STTR Phase I: Spiral Assessment Loop for Rubrics

This Small Business Technology Transfer (STTR) Phase 1 project seeks to improve educational assessment rubrics. Accreditation requires adequate assessment rubrics to substantiate expected student outcomes. Faculty struggle with specifying outcomes and their rubrics, due to inexperience and lack of proper tools. Our first research objective is to accumulate and classify expected outcomes, including common assessment techniques, e.g. survey, portfolio, and projects, and associated rubrics to evaluate if expectations are exceeded, met, or not met. Classification leads to the second objective defining a many-to-many mapping of outcomes to rubrics. The third objective initiates controlled customization of outcomes, rubrics, and mappings for better evaluation coverage and to satisfy accreditation goals for continuous assessment improvement. The proposed effort defines SALR (Spiral Assessment Loop for Rubrics), based on the spiral model for software engineering, to generate, test, and evaluate assessment rubrics. SALR uses the resulting outcome classification ontology from which the mapping of outcomes to rubrics emerges coupled with current evaluations of achievement to determine needed changes. The anticipated results are documented classification of outcomes, assessment techniques and rubrics for computer science and some broader engineering sciences, ontologically based mappings of outcomes to rubrics, and spiral assessment loop prototype for commercial development. Universities can benefit from software-based rubrics development and evaluation scaled to the collegiate level. Curriculum-based rubrics immediately inform faculty of assessment issues and needed improvements. Commercial software can provide quick and easy storage, retrieval, and evaluation performed on a semester, exam, or project basis. Students are well-served by achieving success in critical education aspects. With performance-based rubrics at hand, students can understand expectations and how to reach them. Without explicit expectations, all may be surprised at failed outcomes. Unfortunately, few in higher education know how to construct proper rubrics. Current tools do not scale to university levels. Thus, rubrics must often be tailored or created from scratch. Thus, this research and development effort can remove barriers to assessment, benefiting faculty and student understanding of successful outcomes.