Daniel E. Ho

Randomization Inference with Natural Experiments: An Analysis of Ballot Effects in the 2003 California Recall Election

Since the 2000 U.S. Presidential election, social scientists have rediscovered a long tradition of research that investigates the effects of ballot format on voting. Using a new dataset collected by the New York Times, we investigate the causal effect of being listed on the first ballot page in the 2003 California gubernatorial recall election. California law mandates a complex randomization procedure of ballot order that approximates a classical randomized experiment in a real world setting. The recall election also poses particular statistical challenges with an unprecedented 135 candidates running for the office. We apply (nonparametric) randomization inference based on Fishers exact test, which incorporates the complex randomization procedure and yields accurate confidence intervals. Conventional asymptotic model-based inferences are found to be highly sensitive to assumptions and model specification. Randomization inference suggests that roughly half of the candidates gained more votes when listed on the first page of the ballot.

Measuring Explicit Political Positions of Media

We amass a new, large-scale dataset of newspaper editorials that allows us to calculate fine-grained measures of the political positions of newspaper editorial pages. Collecting and classifying over 1500 editorials adopted by 25 major US newspapers on 495 Supreme Court cases from 1994 to 2004, we apply an item response theoretic approach to place newspaper editorial boards on a substantively meaningful — and long validated — scale of political preferences. We validate the measures, show how they can be used to shed light on the permeability of the wall between news and editorial desks, and argue that the general strategy we employ has great potential for more widespread use.

Randomization Inference With Natural Experiments

Since the 2000 U.S. Presidential election, social scientists have rediscovered a long tradition of research examining the effects of ballot format on voting. Using a new dataset collected by The New York Times, we investigate the causal effect of being listed on the first ballot page in the 2003 California gubernatorial recall election. California law mandates a unique randomization procedure of ballot order that, when appropriately modeled, can be used to approximate a classical randomized experiment in a real world setting. We apply randomization inference based on Fishers exact test, which directly incorporates the exact randomization procedure and yields accurate nonparametric confidence intervals. Our results suggest that being listed on the first ballot page causes a statistically significant increase in vote shares for more than 40% of the minor candidates, whereas there is no significant effect for the top two candidates. We also investigate how randomization inference differs from conventional estimators that do not fully incorporate Californias complex treatment assignment mechanism. The results indicate appreciable differences between the two approaches.

Improving the presentation and interpretation of online ratings data with model-based figures

Online ratings data are pervasive, but are typically presented in ways that make it difficult for consumers to accurately infer product quality. We propose an easily understood presentation method that has the virtue of incorporating a parametric model for the underlying ratings data. We illustrate the method with new data on the content quality of news outlets, and demonstrate its reliability and robustness with an experiment of online users and a simulation study. Our simple approach is easy to implement and widely applicable to any presentation of ratings data.

Assessing Political Positions of Media

Although central to understanding the role of the media, few quantitative measures of the political positions of media exist. Collecting and classifying editorials adopted by 23 major U.S. newspapers on 495 Supreme Court cases from 1994-2004, we apply an item response theoretic approach to place newspapers on a substantively meaningful - and long validated - scale of political preferences. Our results provide significant insights into the study of the media. We show that 17 of the 23 papers are more likely to the left of the median Justice for this period, but also find considerable evidence that this may be an artifact of the liberalness of urban, elite, high circulation papers.

The Impact of Partisan Electoral Regulation: Ballot Effects from the California Alphabet Lottery, 1978-2002

How does partisan regulation of political markets affect elections? We investigate how the partisan control of ballot format, which is expressly regulated - often to the apparent advantage of incumbents and major parties - in all U.S. states, affects voting. Through the analysis of a unique natural experiment, we focus specifically on the longstanding question of whether the name order of candidates on ballots affects election outcomes. Since 1975, California law has mandated randomizing the ballot order with a lottery. Previous studies, relying overwhelmingly on observational data, have yielded largely conflicting results. Using improved statistical methods, our analysis of statewide elections from 1978 to 2002 reveals that ballot order might have changed the winner in twelve percent of all primary races, including major and minor party races. We propose that all electoral jurisdictions should randomize ballot order to minimize ballot effects, and show that randomization may be substantially more cost-effective at reducing voting bias than currently proposed voting technology reforms.

Does Class Size Affect the Gender Gap? A Natural Experiment in Law

We study a unique natural experiment in which Stanford Law School randomly assigned first-year students to small or large sections of mandatory courses from 2001 to 2011. We provide evidence that assignment to small sections closed a slight (but substantively and highly statistically significant) gender gap existing in large sections from 2001 to 2008; that reforms in 2008 that modified the grading system and instituted small graded writing and simulation-intensive courses eliminated the gap entirely; and that women, if anything, outperformed men in small simulation-based courses. Our evidence suggests that pedagogical policy—particularly small class sizes—can reduce, and even reverse, achievement gaps in postgraduate education.