From the abstract: "There has been perennial interest in personal qualities other than cognitive ability that determine success, including self-control, grit, growth mind-set, and many others. Attempts to measure such qualities for the purposes of educational policy and practice, however, are more recent. In this article, we identify serious challenges to doing so. We first address confusion over terminology, including the descriptor noncognitive. We conclude that debate over the optimal name for this broad category of personal qualities obscures substantial agreement about the specific attributes worth measuring. Next, we discuss advantages and limitations of different measures. In particular, we compare self-report questionnaires, teacher-report questionnaires, and performance tasks, using self-control as an illustrative case study to make the general point that each approach is imperfect in its own way. Finally, we discuss how each measure’s imperfections can affect its suitability for program evaluation, accountability, individual diagnosis, and practice improvement. For example, we do not believe any available measure is suitable for between-school accountability judgments. In addition to urging caution among policymakers and practitioners, we highlight medium-term innovations that may make measures of these personal qualities more suitable for educational purposes."
This is an important article to read for stakeholders who are considering the use of such measures in accountability systems. Professors Angela Duckworth and David Yeager, experts who study personal qualities other than cognitive ability that lead to student success, detail their concerns in using the data from these measures for purposes beyond research. Currently, the measures for social, emotional, and character traits - and the data they generate - are used for research purposes and have not been refined for use in educational settings. There are inherent advantages in using self-report questionnaires, teacher-report questionnaires, and performance tasks; however, their limitations make them invalid for use in accountability purposes. The authors provide recommendations on how to measure social, emotional, and character traits, such as using multiple measures and selecting the "most valid measure for [the] intended purpose” (p. 245).
Note: Access to the full journal article may require payment or a subscription. Education Week has summarized this article, and the American Educational Research Association has a press release with author interviews.