The objective of this study is to employ the quasi-experimental research method to explore third-grade students fraction learning situations. Two different ways of learning, "diagnostic teaching" and "programmed instruction", are applied to explore the differences between the learning performance and learning attitude on the concept of fractions. The participants of the research experiments consisted of 88 third-grade students (44 for diagnostic teaching and 44 for programmed instruction). After the experiment, ANCOVA analysis and t-test analysis on the learning outcomes and learning attitude before and after the test are conducted, and the results are as follows: First, the learning performances of the experimental group instructed by the "diagnostic teaching" approach are better than those of the control group instructed by "programmed instruction" approach. Second, these two digital learning approaches can improve the students' interest in learning and learning attitude.