Abstract: The global numbers are stuck. Despite two decades of “women in tech” campaigns, women’s representation in computer science has barely moved in most countries. At TED University in Ankara, Turkey, our CS graduating cohorts are now 38% women, nearly double the global average. We have tracked what changed that number and what did not. Context matters. We are IEEE Region 8, in a country where institutional budgets, family structures, and gender dynamics look nothing like Silicon Valley or Cambridge. What works here tends to travel, because it was never the expensive version to begin with. Over ten years we tracked enrollment, completion, employment, graduate program continuation, and early-career leadership outcomes by gender across 465 graduates. The popular interventions moved nothing. Awareness campaigns, summer camps, and token scholarships showed no measurable effect on our retention or outcome numbers. Three unglamorous mechanisms moved a lot: redesigning project grading rubrics, restructuring mentorship as peer-led networks instead of top-down assignments, and integrating women students into active research groups in their second year, not their fourth. This session shares (1) the measurement framework we used; (2) the three interventions that actually moved the numbers; (3) the popular initiatives that did not and why; and (4) a transferable template for other institutions across cultural and resource contexts, especially those operating under real budget and cultural constraints. For community builders, educators, industry leaders, and policymakers who want to move past motivation into what the data says actually works.
