SUMMARY
Good instructional materials are arranged systematically and they enable students to learn independently, up to date, easily digestible and designed in accordance with the applicable curriculum. This research aimed to determine the process and results of the development of integrated statistical instructional materials EPUB-based statistical literacy properly (valid, practical, and effective) and to determine whether there were significant differences between the pretest-posttest results. The type of research used was Research and Development (R&D) with the ADDIE type development model. The subjects in this research were 53 students of the mathematics education study program at Universitas Palangkaraya (UPR) and 41 students of the PGSD study program at Universitas Mushammadiyah Palangkaraya (UMPR). Data analysis used T-test and N-Gain test. The research results obtained were as follows: (1) The instructional materials developed contained statistical literacy and multimedia based, namely EPUB (2) The average validity obtained from the validator's assessment was 89.42% very valid of 89.42% (3) The average percentage of student responses to EPUB-based statistical literacy integrated instructional materials was 89.82%. Overall, instructional materials were in the very practical category. (4) student learning completeness reached 80.76%, there were 30 of 31 UPR students completed and student learning completeness reached 78, 80%, there were 22 UMPR students completed and it can be concluded that the use of EPUB was effective. (5) Furthermore, a significant difference can be seen from the UPR average of the experiment class posttest was 84.23 while the posttest control class was 61.23. It means that the post-test results of the experiment class were higher than the post-test results of the control class with a difference of 23 and the result of the N-gain score for the experiment class is 56.3% while the control class is only 4.24%. Meanwhile, a significant difference can also be seen from the average UMPR of the posttest experiment class was 78.14 while the posttest control class was 63.00. It means that the post-test results of the experiment class were higher than the post-test results of the control class with a difference of 15.14 and the results of the N-gain score for the experiment class are 56% while the control class was only 10%.