My methodological research is primarily in the intersection of causal inference and big data--how can we extract what is most useful, or most interesting, from a high-dimensional or complex dataset, within a causal framework? This question encompasses two overarching research projects. In one, I incorporate machine learning models trained on a separate sample to improve statistical precision (i.e. standard errors) of design-based estimators (in randomized experiments) and/or decrease bias of matching estimators (for observational studies). In another, I incorporate latent variables--based on item response theory or model-based cluster analysis--into Bayesian principal stratification models of randomized trials, in order to use high-dimensional or complex data to understand when, for whom, or how treatments work best. I am also interested in regression discontinuity designs, natural experiments, latent variable models, hierarchical/multilevel models, and randomization or permutation inference. 

My applied statistics work is mostly in social sciences, and especially education. I have worked extensively with data from intelligent tutoring systems--computer-based educational programs--including the Cognitive Tutor and ASSISTments. As the director of statistical consulting at the UT-Austin College of Education, I work on a broad range of quantitative education research questions, including designing field trials and analyzing data. I also collaborate with researchers in psychology, health, political science and other fields. 

Representative Publications

For a complete and up-to-date list, please see my CV

• Big Data for Design-Based Causal Estimators

  • Sales, AC, Hansen, BB, and Rowan, B. (2017) "Rebar: Reinforcing a Matching Estimator with Predictions from High-Dimensional Covariates." Journal of Educational and Behavioral Statistics. Vol 43, Issue 1, pp. 3–31.

  • Sales, AC, Botelho, A, Patikorn, T and Heffernan, NN. (2018) “Using Big Data to Sharpen Design-Based Inference in A/B Tests.” Proceedings of the 11th International Conference on Educational Data Mining.

  • Sales, AC, and Hansen, BB (2019) "Limitless Regression Discontinuity" Journal of Educational and Behavioral Statistics.

• Principal Stratification for Intelligent Tutors

  • Sales, AC and Pane, JF. (2019) "The Role of Mastery Learning in Intelligent Tutoring Systems: Principal Stratification on a Latent Variable." Annals of Applied Statistics. Vol 13, No 1, 420-443.

  • Sales, AC and Pane, JF. (2017) Principal Stratification for Intelligent Tutors: A Tutorial (Working paper).

  • Sales, AC, Wilks, A, and Pane, JF. (2016) “Student Usage Predicts Treatment Effect Heterogeneity in the Cognitive Tutor Algebra I Program.” Proceedings of the 9th International Conference on Educational Data Mining.

  • Williams, JJ, Botelho, A, Sales, AC, Heffernan, N, and Lang, C. (2016) “Discovering ‘Tough Love’ Interventions Despite Dropout.” Proceedings of the 9th International Conference on Educational Data Mining.

• Applied Research

  • Berger, W J, & Sales, AC. (2019). Testing epistemic democracy’s claims for majority rule. Politics, Philosophy & Economics. 

  • Johnson, K, Sales, AC, Rew, L, Garing, JH, and Crosnoe, R. (2019)  Using polytomous latent class analysis to compare patterns of substance use and co-occurring health-risk behaviors between students in alternative and mainstream high schools. Journal of Adolescence 75, 151-162.

  • Calzada, E. and Sales, AC (2018). Depression Among Mexican-Origin Mothers: Exploring the Immigrant Paradox. Cultural Diversity and Ethnic Minority Psychology. 

  • Garberoglio, C. L., Cawthon, S., and Sales, AC. (2017). Deaf People and Educational Attainment in the United States: 2017. Washington, DC: U.S. Department of Education, Office of Special Education Programs, National Deaf Center on Postsecondary Outcomes.

  • Thurman, W, Johnson, K, Gonzalez, D, & Sales, AC. (2018). “Teacher support as a protective factor against sadness and hopelessness for adolescents experiencing parental incarceration: Findings from the 2015 Texas Alternative School Survey.” Children and Youth Services Review. Vol 88. pp. 558–566.

Links to Working Papers

• Sales, AC, Gagnon-Bartsch, J, Wu, E, Botelho, A, Heffernan, NT, Patikorn, TM and Miratrix, L.  Precise Unbiased Estimation in Randomized Experiments using Observational Auxilliary Data.

• Israni, A., Sales, A.C., and Pane, J. Mastery Learning in Practice: A (Mostly) Descriptive Analysis Based on the Cognitive Tutor Algebra I Eff.ectiveness Trial. In Revision

• Sales A.C. Sequential Specification Tests to Choose a Model: A Change Point Approach.

• Sales, A.C., and Pane, J.F. Student Log-Data from a Randomized Evaluation of Educational Technology: A Causal Case Study. Conditionally accepted.

• Sales, A.C., and Pane, J.F. Principal Stratification for Intelligent Tutors: A Tutorial