Blog 10: The Campaign Narrative

Hi everyone! Welcome back to the blog! While I’ve spent the past 9 weeks focusing on how to predict elections and analyze forecasts, this week we’ve decided to culminate our studies with a case study: Pennsylvania District 8. Since analyzing numbers and regressions in the abstract oftentimes feels so removed from having consequence on anything meaningful in our political system, I’m hoping to contextualize our work. As such, I will start by telling the story of the race from district demographics to the campaign messages.

Blog 9: Election Reflection

Hi everyone! Welcome back to the blog. Election day is finally over and now it’s time to analyze what worked and what didn’t in my prediction forecast. To see the original forecast look here at my last blog post, but no worries if you want the TLDR as I’ll provide a quick recap below. Model Recap My model was a national level model that aimed to predict Democratic Two Party Vote Share from 5 variables:

Blog 8: Final Prediction

TLDR; If you read nothing else, my model predicts that democrats will win approximately 46.84% of the two party vote share while Republicans will win 53.16% of the two party vote share for the House midterm election in 2022. Hi everyone! We’re back this week and ready for election day! This is the last stop on my election prediction journey and so I will take the time to synthesize my findings from the past 7 weeks to formally supply an estimation of the two party vote share outcome for the 2022 midterm elections.

Blogpost 7: Shocks

Hi everyone! We’re back this week with and just two weeks out from election day! This week we covered a lot of topics but especially shocks! Shocks are bound to happen in any election for many reasons, whether it be the volatility of the economy, foreign policy events, or personal scandals brought to light by the media. The very nature of elections is unpredictable and shocks are bound to happen, hence the infamous “October Surprise”.

Blog Post 6: The Ground Game

Hi everyone! We’re back this week with another set of models this time exploring the important concept of turnout. This week through building models using a measure of turnout, I’ve set out to investigate: How much does voter turnout actually explain election outcomes on average and does turnout benefit one party over another? To begin looking at the relationship between party vote share and turnout, I graphed regressions of both the republican or democrat party vote share on turnout.

Blogpost 5: Air Campaigns

Hi everyone! Welcome back to this weeks election analytics blog post. While last week we were looking at the effect of expert prediction and polls, this week I am turning our attentions to the efficacy of air campaigns and it’s role in predicting election outcomes. Political campaigns are extraordinary operations with thousands of people working and volunteering each year. People dedicate their lives to running and studying these campaigns. With all this attention focused on persuading others and turning out support, the question becomes: How much does the adveretising element of campaigns matter?

Blog Post 4: Expert Prediction and Incumbency

## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ── ## ✔ ggplot2 3.3.5 ✔ purrr 0.3.4 ## ✔ tibble 3.1.6 ✔ dplyr 1.0.7 ## ✔ tidyr 1.2.0 ✔ stringr 1.4.0 ## ✔ readr 2.1.2 ✔ forcats 0.5.1 ## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ── ## ✖ dplyr::filter() masks stats::filter() ## ✖ dplyr::lag() masks stats::lag() ## ## Please cite as: ## Hlavac, Marek (2022). stargazer: Well-Formatted Regression and Summary Statistics Tables. ## R package version 5.

Blog post 3: Polling

Blog Post 3: Polling

Hi everyone! Welcome back to my election analytics blog. While last week we were looking at purely economic factors, this week I am turning our attentions to the power of the poll and it’s role in predicting election outcomes. Nate Silver crafts Fivethirtyeight’s election prediction in 2022 by creating 3 versions of one model: the lite, classic, and deluxe models. Each version of the model builds upon each other iterratively in complexity.

Blogpost 2: Economic Fundamentals

Hi everyone! Welcome back to my election analytics blog. This week I am looking at pure economic factors and trying to understand the age old question: Does the status of the economy influence the vote share of incumbent party candidates during elections? Scholars have argued for decades about the extent to which candidates are rewarded or penalized for the status quo of the economy during elections. The traditional thesis that has prevailed is that incumbent candidates party vote share increases when the economy is doing well but decreases when the economy is doing badly.