I am an assistant professor of Political Science at the University of North Carolina, Chapel Hill. My research focuses on the use and development of novel computational methods for quantitative political research (particularly Bayesian and Machine Learning probabilistic models), the measurement of latent traits, and the political consequences of electoral and legislative institutions. You can download my CV here.
- Assistant Professor
- 2017 - Present · University of North Carolina at Chapel Hill
- Visiting Assistant Professor
- 2019 - 2020 · Harvard University
- Visiting Associate Research Scholar
- 2015 - 2017 · Princeton University
- Assistant Professor
- 2013 - 2015 · University of Miami (Miami, FL)
- Ph.D. in Political Science
- 2007 - 2013 · Washington University in St. Louis
- M.A. in Political Science
- 2007 - 2009 · Washington University in St. Louis
- B.A. in Political Science
- 2001 - 2006 · Universidad de los Andes (Bogotá, Colombia)
2018. Olivella, S., and Montgomery, J. B. Tree Based Methods for Political Science. American Journal of Political Science 62(3): 729–744.
2017. Uscinski, J. and Olivella, S. . The Conditional Effect of Conspiracy Thinking on Attitudes toward Climate Change. Research & Politics October-December 2017: 1–9.
2017. Olivella, S., Kanthak, K. and Crisp, B. ...And Keep Your Enemies Closer: Cosponsorship Patterns as Reputation Building. Electoral Studies 46(1): 75-–86.
2015. Montgomery, J., Olivella, S., Potter, J., and Crisp, B. . An Informed Forensics Approach to Detecting Vote Irregularities. Political Analysis, 23(4): 488–505.
2015. Potter, J. and Olivella, S. Electoral strategy in geographic space: Accounting for spatial proximity in district-level party competition. Electoral Studies, 40(1): 76–86.
2014. Atkinson, M., Mann, C., Olivella, S., Simon, A. and Uscinski. J. (Where) Do Campaigns Matter? The Impact of National Party Convention Location. The Journal of Politics, 76(4): 1045–1058.
2014. Crisp, B., Olivella, S., Potter, J. and Mishler, W. . Elections as Instruments for Punishing Bad Representatives and Selecting Good Ones. Electoral Studies, 31(1): 1–15.
2013. Olivella, S. and Tavits, M. Legislative Effects of Electoral Mandates. British Journal of Political Science, 44(2): 301–321.
2013. Olivella, S. and Tavits, M. Vote-Earning Strategies in Flexible List Systems: Seats at the Price of Unity. Electoral Studies, 32(4): 658–669.
2013. Crisp, B., Olivella, S., and Potter, J. Party System Nationalization and the Scope of Public Policy. Comparative Political Studies, 46(4): 431–456.
2012. Crisp, B., Olivella, S., and Potter, J. Electoral contexts that impede voter coordination. Electoral Studies, 31(1): 143–158.
2006. Olivella, S. and Vélez, C. Will Uribe’s coalition survive?” (in Spanish). Colombia Internacional, 64(2): 194–205.
2020. Crisp, B. F., Olivella, S. and Rosas, G. The Chain of Representation: Preferences, Institutions, and Policy in Latin America’s Presidential Systems. Cambridge University Press.
2020. Shoub, K., and Olivella, S. . 'Machine Learning in Political Science' (in The SAGE Handbook of Research Methods in Political Science and International Relations ) . SAGE Publications.
2013. Crisp, B., Olivella, S., and Potter, J. 'A Comparison of different indicators of party system consolidation' (in Spanish) in Torcal, M. (Ed.) Sistemas de partidos en América Latina. Causas y consecuencias de su equilibrio inestable. Editorial Anthropos.
2011. Olivella, S. 'Cross-Tabular Analysis.' in Badie B. et al. (Eds.) International Encyclopedia of Political Science. SAGE Publications.
2009. Olivella, S. and Rodríguez-Raga, J. C. . 'What is spatial is special: Proximity voting in Colombia' (in Spanish). in Botero, F. (Ed.) Juntos pero no revueltos? Partidos, candidatos y campañas en las elecciones legislativas de 2006 en Colombia. Uniandes - CESO.
- NetMix (R package)
- Variational EM estimation of mixed-membership stochastic blockmodel for networks, incorporating node-level predictors of mixed-membership vectors, as well as dyad-level predictors.
- poisbinom (R package)
- Provides the probability, distribution, and quantile functions and random number generator for the Poisson-Binomial distribution.
Work in progress
- Dynamic Stochastic Blockmodel Regression for Social Networks: Application to International Militarized Conflicts (w/ T. Pratt and K. Imai)
- A primary goal of social science research is to understand how latent group mem- berships predict the dynamic process of social network evolution. In the modeling of international conflicts scholars hypothesize that membership in geopolitical coalitions shapes the decision to engage in militarized conflict. Such theories explain the ways in which nodal and dyadic characteristics affect the evolution of relational ties over time via their effects on group memberships. To aid the empirical testing of these arguments, we develop a dynamic model of social networks by combining a hidden Markov model with a mixed-membership stochastic blockmodel that identifies latent groups underlying the network structure. Unlike existing models, we incorporate co- variates that predict node membership in latent groups as well as the direct formation of edges between dyads. While prior substantive research often assumes the deci- sion to engage in militarized conflict is independent across states and static over time, we demonstrate that conflict patterns are driven by states’ evolving membership in geopolitical blocs. Changes in monadic covariates like democracy shift states between coalitions, generating heterogeneous effects on conflict over time and across states.
- Representing communities of economic interest in the U.S. House (w/ S. Treul and N. Pinnell)
- Communities of shared interest can easily fail to coincide with congressional dis-tricts. When areas that demarcate these communities span multiple congressionaldistricts, do representatives shift their efforts to reflect the interests of a broader set ofcitizens, or do they maintain a focus on the local interests of their own constituents?Using journey-to-work data for millions of Americans, we measure the extent to whichone such community — the labor market area — crosses congressional district bound-aries, and evaluate the effect of these shared economic interests on the representationalefforts of House members. We find that members coming from a common labor marketarea tend to collaborate more often, and that members representing districts embed-ded in larger labor market areas sponsor more legislation on national issues. Thesepatterns may exacerbate polarization in Congress and hurt representation of urbanminority populations, even in the absence of malapportionment.