This manuscript describes the comprehensive validation work undertaken to develop the Recognizing Effective Special Education Teachers (RESET) observation system, which was designed to provide evaluations of special education teachers' ability to effectively implement evidence-based practices and to provide specific, actionable feedback to teachers on how to improve instruction. Following the guidance for developing effective educator evaluation systems, we employed the Evidence-Centered Design framework, articulated the claims and inferences to be made with RESET, and conducted a series of studies to collect evidence to evaluate its validity. Our efforts and results to date are described, and implications for practice and further research are discussed. [This paper will be published in "Educational Assessment."]