USING A PATIENT ADMINISTRATIVE SYSTEM - A PERFORMANCE EVALUATION AFTER END-USER TRAINING

We first review Swedish experimental research with relevance to computer application program training and Scandinavian research on usability aspects of computer program development projects. Then we report an empirical study of eight end-users' performance as they interacted with a newly installed patient administrative program. After extensive training on the program, organized by their employers, the users were asked to carry out four routine work tasks. The results show that none of the users could carry out all four tasks without errors or assistance from the experimenter It was also found that experimenter assistance in the form of direct help with specific content accounted for more than two fifths of all assistance provided. The program's Help function was found to be of no help at all to the users. Finally, most of the users' difficulties fell into just a few categories. The program has not been user-tested, but our results argue for the benefits of user-testing both for program development and for the planning of user training.