G. Elliott Morris's Newsletter $5 / month

G. Elliott Morris's Newsletter delves into the intricacies of polls, public opinion, and democracy, emphasizing the analysis and improvement of polling techniques, voter behavior, and the implications for electoral outcomes. It offers critiques, updates on methodologies, and reflections on the dynamics within the Republican party, particularly concerning Donald Trump.

Polling Analysis and Methodologies Public Opinion and Voter Behavior Election Forecasts and Predictions Political Commentary Democracy and Electoral Systems Data Journalism Republican Party Dynamics

The hottest Substack posts of G. Elliott Morris's Newsletter

And their main takeaways

A revised method for measuring pollster quality, now published at 538

157 implied HN points • 25 Jan 24
🇺🇸 U.S. Politics
  1. 538 has updated their pollster ratings for the 2024 general election, focusing on empirical accuracy and methodological transparency
  2. The new methodology ranks pollsters based on bias, recency, sample size, time to election, and methodological transparency
  3. Pollscore and Transparency Score are combined to evaluate pollster quality, with a star grading system used to represent performance

Happy Election Day -300!

196 implied HN points • 11 Jan 24
🇺🇸 U.S. Politics Elections Polls Presidential Forecasting
  1. Polls are not very useful 300 days before an election, but become more reliable as the election day approaches.
  2. Poll discourse before a certain point historically lacks accuracy and predictive value.
  3. Analyzing crosstabs for major shifts in vote choice between elections may lead to misleading interpretations.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

A new framework for media polling analysis

452 implied HN points • 22 May 23
🇺🇸 U.S. Politics Data journalism Election forecasting Media Analysis
  1. Polls should be adjusted based on both accuracy and bias, evaluated based on all factors of data-generating process.
  2. Future polling aggregation may move towards Bayesian statistical techniques for measurement-oriented models.
  3. Models focusing on measurement over prediction can help explore polling components directly, improving journalistic storytelling.

The philosophical and empirical cases for ignoring bad or biased pollsters | #210 - February 26, 2023

157 implied HN points • 27 Feb 23
🇺🇸 U.S. Politics Philosophical Bias
  1. Aggregating public opinions may mean slightly less accurate election forecasts, which is acceptable.
  2. Polls dramatically overperformed expectations in the 2022 midterms, despite popular misconceptions.
  3. It's important to be cautious of biased pollsters and consider the methodology, credibility, and track record of pollsters before including their data in models.

Links for February 19-25, 2023 | Americans in most states support legal abortions; New graphs for showing polling uncertainty at PBS; And evidence of economic voting from the 1790s

137 implied HN points • 25 Feb 23
🇺🇸 U.S. Politics Abortion Polling Presidential elections
  1. Majority of Americans, including 40% of Republicans, support legal abortions
  2. Americans prefer government compromise over gridlock, especially regarding debt-ceiling decisions
  3. Historical evidence shows voters have historically punished presidents for poor economic performance

Artificial intelligence and "big data" cannot replace public opinion polls | #213 - April 10, 2023

117 implied HN points • 10 Apr 23
🇺🇸 U.S. Politics Polling Public Opinion Data science Political Analysis Social Trends
  1. Artificial intelligence and big data cannot fully replace public opinion polls, as they rely on polls for calibration and may not be as reliable for all groups.
  2. Changes in polling methods, like switching from phone to online surveys, can impact results, highlighting the importance of consistency over time.
  3. Studies show genuine change in attitudes, like increasing racial liberalism, but also caution against biases affecting survey responses.

If you’ve ever asked me for job advice, this post is for you

78 implied HN points • 15 Aug 20
  1. The author receives job advice requests despite being young and early in their career.
  2. The author published a post with job advice for upcoming data journalists and analysts.
  3. The author's intention is to help others improve their prospects in the field.

My annual holiday subscription special!

19 implied HN points • 04 Dec 20
  1. Annual holiday subscription special with a 50% discount for new subscriptions made before January 1st, 2021.
  2. Subscribers get access to extra posts, making a huge difference to the writer and supporting their work.
  3. Various premium articles and content available in the archive for paying subscribers.

Free trials of subscriber-only content!

0 implied HN points • 07 Oct 19
  1. You can sign up for free trials of subscriber-only content until November 1st, 2019.
  2. Check out the link for a 14-day free trial of the newsletter.
  3. Subscription includes access to exclusive content, like the recent email on impeachment.