Background/Context: The last few decades of research on teacher professional learning (PL) show clear evidence that PL programs can shift instructional practice and accelerate student learning. Yet many programs continue to be ineffective, raising the question: what are features of effective PL design? Prior work attempting to answer this question is almost exclusively correlational and has been increasingly cast in doubt (Hill & Papay, 2022a, 2022b; Sims & Fletcher-Wood, 2021). These gaps in knowledge create challenges for district leaders and PL designers, who are routinely tasked with making myriad decisions about PL design. In this study, we begin to address these challenges by using a randomized controlled trial (RCT) to compare multiple versions of two design features, teacher agency over PL content and the format for follow-up. This study was co-designed with Montgomery County Public Schools' (MCPS) to help the district make informed decisions about the design of their district-wide PL programming. One question that MCPS, and other districts, are wrestling with is how much choice teachers have over PL content. Theory and descriptive evidence suggest that providing teachers with choice over their learning can increase engagement and motivation (Knowles et al., 2012; Merriam, 2001; Patrick et al., 2021). However, whether increased teacher choice makes PL more effective is unclear. The benefits of increased teacher motivation may not translate to improved instruction if teachers' choices are poorly aligned with their learning needs. Furthermore, providing greater teacher choice may mean sacrificing the benefits of more coherent and collaborative PL focused on common developmental needs (Patrick, 2022). Another question MCPS wrestles with is how to follow-up with teachers to support their implementation of PL content. Embedding follow-up opportunities within existing collaborative structures (e.g. professional learning communities) may be an efficient way to motivate and hold teachers accountable to implementing new practices (Guskey, 1995; Kim, et al., 2017; Lynch et al, 2019). It may also create greater coherence and connection between district-wide professional development and teachers' daily activities. On the other hand, these interactions may use up limited collaboration time and feel disjointed for teachers, especially if teachers have chosen different PL content to focus on. Independent reflection outside of existing structures may therefore offer a less disruptive and potentially equally effective option (Franek et al., 2001). Research Questions: In this study we conduct an RCT comparing a PL design that provides teachers with greater choice over the content of the PL with a design that provides teachers with less choice, instead giving more choice to school leaders. We also compare PL that includes a structure for teachers to check-in with peers about how they are implementing their learning with a design that includes a structure for individual reflection. Our 2x2 design will allow us to answer the following research questions: (1) To what extent do teacher choices about PL focus areas differ from the choices that school leaders make?; (2) Does providing teachers with greater choice about the content of their PL lead to greater teacher engagement and uptake of target instructional practices?; (3) Does embedding structures for peer reflection lead to greater engagement in PL and greater uptake of target instructional practices than structures for individual reflection?; and (4) Does the effectiveness of social reflection vary depending on teacher choice and vice versa? The fourth research question will provide insight into the interaction between teacher agency and follow-up format. For example, we could hypothesize that the effects of peer reflection will be larger when teachers have less choice because teachers are focused on similar goals for improvement. Alternatively, we could hypothesize that peer reflection will have positive effects on teacher motivation regardless of the level of choice provided to teachers. Setting and Population: In Fall 2024, all MCPS teachers are required to participate in district-wide professional learning focused on meeting the needs of multilingual learners, students receiving special education services, and Black and Latino/a learners. All teachers are required to first complete online learning modules asynchronously. After completing the modules, all teachers participate in informal observations and debrief meetings with their school leader to receive targeted feedback on their implementation of the online module content. Intervention and Research Design: We examine the impact of teacher choice and peer reflection using a 2x2 RCT where each of the 207 MCPS schools are assigned to one of the following groups: (1) Expanded teacher choice and individual reflection; (2) Limited teacher choice and individual reflection; (3) Expanded teacher choice and peer reflection; and 4. Limited teacher choice and peer reflection. In the expanded teacher choice condition, teachers first choose one of four topics to focus on in the online modules and then further choose a specific instructional strategy aligned to the selected topic area. In the limited teacher choice condition, school leaders select the focal topic and then teachers choose a specific instructional strategy aligned to the selected topic area. In the individual reflection condition, teachers receive an email in November prompting them to reflect individually on their implementation of their chosen instructional strategy. In the peer reflection condition, teachers will discuss implementation during a PLC meeting. Data: In addition to district administrative data, including classroom observation data and student assessment data, we leverage teacher survey data. Teachers complete a pre-survey before completing the modules, a post-survey immediately after completing the modules, and a final survey three months later. Data from the pre-survey provide baseline information about how teachers value choice and peer interactions. Data from the two post-surveys will serve as proximal outcomes, measuring teachers' sense of agency, engagement with PL, and motivation to implement what they've learned. Classroom observation data and student assessment data will serve as distal outcome measures. We also employ meta-data from the online modules to understand teachers' choices and assess treatment compliance. Analysis, Findings and Conclusion: Implementation and data collection is complete, and we are waiting on data transfer. We expect to have results prior to the conference in September. Our findings will offer practical guidance for PL designers in other districts who face similar questions about teacher choice and follow-up opportunities. These results will also provide broader learnings about how districts can balance district-wide goals and constraints with teachers' individual needs.