High quality student outcome measures provide the foundation for effective learning assessment. These outcomes can be organized into different categories, such as skills or habits of mind, and evidence types, such as direct or indirect. Prevailing models of assessment, however, focus only on the distinction between direct and indirect evidence without incorporating the outcome categories. Such a binary model of evidence can be limiting, privileges measures of skills and knowledge over habits of mind and values, and risks excluding validated psychometric tools as a source of measurement. An argument is made for a new 2x2 outcome matrix which incorporates outcome category and evidence type as a possible alternative to the dominant direct vs. indirect dichotomy.