The dependent variables of this research model are learning score and learning satisfaction. The instruments of the learning score include the comprehensive test sheets designed for the two course units. Each test sheet was designed by two senior subject teachers with at least 10 years of teaching experience and refined by university professors. The test sheet for Chinese poem “The River Mechant’s Wife” has 10 multiple-choice tasks and five exegesis tasks. The total score is 100. The test sheet of the number system transformation has 15 multiple-choice tasks and the total score is also 100. The final score for each test sheet was the average marks of scorings by two teachers and the scores were immune to the experiment conditions.
Since the learning satisfaction dependent variable is to measure the learner’s satisfaction with the benefits of the media design in helping learning, we adapt the usefulness instrument developed and tested by Davis (1989) to measure this construct. The usefulness instrument has been used and validated in a wide range of studies and it has been shown to have high reliability and validity. A pretest for the instrument was administered to 32 students to ensure the validity of this measure. After the pretest, five items rated on a scale varying from one being “strongly disagree” to seven being “strongly agree” (i.e., 7-point Likert scale) used to measure the learning satisfaction are shown in Table 1. The learning satisfaction construct satisfies the criteria of reliability with the coefficient of reliability (Cronbach α) at 0.77, exceeding the standard high water mark of 0.7.