The Politics of Experimentation: Political Competition and Randomized Controlled Trials

Dorsch, Michael; Huci, Cristina Corduneanu; Maarek, Paul

doi:10.2139/ssrn.3648321

Cited by 1 publication

(1 citation statement)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While a few reviews on this topic have been conducted, these do not address the effects of political competition interventions. Previous studies have focused on the relationship between highly politically competitive societies and the number of RCTs used to evaluate public policy (Dorsch et al, 2020), the relationship between electoral promises and voting behavior (Guo, 2020), and the impact of a range of development interventions on power-related outcomes (IRC, 2016). In addition, meta-analysis has been used to analyze the variables which affect voting behavior (Stockemer, 2017) and voter turnout (Cancela & Geys, 2016), and one qualitative systematic review has explored disability inclusion during elections (Virendrakumar et al, 2017).…”

Section: Executive Summarymentioning

confidence: 99%

Strengthening good governance in low- and middle-income countries: an evidence gap map

Parrao¹,

Lwamba²,

Khan³

et al. 2023

View full text Add to dashboard Cite

Using 3ie's EGM platform, we created an online map of studies according to the interventions and outcomes covered by included studies. This platform also allows users to explore the available evidence through filters (e.g., regions, populations, democracy levels, etc.). The interactive online map can be viewed at https://developmentevidence.3ieimpact.org/egm/political-competition-through-electionsevidence-gap-map. Main findingsAn extensive search of peer-reviewed and grey literature, as well as citation tracking for included IEs and SRs, was conducted in October 2021 and January 2022. These searches returned a total of 34,139 records. After removing duplicates, we screened 25,648 studies by title and abstract. We identified 766 potentially relevant studies, which were screened based on full-text. We included 194 studies in the EGM, of which 188 used quantitative impact evaluation methods, four used qualitative methods that seek to establish causal inference, and two were systematic reviews. These studies were published from 2003 and their publication has expanded rapidly since 2010, with 31 studies being published in 2020 alone.South Asia was the region with the highest number of evaluations. This is driven by India, which covered 27 percent of the evidence. Research was also focused on Latin America and the Caribbean and Sub-Saharan Africa. The majority of included studies were conducted in electoral democracies and in countries which were not classified as fragile and conflict-affected settings. Information on specific populations targeted by included interventions was generally lacking. The high proportion of studies evaluating interventions targeting the entire population can be driven by the number of interventions that act as legislative changes affecting the whole electorate; for instance, electoral system or electoral rules reforms.The most common intervention category identified was voter information, voter education, and GOTV. This category included interventions that typically provided voters with information and reminders on where and when to vote, information on the incumbent's performance, information on corruption, or conducted anti-vote buying campaigns. The second largest intervention category was quotas for elected positions. Almost all of these studies evaluated the introduction of gender and caste-based quotas in India.Outcomes related to inclusive participation and competition were the most common measures in two-thirds of the evidence. Over half of included studies measured immediate voting behavior, primarily through indicators of turnout and party/candidate vote share.The two systematic reviews identified in the map focused on quotas for elected positions and on aspects of voter information, voter education, and GOTV. However, both were assessed as having low confidence, meaning their results are subject to a high risk of bias.iv Conclusion and implicationsThis body of evidence is suited to shed a light on interventions implemented by public institutions at the national level and targeting...

show abstract