Rain, Rain, Go Away: 195 Potential Exclusion-Restriction Violations for Studies Using Weather as an Instrumental Variable

Mellon, Jonathan

doi:10.31235/osf.io/9qj4f

Cited by 24 publications

(18 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our results are also robust to relaxing the exclusion restriction, thus making them immune to Mellon's (2021) criticism that many weather-based instrumental variables fail to meet the exclusion restriction.…”

Section: Introductionmentioning

confidence: 72%

Is Demonstrating Against the Far Right Worth it? Evidence from French Presidential Elections

2022

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 72%

Is Demonstrating Against the Far Right Worth it? Evidence from French Presidential Elections

2022

View full text Add to dashboard Cite

“…One case where we should be particularly cautious of exclusion restriction violations through indirect causal chains is when the same instrument is used for multiple different treatments-suggesting that the instrument may affect many different things. Mellon (2021) found that weather has been used as an instrument in at least 111 studies across the social sciences. The fact that weather appears to affect so many different treatments raises concerns about its validity in any given application.…”

Section: What "No Direct Effect" Means In the Context Of The Exclusio...mentioning

confidence: 99%

“…If sunlight improves mood through pathways other than outdoor activity, and mood itself affects math performance, the exclusion restriction could be violated. 7 Indeed, Mellon (2021) locates a number of studies showing that weather affects mood and that mood in turn influences a wide range of other outcomes. Historical instruments-those that occurred long before the treatment-present another complicated case for exclusion restriction assumptions.…”

Section: What "No Direct Effect" Means In the Context Of The Exclusio...mentioning

confidence: 99%

Handle with Care: A Sociologist's Guide to Causal Inference with Instrumental Variables

Felton

Stewart

2022

Preprint

View full text Add to dashboard Cite

Instrumental variables (IV) analysis is a powerful, but fragile, tool for drawing causal inferences from observational data. Sociologists have increasingly turned to this strategy in settings where unmeasured confounding between the treatment and outcome is likely. This paper provides an introduction to the assumptions required for IV and consequences of their violations for applications in sociology. We review three methodological problems IV faces: identification bias (asymptotic bias from assumption violations), estimation bias (finite-sample bias that persists even when assumptions hold), and type-M error (exaggeration of effects given statistical significance). In each case, we emphasize how weak instruments exacerbate these problems and make results sensitive to minor violations of assumptions. Our discussion is informed by a new survey of IV papers published in top sociology journals showing that assumptions often go unstated and robust uncertainty measures are rarely used. We provide a practical checklist to show how IV, despite its fragility, can still be useful when handled with care.

show abstract

“…Specifically, we instrument fire radiative power with fire risk, a formula developed by INPE that takes into account variations in weather conditions, such as precipitation and temperature. Aware of the challenges of using weather instruments in the social sciences, we follow the recent literature and rely on short-term variation in weather conditions while controlling for longer-term weather patterns (Cooperman, 2017;Mellon, 2021). Second, we construct a stacked differences-in-differences design.…”

Section: Introductionmentioning

confidence: 99%

Global Harms, Local Profits: How the Uneven Costs of Natural Disasters Affect Support for Green Political Platforms

Pianta¹,

Rettl²

2022

Preprint

View full text Add to dashboard Cite

As the climate crisis worsens, it becomes increasingly important to understand how voters respond to first-hand experience of natural disasters. Conventional wisdom holds that exposure to natural disasters fosters environmental concern, thereby increasing support for green parties and candidates. Looking at the impact of wildfires on voting behavior in Brazil, we argue instead that exposure to natural disasters increases support for green candidates only when the costs of disasters outweigh their benefits. While fires have unambiguously negative environmental and health effects, their economic implications are not necessarily negative. In areas where fires destroy natural vegetation, newly ``cleared'' land may represent an opportunity for land grabbing and the expansion of agricultural and livestock production. We source satellite, administrative, and electoral data from Brazil and use it in two different identification strategies. Our results show that exposure to fires increases support for the main green candidate only in municipalities with low shares of employment in sectors that are likely to benefit from the degradation of the natural environment. Our findings shed light on the distributional implications of environmental degradation and their political consequences.

show abstract

Rain, Rain, Go Away: 195 Potential Exclusion-Restriction Violations for Studies Using Weather as an Instrumental Variable

Cited by 24 publications

References 25 publications

Is Demonstrating Against the Far Right Worth it? Evidence from French Presidential Elections

Is Demonstrating Against the Far Right Worth it? Evidence from French Presidential Elections

Handle with Care: A Sociologist's Guide to Causal Inference with Instrumental Variables

Global Harms, Local Profits: How the Uneven Costs of Natural Disasters Affect Support for Green Political Platforms

Contact Info

Product

Resources

About