search

TheEconomist / us-potus-model

Code for a dynamic multilevel Bayesian model to predict US presidential elections. Written in R and Stan.
https://projects.economist.com/us-2020-forecast/president
MIT License
1.24k stars 188 forks source link

Incorrect data in poll sheet #8

Closed nicholasbaileyrally closed 4 years ago

nicholasbaileyrally commented 4 years ago

Hi there.

I noticed today that the downloadable sheet you have of current polling data on the economist website has incorrect values for the Biden vote share on the latest RMG poll. That poll has a 46% vote share for Biden, not the 48% in your data. I don't know how much that will adjust your model - I assume mostly a wiggle since it's one poll, but wanted to alert you to the error.

If this is a deliberate adjustment to weight a partisan poll or something, ignore this issue!

elliottmorris commented 4 years ago

Thank you. Issue resolved. On Jul 13, 2020, 15:01 -0400, Nicholas Bailey notifications@github.com, wrote:

Hi there. I noticed today that the downloadable sheet you have of current polling data on the economist website has incorrect values for the Biden vote share on the latest RMG poll. That poll has a 46% vote share for Biden, not the 48% in your data. I don't know how much that will adjust your model - I assume mostly a wiggle since it's one poll, but wanted to alert you to the error. If this is a deliberate adjustment to weight a partisan poll or something, ignore this issue! — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.

nicholasbaileyrally commented 4 years ago

No thank you for your awesome modeling work!

tomsc87 commented 4 years ago

Similar problem with the latest Redfield & Wilton poll. Trump got 40% (not 39%), there were 2,000 respondents (not 1,853), and the respondents were registered voters (not likely voters).

elliottmorris commented 4 years ago

We must be looking at different toplines then. On Jul 14, 2020, 08:04 -0400, tomsc87 notifications@github.com, wrote:

Similar problem with the latest Redfield & Wilton poll. Trump got 40% (not 39%) (meaning also that undecided/don't know was 8%, not 9%), there were 2,000 respondents (not 1,853), and the respondents were registered voters (not likely voters). — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub, or unsubscribe.

tomsc87 commented 4 years ago

I'm looking at the source you provided. Nowhere on the Redfield & Wilton page does it mention 1,853 (with or without the thousands separator).

Source Report

elliottmorris commented 4 years ago

That may be my fault. I think their full numbers (with Kanye West) are at a different link. https://twitter.com/ppollingnumbers/status/1282768444484734978?s=21 On Jul 14, 2020, 11:48 -0400, tomsc87 notifications@github.com, wrote:

I'm looking at the source you provided. Nowhere on the Redfield & Wilton page does it mention 1,853 (with or without the thousands separator).

— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub, or unsubscribe.