| Abstract: | This investigation replicates Kember's (1995) Open Learning Courses for Adults: a model of student progress (Englewood Cliffs, NJ, Education Technology)] model of student progress, using students enrolled on four business courses at the Open University of the United Kingdom. Kember's model identified four key constructs: social integration, academic integration, external attribution, and academic incompatibility. Kember built these constructs, together with background characteristics, into a causal model of student progress and then tested it using path analysis. He concluded that the model was robust, accounting for 80% of the variance in adult student persistence. However, the empirical findings from the present study showed little internal consistency in the sub-scales for the key constructs in Kember's model. Furthermore, few of the causal relationships achieved statistical significance. These results suggest that Kember's path model did not fit the data derived from the present sample. While Kember's recommendations for reducing student dropout have intuitive appeal, their empirical foundations are questionable. |
