This two-article dissertation examined the impact of enhanced educational dialogue, in terms of periodic email feedback on course progress and an invitation to participate in a discussion board, on student achievement and course satisfaction in an introductory statistics course offered in an independent study setting. Participants in the study were students enrolled in the year-long online course. They were randomly assigned to different types and levels of educational dialogue and their completion status, final exam scores, average quiz scores, and course satisfaction ratings were compared after controlling for the following covariates of interest: age, gender, high school GPA, Math ACT, learner autonomy, attitude on the usefulness of statistics, and confidence in learning statistics. The different types and levels of educational dialogue used in this study were: email reminders only, discussion board only, email and discussion board, and no email or discussion board. Successful completion of introductory statistics courses in online learning environments can be predicted by student's attitude toward statistics and learner autonomy, in addition to the conventional measures of mathematics aptitude (ACT Math score) and effort as measured by High School GPA; however, there is a scarcity of psychometrically sound and brief measures of these constructs. The first article developed and validated the following scales as measures of attitude toward statistics and learner autonomy: perceived value of statistics (4 items), confidence in learning statistics (4 items), and learner autonomy (3 items). These abbreviated scales are shown to have content and discriminant validity. They can be used by statistics education researchers with confidence. The second article used MANCOVA and logistic regression to analyze the data collected from the randomized controlled experiment. The MANCOVA results show that students who have higher confidence in learning statistics have higher final exam scores and higher course satisfaction at the 5% level of significance. In addition, students assigned to the email group have the highest average quiz scores. Logistic regression results show that older students and those who have high confidence in learning statistics are more likely to complete the course. Overall, the completion rate for this study is significantly higher than the previous sections of the course. One of the implications of this study is that basic course progress feedback to students with minimal teacher-student interaction may have a beneficial impact on student achievement in online courses.



College and Department

David O. McKay School of Education; Instructional Psychology and Technology



Date Submitted


Document Type



instructional dialogue, course progress feedback, learner autonomy, course satisfaction, student achievement



Included in

Education Commons