A main question this dissertation addresses is: what variation in teaching and teacher training matter? This question is examined within a specific but important context: the scale-up of a technology-rich intervention focused on the algebra strand of 8th grade mathematics. I conducted a multi-level case study by gathering and analyzing data at all three levels of a train-the-trainer model of teacher professional development: from training of regional trainers, to teacher training, to classroom enactment. This case study was contextualized by a larger randomized experiment in the Scaling up SimCalc project. The larger project demonstrated the SimCalc intervention produced robust effects on student learning. Although treatment classrooms outperformed control classrooms, there was variation in student outcomes among teachers who used SimCalc. In the multi-level case studies, I sought to understand why two particular teachers had very different levels of student outcomes. This puzzle was unraveled using a mixed methodology by first searching for distinctive features of their enactments that may have influenced student outcomes, and then looking for connections between these features and the teacher-training workshop attended. Within this framework, the investigation provides arguments for these key findings: (1) Within the specific context of the SimCalc intervention, a wide variety of enactments may be acceptable and successful, provided (a) the main ideas are presented accurately, (b) students are given adequate time and (c) students are given a reasonable amount of autonomy with the materials. (2) Assuming a robust intervention, there may be unexpected benefits in allowing teachers to enact materials within a comfort-zone of teaching that he/she finds effective in his/her classroom. (3) The training workshops were successful in broad goals, but less successful in communicating other more pedagogically-based goals to all teachers. This dissertation is significant in that (1) it documents successful teaching practices in a prominent, successful scale-up experiment, (2) it investigates a complete "train-the-trainer" process, (3) it sheds light on the complex relationship between a teacher's mathematical knowledge for teaching (MKT) and the quality of enacted instruction, and (4) it provides practical insight to trainers of short-term workshops, most notably: be realistic about what training can and cannot accomplish. [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.]