Profile picture
Charles Franklin @PollsAndVotes
, 9 tweets, 3 min read Read on Twitter
I posted a thread on trends in party ID earlier today, linked below.

Now let’s do a little self-critique. What issues remain?

1. I used Gallup. You mileage might vary with a different pollster. Gallup has the longest history and a stable question, so good. But 1 pollster. /1
2. I used the 2004-2018 time frame. That’s pretty long but what would an even longer timeframe look like? Note the debate IS about Trump’s effect on PID now, so 2017-18 is the critical time, but a longer term perspective might be good for appreciating PID change in general. /2
3. I’m looking at aggregate change but what about individuals? Pew and CBS have great looks at this, as does Bartels. My read of their work is modest change and pretty equal for D and R. Most is in and out of Ind to lean, or lean to party.
Bartels paper:
vanderbilt.edu/csdi/includes/…
4. There is a technical argument over how much is “a lot”. I say a move of 3-4 points isn’t much but some point out that is 10-15% of the BASE level of PID (eg 3ppt change on a 30% base is a “10%” change.) /4
5. Maybe we don’t have enough data. Gallup is releasing monthly PID estimates, which I am using. But that is just 6 estimates so far in 2018, so we are looking at changes on a small N and at the end of a series when trends can be misled by outliers. /5
6. Related, the trend estimator I use, loess, can be sensitive to how much smoothing you do. I use a moderately sensitive estimate, not so smooth. More (or less) smoothing can affect trend estimates, especially at the end. /6
7. Maybe the *structure* of partisanship is shifting even if the aggregate margins are stable. Good point. I’m working on that. /7
8. There are serious people who have different views on this and they should be taken seriously. In the academic world, at least, these discussions have been pleasantly professional with constructive disagreements, which I have much appreciated. /8
9. If you want to go down this rabbit hole there are good reads:
Individual change: people-press.org/2017/05/17/par…

cbsnews.com/news/2017s-big…

vanderbilt.edu/csdi/includes/…

Excellent Original on GOP decline:
zacharypeskowitz.com/PAR_Writeup.pdf

/end
Missing some Tweet in this thread?
You can try to force a refresh.

Like this thread? Get email updates or save it to PDF!

Subscribe to Charles Franklin
Profile picture

Get real-time email alerts when new unrolls are available from this author!

This content may be removed anytime!

Twitter may remove this content at anytime, convert it as a PDF, save and print for later use!

Try unrolling a thread yourself!

how to unroll video

1) Follow Thread Reader App on Twitter so you can easily mention us!

2) Go to a Twitter thread (series of Tweets by the same owner) and mention us with a keyword "unroll" @threadreaderapp unroll

You can practice here first or read more on our help page!

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just three indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member and get exclusive features!

Premium member ($3.00/month or $30.00/year)

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!