About Me


I am a PhD graduate from NYU's Wilf Department of Politics and current Postdoctoral Fellow at Vanderbilt's Data Science Institute with 10+ years of wide-ranging research experience in both methods and applications. I specialize in computational methods --with emphasis on NLP, networks and machine learning-- and their application to the study of political semantics and attitude judgments. My research draws in significant ways from multiple disciplines --cognitive science, computer science, linguistics and social psychology-- and has been published (or is forthcoming) in the Journal of Politics, the British Journal of Politics and the Journal of Labor Economics among others. In 2020 I launched Weavio LLC to help researchers collect and extract robust actionable insights from qualitative data at scale using NLP and AI.

Research


NLP

"Embedding Regression: Models for Context-Specific Description and Inference" (with Spirling, A. and Stewart, B.), under peer review. (link).

"Word Embeddings: What works, what doesn't, and how to tell the difference for applied research" (with Spirling, A.), forthcoming in the Journal of Politics, 2021 (link).

Cognitive Modeling/Machine Learning

"Comparing Models of Semantic Search in Concrete and Abstract Categories" (with Halpern, D.) , PsyArXiv 2018 (link).

"Partisan Representations: Partisan Differences in Semantic Representations and their Role in Attitude Judgments" (with Halpern, D.), CogSci 2018 conference proceedings (link).

Networks

"From Chatter to Action: How Social Networks Inform and Motivate in Rural Uganda" (with Larson, J. and Lewis, J.), forthcoming in the British Journal of Political Science, 2021 (link).

"More is Not Necessarily Better: The Attenuating Effect of Aggregating Networks" (with Larson, J.), 2018 working paper (link).

Lab Experiments

"Taxes, Windfalls and Accountability: The Role of Property Rights" (with Bernabel, R.), 2015, working paper (link).

Causal Inference

"Taking the Easy Way Out: How the GED Induces Students to Drop Out of School” (with Heckman, J., LaFontaine, P. and Humphreys, J.E.), Journal of Labor Economics, 2012, Vol. 30, n. 3 (link).

Venezuela & LATAM

"Backsliding by Surprise: The Rise of Chavismo" (with Kronick, D. and Plunkett, B.), 2021, under peer review (link).

"El petróleo como instrumento de progreso: una nueva relación Ciudadano-Estado-Petróleo" (with Rodríguez, L.R.), IESA Press, currently on its 2nd edition, 2012. (Translation: “Oil as an Instrument for Progress: A New Citizen-State-Oil Relationship”) (link).

"Direct Distribution of Oil Revenues in Venezuela: ¿A Viable Alternative?" (with Monaldi, F. and Morales, J.R.), Center for Global Development Working Paper 306, 2012 (link).

“De subsidiados a propietarios: replanteando el subsidio a la gasolina”, chapter in “Venezuela 2015: Economía, Política y Sociedad”, edited by Ronald Balza, UCAB Press, 2015. (Translation: “From Subsidy Recipients to Proprietors: Rethinking the Gasoline Subsidy”) (link).

“Panorama de los sistemas financieros en América Latina: avances y desafíos” (with Arreaza, A.), chapter in “Servicios Financieros para el Desarrollo: Promoviendo el Acceso en América Latina”, edited by Corporación Andina de Fomento (CAF), 2011. (Translation: “Outlook of Latin America’s Financial Sectors: Progress and Challenges) (link).

"La inflación en Venezuela: marco institucional y un modelo VAR”, Revista BCV N°2/2009. (Translation: “Inflation in Venezuela: Institutional Framework and a VAR Model") (link).

Software


Weavio Voice: App to build short voice surveys with automated transcription, sentiment and keyword extraction (with Ghobashy, M. and Elaraby, M.).

Weavio Qual-AI (demo): Semantic search engine for coding open-ended responses and feedback (with Ghobashy, M.).

conText: An R package for regression-based inferences using text data (with Spirling, A. and Brandon, S.).

dyadRobust: Cluster-robust standard errors for dyadic data (with Bisbee, J.).

shinyGeNNs: R Shiny App to collect semantic fluency data.

turingTT: R Shiny App to perform Turing-style evaluations of word embedding models (as laid out in Rodríguez, P. And Spirling, A., 2021).