To assess educational effect in an adult students' course, three experiments were conducted with a rigid questionnaire related in job performance in which adult students participated in an educational course and their bosses evaluated the adult students' job performance before and after the course (before one week and after one and four weeks). The task of the first group (students and their bosses) in Experiment 1 was to enumerate criteria for evaluation of job performance, and 12 items were selected. In Experiment 2, the second group rated importance of the 12 items in evaluation of job performance, and the results revealed no difference in the rank order rating between the adult students and the bosses. The third group in Experiment 3-1 evaluated the adult student's job performance three times (before one week, and after one and four weeks). The second task in Experiment 3-2 was identical to Experiment 3-1 except that there were no classes. The results in Experiment 3-1 indicated significantly the time lag effect with a specific job-related item that the adult students evaluated advancement of job performance after one week, while their bosses evaluated subordinate's advancement after four weeks. In contrast, the time lag effect was not observed in Experiment 3-2. It is concluded that the evaluation based on the adult students' job performance revealed differences between the adult students and their bosses. These results suggest that self-evaluation in the adult students and others-evaluation in their bosses differ according to confirmatory data sampling and processing.