Standard seeker: A system for aligning content, curricula, and assessment to learning objective benchmarks for standards-based education

This Small Business Innovation Research Phase I project will develop a prototype system for semi-automatically aligning content, curricula, and assessment items to learning objectives to promote standards-based education reform in American K-12 schools. This system, when employed by professional content correlators or novice teachers, will provide accurate matches between learning materials and assessments. It will gauge adequate coverage of state standard learning objectives. Latent Semantic Analysis (LSA), a type of machine learning, will be employed as the text matching engine. LSA has been proven in numerous information processing applications to match human judgments for the conceptual similarity of texts. Research topics include: (1) assessment of reliability of system performance in generation of automatic matches when compared to human performance, (2) usability of the system by professional and novice correlators, (3) automatic classification of the depth of knowledge requirements for learning materials and assessments, and (4) the usefulness of expert generated meta-statements about item content in the alignment process. Pilot experiments in automated matching of assessment items to state standards suggest that expert levels of reliability can be achieved. Phase I will result in a web-based prototype which will suggest standards benchmarks when provided with learning materials or test items.