The Problem. This study focuses on the instructional technique of correspondence instruction, the learner, the setting, and the problem of course noncompletion. The purpose of this study was to develop a model of completion of correspondence courses which identifies factors related to progress toward course completion. The objectives were to estimate the model and identify factors associated with course completion.

Methodology. The causal model for this study was adapted from Bean's synthetic model of student attrition from institutions of higher education. Bean's model uses four categories of variables, (background, environmental, organizational, attitudinal), and a variable of intent to drop out. Another variable, date of submission of the first lesson was added to Bean's model for the purposes of this study.

Data to test the model were obtained from a questionnaire, transcripts, and progress records of students enrolled in Independent Study Program courses and declaring a nursing major. Factor analysis was used to identify similar items on the questionnaire to be summed as composite variables. Six factors and 20 single-item variables were measured. Reliability was determined for the factors and coefficient alpha averaged .82. Content and convergent validity were also established for the questionnaire. Multiple regression and path analysis were used to analyze the data.

The Findings. Eleven variables in the model estimated in this study accounted for 44% of variance in course progress. Students making progress intended completing the course in three months, submitted the first lesson within 40 days, had high SAT scores and GPA's, completed (and not dropped) other correspondence courses, perceived family support, but not employer support, had high goals for completing the program and obtaining the BSN, lived closer to the instructor, and entered the nursing program with college preparation (as a registered nurse).

Conclusions. The model has utility for understanding and explaining progress toward course completion as well as the distance education setting of correspondence instruction. The intent variable explains most variance and can be measured while students are enrolled in the course. The lesson submission variable is theoretically justified.