Students in the Science, Technology, Engineering, and Mathematics (STEM) fields are confronted with multiple external representations (MERs) in their learning materials. The ability to learn from and communicate with these MERs requires not only that students comprehend each representation individually but also that students recognize how the representations are related to one another. Past research has indicated that this integration of MERs is not automatic, and students need instructional support to master the learning and integration of MERs. Several researchers have investigated learning environments that included pre-integrated MERs or interactive capability, but results have been mixed in terms of the effectiveness of these learning environments on enhancing learning. Other researchers have indicated that students need support to guide their cognitive and metacognitive processes while learning with and integrating MERs, so they can engage in deeper learning. Thus, this study aims to investigate the effects of three instructional interventions, self-explanation, metacognitive, and the combined self-explanation and metacognitive instruction on students' learning outcomes in statistics, a domain with extensive use of MERs. The designed interventions supported students' learning with MERs by presenting a rationale, modeling with exemplary use of learning strategies suggested by the intervention, providing opportunities for practicing with materials containing MERs, and applying what they learned about the intervention while studying the statistics materials. This study applied a one-way four-group design, which included: (1) the control group (C); (2) the self-explanation instruction group (SE); (3) metacognitive instruction group (MI); and (4) the combined self-explanation and metacognitive instruction group (SEMI). The control group did not receive any intervention but had the opportunity to practice typing their thoughts while learning. The SE group was informed about why self-explanation can enhance learning, and practiced typing self-explanations by responding to prompts while learning with materials containing MERs. The MI group was informed about why the awareness and integration of MERs is important for learning, and practiced typing their thoughts while learning with materials containing MERs. The SEMI group was informed about why self-explanation can enhance learning, why the awareness and integration of MERs are important for learning, and practiced typing self-explanations by responding to prompts that solicit attention to and integration of MERs. The dependent variable was students' learning outcomes on introductory statistics concepts measured by a posttest that included multiple-choice questions requiring students to integrate either the same type or different types of representations in the question stem and corresponding choices respectively. A total of 120 undergraduate students participated in this study. The findings revealed that there were no significant differences on the learning outcomes between any one of the intervention groups and the control group. Also, there were no significant differences on the learning outcomes between any two intervention groups. It is possible that the lack of intervention effect resulted from ineffective instruction or prompts provided to the intervention groups. Another possible explanation is the requirement to type thoughts for the control group engaged learners in self-explanations and/or MER integration, which helped the control group perform equally well as the intervention groups on learning outcomes. The descriptive analysis indicated that there was a ceiling effect that the majority of participants in this study received near maximum scores, which could fail to reveal the true variability in scores and therefore the true effects of the interventions. Suggestions for future research and improvement in instructional design are provided. The emerging hypotheses regarding the effects of interventions and mere requirement of typing thoughts on learning outcomes are worth further investigation for refining the design of the intervention. In addition, this study contributes to the area of researching MER integration by developing a reliable statistics test that requires integrating MERs in the question stem and the corresponding choices in order to score. This format and development of this test can serve as a preliminary model of measurement development for future studies related to MER integration. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]