leoahrens.bsky.social
PolSci researcher at the Varieties of Egalitarianism project @EXCInequality. Politics of inequality, tax & welfare policies, public opinion, voting behavior.
leoahrens.eu
185 posts
847 followers
132 following
Active Commenter
comment in response to
post
Just like Economists are unaware that multilevel models exist!
comment in response to
post
I'll leave this here
comment in response to
post
That would be much appreciated! Thank you
comment in response to
post
data from a wide variety of countries.
comment in response to
post
more credible identification to lend credibility to the theory. Even in this optimal world, the cross-sectional analysis still has value—it offers evidence supporting the claim that the credible causal effect estimates from, say, an experiment are “replicated” in observational
comment in response to
post
I am fully aware of the weaknesses relating to causal identification of the present paper, but it can show that real-world data is consistent with the presented theory. Optimally, further people will do connected research and use a different approach with
comment in response to
post
In my opinion, important topics should be analyzed with a mix of different methods due to their unique strengths and weaknesses—and cross-sections have their place in this methods mix, despite their weaknesses.
comment in response to
post
But it always remains unclear to what extent the results from the survey/lab actually translate into real-life politics. Experiments offer an artificial environment with often unclear generalizability. For me, that is the unique strength of observational data.
comment in response to
post
I have a similar reservation regarding experiments. Of course, it would be possible to set up a survey or lab experiment that analyzes the research question at hand, for example by triggering people to see their government as incompetent, etc.
comment in response to
post
Multi-country cross-sections, as this paper, analyze a much broader set of contexts, which weakens this concern.
comment in response to
post
And what about social media use other than early-stage Facebook, which was quite different to today? We don’t know from this paper.
comment in response to
post
Economists recently used the geographically staggered rollout of Face-book in a natural experiment for causal estimates. This is great, but what about social media use in other countries?
comment in response to
post
Natural experiments tend to analyze a very specific case, and it remains unclear to what extent the specific results hold across contexts. A good example is the relationship between social media use and depressions.
comment in response to
post
The superior methods (from a causal id perspective) also have problems, or to put it more positively, cross-sections offer some advantages relative to them. The advantages relate to generalizability.
comment in response to
post
I consider the research topic to be important enough to warrant research with a suboptimal methodological approach.
comment in response to
post
The alternative would be to drop the research topic, but this creates a world where researchers only look for evidence where there is an identification strategy, which leaves many blind spots akin to the drunk who only looks for his/her keys under the street light.
comment in response to
post
In an optimal world, there would be a natural experiment available for analysis. But I found no case where there is also access to suitable survey data. In my opinion, the best approach is to take the next best solution, that is, use a cross-section.
comment in response to
post
Next, assuming that the research topic is worthwhile and that it addresses an important topic (which I would argue the present paper does), cross-sectional analyses may be the only approach to study the topic in a broad range of settings.
comment in response to
post
However, I fully agree that causal language should be avoided in the presentation of the empirical results, and I edited the manuscript accordingly (precisely because of the weak identification).
comment in response to
post
Substituting “drives” for “associates with”, etc., strikes me as obfuscation. This is why I decidedly use causal language – it is not because I fully believe in the validity of causal id assumptions underpinning the design.
comment in response to
post
Many authors simply avoid using causal language and depict their research as „descriptive“ to avoid this line of critique. But, if you carefully read their theories, they remain causal (as most theories on the relationship between concepts are).
comment in response to
post
First, the decision to use causal language despite the weak id strategy is conscious. I think it is the most honest approach to make explicit causal claims in the theory because the theory is causal.
comment in response to
post
Unfortunately, this point cannot be addressed easily. While it would be easy to dismiss all cross-sectional research like this paper, I will hereafter offer my perspective of why this research approach, and thus the present paper, is worthwhile nonetheless.
comment in response to
post
and it should be expected that the presented estimates deviate from the true causal effect (despite my best attempts to avoid this).
comment in response to
post
I generally agree and wholeheartedly emphasize with this critique. The assumptions underpinning a “selection on observables” identification strategy (if you can even call it an id strategy) are very strong,
comment in response to
post
"I standardize x so that it has a standard deviation of one. Therefore, all regression coefficients indicate expected changes in y when x increases by one standard deviation."
comment in response to
post
Thank you!
comment in response to
post
The paper was first written as a normal article, then transformed to a short research note, a long paper again, then a note again. (Responding to four reviewers with a 4k word limit is tough). I kept believing! I chalk this up to the randomness of the review process.
comment in response to
post
Full disclosure: I thought this paper would be the easiest to publish from my dissertation because I like the argument and findings. It ended up being rejected eight times. My record by far!
comment in response to
post
With this publication, all of the papers from my dissertation are out! I am very happy to have completed a journey that I began six years ago.
Many thanks to: @thomasrixen.bsky.social @dweisstanner.bsky.social @andreabinder.bsky.social r.bsky.social Lukas Hakelberg
comment in response to
post
Get your free article copies here:
www.tandfonline.com/eprint/JN5C9...
A preprint and a full replication package are available on my website (leoahrens.eu/publications).
comment in response to
post
One implication is that the politics of highly developed countries are less applicable to developing countries.
comment in response to
post
For example, research shows that government quality increases voter turnout of the rich and voting across class lines. My study serves as a micro-level foundation for this.
@pavisuri.bsky.social
comment in response to
post
Assuming that public opinion affects politics and policies, this suggests that the economic left-right conflict is stronger in countries with higher government quality and weaker in those with less government quality.
comment in response to
post
The results primarily support the expectation that, under higher government quality, the opponents of income equalization fear the effectiveness and impartiality of redistribution, thus turning against government action compared to countries with less government quality.
comment in response to
post
Overall, the study suggests that government quality increases the heterogeneity of public opinion on redistribution. It serves as a wedge that drives attitude variance between those who have more/less to gain from redistribution (both materially and ideologically).
comment in response to
post
The results are robust to alternative specifications and a wide range of potential other country-level moderators, such as income inequality.
comment in response to
post
Further evidence shows that this primarily results from those I theorize as opponents of income equalization (i.e. high income / low unfairness perceptions) having disproportionally lower redistribution demand under higher government quality.
comment in response to
post
The results show that effect sizes vary with government quality. Under higher government quality, income and unfairness perceptions considerably shift policy preferences. Under lower government quality, income and unfairness perceptions barely make a difference.
comment in response to
post
On to regression-based evidence. I estimate interaction regression models with country fixed effects, modeling government quality as a moderator of income and unfairness perceptions.
comment in response to
post
At the same time, within-country variance of attitudes increases. This offers first support for my theory – public opinion is more varied under higher government quality.
comment in response to
post
I begin with descriptive statistics. Contradicting the expectations of the trust literature, countries with higher government quality have lower support for redistribution and unemployment assistance, on average.
comment in response to
post
I test my theory with 2009 ISSP data, using a country-level government quality indicator. It includes 40 quite diverse countries.
comment in response to
post
This may explain why the findings from trust studies were inconclusive so far, with many reporting null results, especially in experimental settings. The effect of trust / government quality may be net neutral or even negative.
comment in response to
post
In contrast, I argue that government quality creates a rift in public opinion. It increases the redistribution support of those, e.g., with lower income and decreases the support of those with, e.g., higher income.
comment in response to
post
I draw from research on political trust. The literature expects that trust in public institutions increases demand for redistribution/welfare. This research closely related because government quality fosters trust.
comment in response to
post
Quality of government thus serves as a moderator that increases the effects of preference drivers, in particular income and unfairness perceptions.
As government quality increases, attitudes are more stratified between those with higher/lower income/unfairness perceptions.