they were between not very skilled and somewhat skilled at using Canvas® LMS at pretest. At posttest, the comparison group’s mean indicated that they were above the somewhat skilled choice on the rating scale.

For both groups, the mean measuring the construct of intent (media) decreased slightly from pretest to posttest, and the results were not statistically significant. Means for ease were higher at posttest for both groups; the independent-samples t test resulted in statistical significance for the comparison group with a moderately strong effect size.

The variables of ease (media) and frequency resulted in higher means at posttest for both groups and were statistically significant across four paired-samples t tests. Moderately strong effect sizes were present in the variables of ease and frequency among comparison-group participants. The variables of self-efficacy (media) and frequency (media) resulted in means that signified the lowest levels of agreement among the nine dependent variables. For treatment-group participants, self-efficacy (media) resulted in a large statistically significant effect, and mean increased from pretest to posttest.

Self-efficacy (media) for comparison-group participants increased in agreement from pretest to posttest, and the results were statistically significant with a large effect. Frequency (media) decreased in agreement from pretest to posttest for treatment-group participants, and comparison-group participants’ mean increased slightly from pretest to posttest. No independent-samples t tests resulted in statistically significant findings for the variable of frequency (media).

This research addressed a gap in the literature and illustrated that instructors are willing participate in research. The study participants gained new skills to support their own day-to-day work teaching courses and grading assignments, which was a benefit to them, the university, and the research community.