Regression Discontinuity Designs as Local Randomized Experiments

Abstract

In the seminal paper from 1960, Thistlethwaite and Campbell (1960) introduce the key ideas underlying regression discontinuity (RD) designs, which, even if initially almost completely ignored, have then acted as a fuse of a blowing number of studies applying and extending RD designs starting from the late nineties. Building on the original idea by Thistlethwaite and Campbell (1960), RD designs have been often described as designs that lead to locally randomized experiments for units with a realized value of a so-called forcing variable falling around a pre-fixed threshold. We embrace this perspective, and in this discussion we offer our view on how the original proposal by Thistlethwaite and Campbell (1960) should be formalized. We introduce an explicit local overlap assumption for a subpopulation around the threshold, for which we re-formulate the Stable Unit Treatment Value Assumption (SUTVA), and provide a formal definition of the hypothetical experiment underlying RD designs, by invoking a local randomization assumption. A distinguishing feature of this approach is that it embeds RD designs in a framework that is fully consistent with the potential outcome approach to causal inference. We discuss how to select suitable subpopulation(s) around the threshold with adjustment for multiple comparisons, and how to draw inference for the causal estimands of interest in this framework. We illustrate our approach in a study concerning the effects of University grants on students' dropout.

New multivariate tests for assessing covariate balance in matched observational studies

“…Since our tests only need a similarity measure on the sample space, they are not limited to multivariate data, for example, the covariate could be a curve over time, a network, or a text document, and the dissimilarity among these objects could be provided by domain experts (Mozer et al ., 2020). Balance tests are also commonly applied to regression discontinuity designs (Cattaneo et al ., 2015; Mattei and Mealli, 2016; Branson and Mealli, 2018; Sales and Hansen, 2020), where it is assessed whether units above and below a threshold are similar in terms of background covariates. The methods proposed in this paper can be applied to this context as well.…”

The Expediting Effect of Monitoring on Infrastructural Works. A Regression-Discontinuity Approach with Multiple Assignment Variables

“…In the traditional econometric approach, the assignment variable is considered a pretreatment covariate and, under relatively mild continuity (also termed smoothness) assumptions, inference relies on some form of extrapolation at a cutoff point. A handful of recent methodological papers have explored the possibility to view regression discontinuty designs as locally randomised experiments in a region around the cutoff, requiring assumptions other than smoothness for identification (e.g., Mattei and Mealli, 2016;Cattaneo et al, 2015). In this study, we will follow the more established econometric strand of the methodological literature, as it already provides some extensions to cases, like ours, where the assignment variables are more than one.…”

Limitless Regression Discontinuity

“…Rather, a marked seasonal trend is apparent, with death counts being higher in the winter months than during the rest of the year; (1) cannot be sustained. (For RDD methodology founded on (1), see Cattaneo, Frandsen, and Titiunik (2015), Aronow, Basta, and Halloran (2016) or Mattei and Mealli (2016). )…”