Better Inferences at the Congressional District Level

Better Inferences at the Congressional District Level

About a week ago I wrote about how to estimate congressional district level opinion from the Cooperative Congressional Election Survey, which is a large national opinion poll. I sent the post to Andrew Gelman, a statistician and political scientist, who gave me some great feedback on the methodology I used. In this post, I've revised my opinion model based on that feedback, and have made a few other additional enhancements as well. Compared to the previous model, this iteration produces very similar point estimates, but with reduced variance. Here I'll explain the changes and the reasoning behind them.

In the original post, I used a hierarchical student T distribution to model district level regression intercepts, with the standard deviation and degrees of freedom estimated from the data. I did not use a state or regional level regression as is typically done in the literature. There were two main reasons for this, the first is that I tended to get inefficient inferences when I included the state and regional level variation, which I blamed on correlations between the district level predictors and the state/regional regression indicators. However, at Andrew's suggestion I tried using more informative priors for the variance parameters in the model (which for example represent the standard deviation of the group distribution for the district level intercepts) and this problem resolved instantly.

Read More

Analyzing Opinions in Congressional Districts from Large National Polls

Analyzing Opinions in Congressional Districts from Large National Polls

I recently came across an article in Vice by Sean McElwee which made a very compelling case for Democrats moving to the left on issues like immigration and racial justice to mobilize millennial voters instead of moving to the center on these issues in an attempt to win over white working class voters. This is a move that aligns with my personal politics, but it seems to run completely contrary to a good deal of conventional punditry on how Democrats should go about winning elections. The author made his case by analyzing several questions from the 2016 Cooperative Congressional Election Survey, which is a very large national survey that covers a number of topics related to policy and political attitudes.

In this post, I'll attempt a deeper dive into some of the CCES data using Bayesian Inference with PyMC3. I'm mainly interested in how much the CCES data can tell us about how opinions vary by congressional district, with a particular focus on college educated millennials. The results backup the main thesis of the vice article pretty strongly, so I won't try to rehash much of the commentary. Instead, this post will be more focused on the computational and modeling issues surrounding this type of analysis.

Read More

Which GOP Lies on Tax Reform are the Most Effective?

A few weeks ago, we marveled at how the GOP has been using populist messaging on their tax reform bill, which is a massive giveaway to wealthy individuals and profitable corporations. We recently stumbled across a poll which sheds some light on exactly why the GOP can't simply be honest about their bill. The poll was conducted by the American Action Network, a conservative nonprofit issue advocacy group. To get an idea of what the AAN is all about, check out this attack ad they ran against the Consumer Financial Protection Bureau using Soviet imagery. If you are not familiar with the CFPB, it is a independent agency whose creation was largely the work of Massachusetts Senator Elizabeth Warren and investigates predatory practices in the financial industry. 

Aside from being commissioned by a right wing organization with a habit of comparing basic consumer protections to the gulag, only conservatives from Republican districts were surveyed. So if there is any popular support for tax reform, then this is where we should find it. In this group 77% thought that Congress should cut taxes for individuals and 54% thought that Congress should cut taxes for individuals and businesses. A bare majority of support for tax cuts on individuals and businesses is pretty terrible when the group you are polling is the group that should be the most enthusiastic about tax reform. This number is consistent with recent pew polling which found that only 48% of conservative Republicans and 41% of Republicans overall thought taxes should be lowered on large businesses. The same Pew poll also found that 36% of Republicans want taxes cut for households incomes over 250k.

Read More

Your District is More Flippable Than You Think

Your District is More Flippable Than You Think

In our series on voter turnout, we previously looked at the congressional abstention rates (casting a ballot but not voting for a congressional candidate) in New Jersey Municipalities in the 2016 election. After accounting for the statewide trend of higher congressional abstention in municipalities with lower incomes, we found a pretty strong district level trend where turnout was higher in districts that ended up being more competitive races. Since voters can't know the outcome of the election before it happens, any direct effect on turnout must be due to perceived competitiveness. This made us wonder about how perceptions of competitiveness can be manipulated to change the actual competitiveness of a race. With this in mind, in this post we will look at New Jersey's 11th district, which has been held for over two decades by Republican Rodney Frelinghuysen. 

Frelinghuysen has won reelection very comfortably for his whole career, most recently defeating Democrat Joe Wenzel and winning 58% of the vote in 2016. Republicans generally dominate state and local public offices in the district as well, and the conventional wisdom is that the district is a fairly safe Republican seat. As far as we can tell however, the safety of Frelinghuysen's seat depends on his campaigns consistently turning out a Republican base that is not nearly as large as Frelinghuysen would like opponents to believe. This job is made even more difficult because he needs to convince substantial numbers of these core supporters to consistently vote for the Republican party against their own policy interests. 

Read More

The GOP's Populist Messaging on Tax Cuts for the Ultra Wealthy

The GOP's Populist Messaging on Tax Cuts for the Ultra Wealthy

GOP messaging on tax reform has been interesting to say the least. Tax cuts for the wealthy are very unpopular, except among the GOP donor class. 'Trickle Down' economic theory, which asserts that low taxes on corporations and wealthy individuals spurs economic growth that ultimately benefits everyone in society, is also a tough sell. Empirical studies have shown that in reality, such cuts do the opposite of what trickle down theory says they should; they increase budget deficits (which Republicans supposedly care about), and even diminish economic growth. The abject failure of high tax cuts in Kansas serves as a recent warning of just how damaging these policies can be to the economy.

At the end of the day, the GOP donors want their tax cuts though, so it is up to the GOP to pitch them to the American people. They have set up a website fairandsimple.gop to make their case. Read the whole site if you have time, it is truly fascinating, and the way they talk about elimination of various tax deductions is particularly interesting. Elimination of certain itemized deductions are discussed under the heading: "Eliminates loopholes for the wealthy, protects bedrock provisions for middle class."

This clearly invokes some of the populist messaging that has been used by various candidates that appeals to the sense that many have that the wealthy in the country do not pay their fair share in taxes. Combine that with the fact that the GOP plan doubles the standard deduction, which lowers the tax liability for those who do not have very many itemized deductions, and it seems like perhaps the GOP is really making an honest effort to lower taxes for the middle class.

Read More