In this article, we explore the ability of demographic and attitudinal variables to predict student scores on the Quantitative Reasoning for College Science (QuaRCS) Assessment. Variables measured by the assessment include: students' academic choices and plans, attitudes and perceptions regarding mathematics, self-reported effort level, and basic demographics such as age, race/ethnicity, gender and disability status. As in previously published numeracy studies, we find significant score deviations according to gender, race/ethnicity, and disability status; however, the effect size of these correlations pale in comparison to the effect size of affective/ attitudinal variables on QuaRCS score. A large number of variables with significant effects on QuaRCS score make the data well-suited to dimension reduction, and Factor Analyses reveal that a majority of affective variables can be collapsed into three underlying factors, which we call numerical self-efficacy, numerical relevancy and academic maturity. These three composite variables alone account for 32.4% of the variance in QuaRCS score. Two additional affective variables -self-reported effort and calculator usage -add 15.9% to the regression model. Together, these five variables account for nearly half of the variance in QuaRCS score. In contrast, academic and basic demographic variables, account for only 0.3% and 0.1% of the remainder, respectively. Furthermore, most demographic variables (including race and gender) do not have a significant effect on the regression model once affective variables have been accounted for. Kate Follette is a NASA Sagan Postdoctoral Fellow at the Kavli Institute for Particle Astrophysics and Cosmology at Stanford University. This work was begun as a graduate student at the University of Arizona, and was supported by an NSF Transforming Undergraduate Education in STEM grant (NSF-DUE # 114-0398) and an NSF Graduate Research Fellowship. She taught general education astronomy courses as an adjunct instructor at Pima Community College from 2009-2014, and will be an assistant professor at Amherst College beginning January 2017. Her science research focuses on finding and characterizing young extrasolar planets, as well as the disks of gas and dust from which they form.Sanlyn Buxner is an assistant research professor in the Department of Teaching, Learning and Sociocultural Studies at the University of Arizona. She teaches introductory science and research methods courses and supports institutional assessment of graduate and undergraduate programs. Her research includes examining science literacy and quantitative literacy in undergraduate science students and studying the impact of research and industry experiences for K12 teachers on classroom practice and student outcomes.This article is available in Numeracy: http://scholarcommons.usf.edu/numeracy/vol10/iss1/art5Erin Dokter is an associate professor of practice in the Office of Instruction and Assessment at the University of Arizona, where she serves as the Coordinator for th...