In this dissertation, I present readers with three studies about the ways in which science, technology, engineering, and mathematics (STEM) research experiences impact undergraduate learning, perspectives, and academic/career activities. In the first two studies I focus on this topic by investigating the ways in which undergraduates were impacted by completing a STEM internship. The first study is unique in focusing on an understudied population: community college students interested in STEM fields. Both the first and second studies make a novel contribution to the STEM education literature by investigating U.S. Department of Energy (DOE) national laboratories as learning environments. These institutions are largely missing from the wealth of academic literature about undergraduate education in STEM disciplines. In the third study I define teaching and mentoring in STEM research experiences, which are terms that are often used, but rarely defined by scholars. In the third study I apply these definitions and a new instrument to determine which teaching and mentoring practices graduate students used when working with undergraduates on STEM research projects, and produce a set of practices that both scholars and practitioners can use in the future. Study 1 focuses on the perspectives of individuals who were taking STEM coursework while enrolled in community colleges. The majority of undergraduates who enroll in community college and/or begin their studies in STEM to obtain a degree do not meet this goal. Previous studies have shown that participation in technical and research internships can increase undergraduate academic achievement, graduation rates, confidence, and STEM persistence. However, very little is known about the benefits of these activities a) for community college students, b) when hosted by DOE national laboratories, and c) beyond the first few years after the internship. I applied the Social Cognitive Career Theory (SCCT) to investigate alumni perspectives about how the Community College Internship (CCI) at a specific DOE national laboratory -- Lawrence Berkeley National Laboratory (LBNL) -- impacted their academic/career activities. Specifically, I collected survey and interview data from 43 "CCI alumni," most of whom majored in civil engineering, mechanical engineering, or chemistry as undergraduates. Analysis of this data revealed that CCI alumni had low confidence and expectations of success in STEM as community college students, and were negatively impacted by the stigma associated with community colleges. Participation in CCI increased their professional networks, expectations of success, and STEM skills, identity, and self-efficacy/confidence. Hispanic/Latinx alumni recalled the positive impact of mentors who prioritized personal connections, and women valued "warm" social environments. These findings highlight program components and mentoring practices that had long-lasting effects on these individuals. Additionally, I propose several additions to the SCCT model, to better reflect the supports and barriers to STEM persistence for community college students. Future studies could build on this work to expand what is currently known about the impact of STEM educational opportunities at DOE national laboratories for community college students. Study 2 is closely related to Study 1 because of its focus on DOE national laboratories as learning environments, which is a topic with very little representation in academic literature. For students attending baccalaureate granting institutions, it is well-established that participation in research experiences or internships hosted by colleges and universities can support retention in STEM degree programs. However, limited research has been published about these opportunities for community college students and/or hosted by the DOE national laboratories. Data was collected from individuals who participated in the Community College Internship (CCI) and Science Undergraduate Laboratory Internship (SULI) programs between 2009 and 2016, at Lawrence Berkeley National Laboratory (LBNL). The CCI and SULI programs are part of a suite of programs hosted at DOE national laboratories designed to expose students and recent graduates to career opportunities at these institutions. Of the CCI alumni, 94% transferred to a baccalaureate granting institution, 90% graduated with a STEM bachelor's degree, and 88% are on a STEM career pathway. Based on what is known about graduation and transfer of community college students in the U.S., CCI alumni transferred and graduated with bachelor's degrees at higher rates than expected. Of the SULI alumni, 91% graduated with a STEM bachelor's degree, and 71% are on a STEM career pathway. These findings suggest that -- as compared to students attending baccalaureate granting institutions -- community college students who engage in STEM professional development activities are likely to persist in STEM careers at similar rates. Additionally, participation in STEM professional development activities may increase the likelihood that community college students complete their academic degrees in STEM disciplines. Shifting the focus "away" from the perspectives and activities of undergraduates, I investigate the topic of teaching and mentoring in undergraduate research experiences (UREs) in Study 3. It is common practice for undergraduates in STEM degree programs to participate in discipline-specific research experiences to gain new skills and knowledge, and to explore STEM careers. Similarly, many graduate students are expected to collaborate with undergraduates on a research project, which requires them to employ a combination of teaching and mentoring practices. Studies have shown that practices can vary dramatically between individuals, and these differences impact the overall experience for undergraduates (positively or negatively). However, the details provided in previous studies about STEM research experiences are insufficient to determine which teaching or mentoring practices are being used in a particular learning environment, by whom, and how often. Many scholars have described benefits of understanding how teaching and mentoring activities differ from each other, which supports student learning and well-being, communication between research team members (including undergraduates), and the evaluation of teaching/mentoring quality in STEM research experiences. Study 3 describes the new "Berkeley Undergraduate Research Evaluation Tools Teaching and Mentoring (BURET-TaM)" instrument, which I developed collaboratively with research team members to identify the teaching and mentoring practices used in STEM research experiences. Additionally, I provide definitions for teaching and mentoring for UREs relevant to this new instrument that are informed by the literature and can be used by educational researchers, students, and practitioners alike. I applied the "BURET-TaM" instrument to written reflections from and interviews with 46 graduate students working with undergraduate researchers in faculty-led research teams at the university and/or at the nearby DOE national laboratory, and generated a list of teaching and mentoring practices used by this group. My findings suggest that a) teaching and mentoring practices are often intertwined in this context, b) teaching scientific concepts and processes can support undergraduates to learn how new scientific knowledge is created, and c) research environments can impact student learning, well-being, and success. In the future, departments or research teams could use this new instrument to implement training sessions or materials to support scientists and professionals in learning how to teach/mentor undergraduate researchers, or to improve their skills in these areas. [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.]