Abstract
This paper uses regression analysis to test design features and assignment types on 411 essays scored with the Washington State Critical Thinking rubric over the period of 11 semesters (2013-2019). Quizzes, exams, and specially designed critical thinking exercises (CTEs) are not found to boost critical thinking for online students. Participation in discussions (mainly through post reading), reading, and essay writing are all found to have a positive impact on subsequently measured CT, while the overall number of assignments in a semester tends to operate in the opposite direction. Comparatively, asynchronous discussion is not clearly superior to essay writing when it comes to promoting CT. However, the greatest payoff in online political theory courses might be to find ways to get students to read more (discussion posts and posted materials).