Assemblers and compilers provide feedback to a programmer in the form of error messages. These error messages become input to the debugging model of the programmer. For the programmer to fix an error, they should first locate the error in the program, understand what is causing that error, and finally resolve that error. Error messages play an important role in all three stages of fixing of errors. This thesis studies the effects of error messages in the context of teaching programming. Given an error message, this work investigates how it effects student’s way of 1) understanding the error, and 2) fixing the error. As part of the study, three error message types were developed – Default, Link and Example, to better understand the effects of error messages. The Default type provides an assembler-centric single line error message, the Link type provides a program-centric detailed error description with a hyperlink for more information, and the Example type provides a program centric detailed error description with a relevant example. All these error message types were developed for assembly language programming. A think aloud programming exercise was conducted as part of the study to capture the student programmer’s knowledge model. Different codes were developed to analyze the data collected as part of think aloud exercise. After transcribing, coding, and analyzing the data, it was found that the Link type of error message helped to fix the error in less time and with fewer steps. Among the three types, the Link type of error message also resulted in a significantly higher ratio of correct to incorrect steps taken by the programmer to fix the error.

Web-based learning resources have been criticized as being developed with minimal consideration as to the effectiveness of the design principles or guidelines used to create them. Extraneous material is oftentimes present and necessary for learners to engage in effective learning with multimedia learning material. Signaling learners towards important information between images and corresponding text has been shown to be an effective method for providing learners a way to quickly find information between the two parts of the learning material. However, not all signaling methods are equally effective in all applications. This study investigates a novel signaling method, using spatial isolation of text, as a way to signal learners in a web-based format compared to a traditional highlighting method and a non-signaled control group. Improved learning performance was observed for knowledge retention using text isolation as the signaling method, but no other significant effects were observed between the other conditions. Additionally, transfer of knowledge across all conditions showed no significant differences either. While minimal support for the effectiveness of isolated text signaling was demonstrated, the statistical means trend across all post-test knowledge assessments suggest that further evaluation of the novel signaling method is justified.

One of the primary objective in a computer science related course is for students to be able to write programs implementing the concepts covered in that course. In educational psychology, however, learning gains are more commonly measured using recall or problem solving questions. While these types of questions are relevant to computer science exams, they do not necessarily reflect a student’s ability to apply concepts by writing an original program to solve a novel problem.

This thesis investigates the effectiveness of including questions within instructional multimedia content to improve student performance on a related programming assignment. Similar techniques have proven effective in educational psychology research using other measures. The objective of this thesis is to apply educational techniques used in other domains to an experiment with real world measures of students in a computer science course. The findings of this paper demonstrate that the techniques used were promising in improving student performance on a programming assignment.

Safe headway learning plays a core role in driving education. Traditional safe headway education just use the oral and literal methods to educate drivers the concept of safe headway time, while with the limitation of combining drivers subject and situational domains for drivers to learn. This study investigated that whether using ego-moving metaphor to embody driver's self-awareness can help to solve this problem. This study used multiple treatments (ego-moving and time-moving instruction of safe time headway) and controls with pretest experimental design to investigate the embody self-awareness effect in a car-following task. Drivers (N=40) were asked to follow a lead car at a 2-seconds safe time headway. Results found that using embodied-based instructions in safe headway learning can help to improve driver's headway time accuracy and performance stability in the car-following task, which supports the hypothesis that using embodied-based instructions help to facilitate safe headway learning. However, there are still some issues needed to be solved using embodied-based instructions for the drivers' safe headway education. This study serves as a new method for the safe headway education while providing empirical evidence for the embodied theories and their applications.

Intelligence analysts’ work has become progressively complex due to increasing security threats and data availability. In order to study “big” data exploration within the intelligence domain the intelligence analyst task was abstracted and replicated in a laboratory (controlled environment). Participants used a computer interface and movie database to determine the opening weekend gross movie earnings of three pre-selected movies. Data consisted of Twitter tweets and predictive models. These data were displayed in various formats such as graphs, charts, and text. Participants used these data to make their predictions. It was expected that teams (a team is a group with members who have different specialties and who work interdependently) would outperform individuals and groups. That is, teams would be significantly better at predicting “Opening Weekend Gross” than individuals or groups. Results indicated that teams outperformed individuals and groups in the first prediction, under performed in the second prediction, and performed better than individuals in the third prediction (but not better than groups). Insights and future directions are discussed.

This study investigated the ability to relate a test taker’s non-verbal cues during online assessments to probable cheating incidents. Specifically, this study focused on the role of time delay, head pose and affective state for detection of cheating incidences in a lab-based online testing session. The analysis of a test taker’s non-verbal cues indicated that time delay, the variation of a student’s head pose relative to the computer screen and confusion had significantly statistical relation to cheating behaviors. Additionally, time delay, head pose relative to the computer screen, confusion, and the interaction term of confusion and time delay were predictors in a support vector machine of cheating prediction with an average accuracy of 70.7%. The current algorithm could automatically flag suspicious student behavior for proctors in large scale online courses during remotely administered exams.

Observational tutoring has been found to be an effective method for teaching a variety of subjects by reusing dialogue from previous successful tutoring sessions. While it has been shown content can be learned through observational tutoring it has yet to been examined if a secondary behavior such as goal-setting can be influenced. The present study investigated if observing virtual humans engaging in a tutoring session on rotational kinematics with embedded positive goal oriented dialogue would increase knowledge of the material and perpetuate a shift an observer's goal-orientation from performance avoidance goal orientation (PAVGO) to learning goal orientation (LGO). Learning gains were observed in pre to post test knowledge retention tests. Significant changes from pretest to posttest occurred across conditions for LGO. Additionally, significant changes from PAVGO pretest to posttest were observed in the control condition however PAVGO did not significantly change in the experimental condition.

Multimodal presentations have been found to facilitate learning, however, may be a disadvantage for low spatial ability students if they require spatial visualization. This disadvantage stems from their limited capacity to spatially visualize and retain information from both text and diagrams for integration. Similarly, working memory capacity (WMC) likely plays a key role in a learner's ability to retain information presented to them via both modalities. The present study investigated whether or not the act of self-explaining helps resolve deficits in learning caused by individual differences in spatial ability, working memory capacity, and prior knowledge when learning with text, or text and diagrams. No interactions were found, but prior knowledge consistently predicted performance on like posttests. The author presents methodological and theoretical explanations as to the null results of the present study.