Using questionnaire data from the 2010 Survey of Academic Chairs, the study focuses on decision autonomy, a component of the power wielded by science, technology, engineering and mathematics (STEM) department chairs in U.S. research extensive universities. A "power index" is developed to measure chairs' decision autonomy, specifically their control of resources employed in negotiations with faculty job candidates. The study asks: What determines the degree of decision autonomy power possessed by department heads; and, what are the strategic implications of department heads' degree of this particular aspect of power? Results of an ordered logistic regression model show that having more power is associated with being hired from outside the current university, being male, and with department size. The power index is employed to predict departmental strategic priorities. Results show that the power index is positively associated with a strategic priority for research. The results show a negative relationship between degree of chair decision autonomy and a priority to increase faculty lines. A student-focused strategy is not predicted by the power index but is related to the size, with larger departments placing less emphasis on numbers or quality of students.