Date of Award


Degree Type

Closed Dissertation

Degree Name

Doctor of Philosophy (PhD)


Learning Technologies Division

First Advisor

Jonathan Cohen

Second Advisor

Brendan Calandra

Third Advisor

Maggie Renken

Fourth Advisor

Teri Holbrook


Computational thinking (CT) is a complex problem-solving process that develops over an extended period of time. After Wing (2006) popularized CT as a concept, there has been growth in studies of it, with the majority taking place in computing courses. Although previous research has demonstrated the relationship between learner characteristics and programming success in higher education (Watson et al., 2014), a comprehensive approach to understand the relationships between learner characteristics and CT in computing courses in K-12 education is lacking. The aim of this dissertation was to address this gap by exploratory analysis to determine how a set of learner characteristics were related to a group of middle school students’ CT, and to determine which of these factors had the strongest association with participants’ CT in two computer science educational contexts. vii This research took place in two study sites in different districts in the U.S. In total, 314 students participated in this research. Students completed a CT quiz and a learner profile survey and developed digital artifacts in an app-building computing course. Artifact analysis was conducted to examine the CT practices in the artifacts. Correlational analysis followed by regression analysis was used to examine the relationships between student variables, including self-efficacy, interest, prior experience with creative computing, and goal orientation, and CT measures such as quiz score and CT practice, after controlling for gender, and grade level. The results of this study demonstrated that self-efficacy had a significant relationship with CT on both of the study sites. The regression analysis showed that none of the other learner characteristics explained significant amount of variation of CT. However, among the control variables, only gender had a significant correlation with CT practices profile; there were significantly more male than female students who demonstrated CT practices in their digital artifacts. Taken together, the findings of this study have provided evidence on which learner characteristics are related to CT for middle school-aged students. Instructional designers, educators, and researchers should consider these learner characteristics in their design in CT-infused, middle school computing courses.


File Upload Confirmation