Political scientist ~ Newhouse Professor of Civic Studies at Tufts ~ C(C)ES Whisperer ~ #COYS #GoDawgs
Somerville, MA
- Tweets 10,407
- Following 1,234
- Followers 8,311
- Likes 9,985
Pinned Tweet
🚨🚨🚨 The Cooperative Election Study's 2024 vote validation data is now available for download!
Microdata here: doi.org/10.7910/DVN/X11EP6
Or easy interactive analysis of the data on...
vote choice: cooperativeelectionstudy.shi…
demographics of voters: cooperativeelectionstudy.shi…
1
21
70
17,664
just released: national estimates from CES pre-election interviews with over 70k american adults
National LVs (Oct. 1 - 25):
Harris: 51%
Trump: 47%
Dynamic crosstabs: cooperativeelectionstudy.shi…
Release: sites.tufts.edu/cooperativee…
113
442
2,311
1,377,314
CES likely voter estimates for 5 key swing states:
FL (N=3,755): Biden 49 - Trump 47
GA (N=1,456): Biden 48 - Trump 47
NC (N=1,627): Biden 49 - Trump 45
PA (N=2,703): Biden 52 - Trump 44
TX (N=2,947): Trump 49 - Biden 47
Details and data tool:
bfschaffner.shinyapps.io/ces…
120
263
1,368
we now have estimates for key states
AZ: Trump 51/Harris 47
GA: Trump 51/Harris 46
MI: Harris 51/Trump 46
NV: Harris 51/Trump 47
NC: Trump 50/Harris 48
PA: Harris 49/Trump 48
TX: Trump 51/Harris 47
WI: Harris 50/Trump 47
Full results here: cooperativeelectionstudy.shi…
just released: national estimates from CES pre-election interviews with over 70k american adults
National LVs (Oct. 1 - 25):
Harris: 51%
Trump: 47%
Dynamic crosstabs: cooperativeelectionstudy.shi…
Release: sites.tufts.edu/cooperativee…
149
219
1,350
1,031,332
always fun to see just how much people misreport changes in their own economic situations.
Republicans whose incomes had increased by $40k or more during the past 2 years were still more likely to tell us that their income had decreased rather than increased
42
281
1,072
338,705
If your polling error take (1) relies on exit polls as a measure of “truth” and (2) can’t explain why the polling errors also existed in senate and house races then please start over.
9
83
998
we’ve been giving 2020 a hard time, but let’s just take a moment to appreciate the fact that the calendar is bringing Election Day almost as early as it could
10
48
803
Interestingly, our estimate of the national vote margin is not too sensitive to our likely voter model (Biden's margin is pretty robust to different turnout assumptions).
But this is very much not the case in Texas, where turnout will clearly make a big difference.
34
131
745
If you keep waiting for Republican women to break from supporting Kavanaugh then this chart is for you. Republican women score nearly as high in hostile sexism as Republican men. And their sexism has increased since Trump's election.
53
369
618
Trump support was partly about economic insecurity, but even moreso about race and gender.
20
507
653
Trump approval by birth year:
1) interesting gap between millennial RVs and all millennial adults
2) gen X is basically just a transition
43
195
571
You know that whole urban/rural divide in American politics? Turns out that when you control for peoples' levels of racial resentment and hostile sexism then it almost entirely disappears. So maybe we should just call a spade a spade?
26
211
504
WI: 9% of Sanders voters voted for Trump.
MI: 8% of Sanders voters voted for Trump.
PA: 16% of Sanders voters voted for Trump.
3/n
103
349
444
the relationship between partisanship and google searches for "hand sanitizer" is tapering now that Trump and Republicans are finally acknowledging the seriousness of the threat
23
177
501
Nationally, among those we can validate as voting in a prez primary and the general election, 12% of Bernie Sanders voters voted for Trump
68
321
387
If you use @pollsterpolls database & pick the SINGLE WORST poll for Clinton over past few weeks & use it to assign the state, she still wins
25
279
406
Some asked for more detail on how Sanders primary voters behaved in general. This graphic shows this, including small % who abstained 2/n
67
230
354
Students: whatever effect you expect to produce from having your mom email me, multiply it by negative 2.
16
23
359
It's here! Preliminary results from the 2020 C(C)ES pre-election survey (N=71,789 adults)
Explore here: bfschaffner.shinyapps.io/CES…
Use in your classes! You can download any graph as a .png by hovering over the plot and clicking the camera icon or share the link with your students!
36
85
380
partisanship is a strong predictor of being anti-vaccine, but much less so when the risks are greater
26
86
351
But 12% defection rate among Sanders voters was higher than I expected. (Which is why I waited for the validation before I believed it.) 5/5
19
89
248
But there does appear to be a racial component to this, as defectors are much more likely to disagree that whites are advantaged in US 9/n
29
160
259
I suspect that @Nike has a good sense of where their target demographic is on the Anthem Protests.
9
83
295
Just posted: "Explaining White Polarization in the 2016 Vote for President: The Sobering Role of Racism and Sexism" people.umass.edu/schaffne/sc…
17
243
248
Most white Democrats already acknowledged that whites have advantages in the U.S., but a lot of independents and Republicans appear to be joining them.
But let's see if it lasts...
dataforprogress.org/blog/202…
8
67
244
Looked at one other thing today -- general election presidential vote based on primary vote, among validated primary & general voters 1/n
14
106
216
I think this is such an under-studied explanation for low turnout. The norm of civic duty is not just to be a voter, but to be an *informed* voter. And young Americans are especially unlikely to realize just how well they stack up to others when it comes to being informed enough.
Since we're talking about youth voting in midterms -- this is four years old, but the results on one question ("Who should vote?") both surprised me and felt like they explained a lot. huffingtonpost.com/2014/11/2…
10
85
238
Trump's Republican Party is hemorrhaging the younger voters that supported them in 2012 and it will be a challenge to get them back. My latest @nytimes piece with @SeanMcElwee, @JesseRhodesPS and @blfraga. nytimes.com/2019/02/16/opini…
12
80
211
In my #MPSA18 paper, I show that exposing white respondents to Trump's quote about Mexicans causes them to say more offensive things not only about Mexicans, but also about other groups as well. umass.box.com/s/x5zz210nor2z…
9
129
222
Trump lost some support, too, of course. 34% of Kasich primary voters voted for Clinton. And 11% of Rubio primary voters did the same. 4/n
14
66
191
she was always gonna wait one day longer than pete, wasn't she?
BREAKING: @amyklobuchar is dropping out of the presidential race, will fly to Dallas tonight to join Biden and endorse
5
6
235
an update on the how the relationship between partisanship and google search traffic for "hand sanitizer" evolved across US media markets during the past couple of weeks
(and my first foray into using gganimate)
more on data and code here: sites.google.com/view/brianf…
10
63
238
Democrats would be wise to focus much more attention on the Obama voters who stayed home in 2016 rather than the ones that switched to Trump, as @SeanMcElwee, @JesseRhodesPS, @blfraga and I argue. nytimes.com/2018/03/10/opini…
12
81
203
Many Republicans embraced Trump's divisive rhetoric after his victory, but in my preliminary analysis of 2018 vote patterns for @DataProgress, I find that GOP House candidates were punished for the party's sexist rhetoric in a way they weren't in '16. wthh.dataforprogress.org/blo…
9
68
184
However, also worth noting partisanship of these voters. This chart shows Bernie -> Trump voters are much less likely to be Dems 6/n
16
97
164
in 2020, the urban-rural divide in presidential vote choice was 32 points. but after we controlled for denial of racism, that same gap dropped to just 8 points.
A Tufts Public Opinion Lab collaboration with (now alumni) @zacharylhertz and @Lucasbenji17
washingtonpost.com/politics/…
4
50
173
in an op-ed appearing in tomorrow's @nytimes, Jesse Rhodes, @raylaraja, and I explore how African Americans receive even worse local political representation in suburbs and small towns as they do in major cities. nytimes.com/2020/08/23/opini…
5
66
170
just a note to say that racist and sexist voters were about as likely to support Biden (or any other Democrat) as they are Harris...which is to say not very likely at all
9
31
155
30,183
Since people have raised the issue: defection rate is same for registered Ds, so this may not be a closed vs open primary story.
6
35
150
Ran a hostile sexism battery on recent NH poll. Here is the effect of sexism on vote for Romney in 2012 & Trump in 2016. Big difference.
24
194
159
We asked Trump voters which photo has more people...15% picked photo A. #alternativefacts washingtonpost.com/news/monk…
19
152
161
Just wrapped up my #APSA2018 paper which investigates how Trump (and his victory in 2016) has affected expressions of sexism among Republicans. Republicans are more willing to support sexist views when they are attributed to Trump & since his election 1/n tufts.box.com/s/wrth10sv6yv7…
5
69
154
can't tweet this out frequently enough...even 40% of Republicans want to ban assault weapons
3
57
145
last cut at this. you can see pretty clearly how responses to Covid19 were quite partisan until Trump/Republicans/Fox News started taking it more seriously.
final code can be found here (incl. models controlling for covid19 cases): sites.google.com/view/brianf…
7
66
157
Patterns of white voting based on education over past few decades are striking. First, separation in 2000, then extreme polarization in 2016
16
154
161
not too surprising, but the single most important predictor of whether someone was vaccinated in the 2021 CES was their 2020 presidential vote (followed by age and education)
9
52
148
We've created a new interactive tool that uses the Cooperative Election Study data to allow you to plot racial resentment and hostile sexism attitudes across different subgroups in the American adult population. cooperativeelectionstudy.shi…
5
37
155
my newest @goodauth piece with @carolinelsoler documents how pollsters are weighting polls differently now compared to 2016 and what that means for numbers they are putting out there goodauthority.org/news/polls…
14
64
157
58,062
I've been scooped!
But a good time to announce that I'm excited to be joining @TuftsPoliSci and @TischCollege as the Newhouse Professor in Civic Studies this fall. Though I'm sad to be leaving my great colleagues @umasspolsci behind, eager to see how this new chapter unfolds!
A big Tufts Jumbo welcome to Professor Brian Schaffner, inaugural holder of Newhouse Chair in Civic Studies, joint Political Science and Tisch College @b_schaffner #mapoli @debbiejsr @eitanhersh @IkeEichenberg #Tufts @peterlevine @JenMcAndrew @TischCollege
34
2
140
We randomize what year we tell white Americans that the U.S. will become a majority-minority nation. Main finding: it doesn't matter to whites *when* this change will occur, just that it will occur.
3
59
133
Honored to have been awarded an Andrew Carnegie Fellowship to support my book project with Stephen Ansolabehere - "American Mosaic" - about the social identities that define our politics and a path to bridge the divide. #CarnegieFellows carnegie.org/awards/honoree/…
17
3
146
10,666
🚨🚨 happy to announce the full release of the 2022 Cooperative Election Study (CES) survey data. interviews with 60,000 American adults conducted by @YouGovAmerica before and after the 2022 election and matched to voter files. lots to explore so dig in! dataverse.harvard.edu/datase…
39
152
37,731
this should go without saying, but a candidate's performance in a state's primary tells you basically nothing about how they will perform in that state in november
2
45
130
by request (cc: @SpencerPiston), Trump approval among registered voters by year of birth and race/ethnicity. a gulf between white boomers and millennial whites.
14
55
127
Here is a graphic from an experiment I ran on the 2016 CCES. Republicans would be much more likely to find the sexist things Trump has said objectionable if somebody else said them.
7
90
130
Just a reminder about who the @NFL is pandering to with its new rule fining teams for anthem protests. washingtonpost.com/news/monk……
7
76
137
a lot of posts on here suggesting that the narrow distribution of polling results is evidence for herding.
herding is prob a thing, but also worth noting that weighting (esp. to past vote/partisanship) will significantly narrow the sampling distribution even w/ unbiased samples
7
31
137
179,989
my paper on the increasing importance of racism & sexism in the 2018 midterm elections is now forthcoming at BJPS. main takeaway: D House candidates gained votes among those with low levels of racism denial/hostile sexism without losing anything from those on high end of scales.
4
45
119
i just noticed that @FiveThirtyEight gives @pewresearch's polls a B/C rating even while Emerson is an A- and I'm assuming these ratings are just coming from a random letter generator
3
5
128
No time to answer every request for more info, but data are here: dataverse.harvard.edu/datase…
And my code here: dropbox.com/s/jqd1lm4tfu91hb…
10/10
8
39
112
how (validated) primary voters compare ideologically to other partisans over time. notable how similar the groups look on the Democratic side.
9
36
116
Along with my co-PIs Steve Ansolabehere and @scluks, I'm happy to announce the release of the preliminary data files for the 2020 Cooperative Election Study, featuring data from 61,000 American adults interviewed before and after the 2020 election.
dataverse.harvard.edu/datase…
2
30
123
For those suggesting this is about trade, note that opposition to TPP is not dramatically different among defectors. 8/n
12
55
96
Barreto rightly notes that what drove vote choice among non-college whites had little to do w/ economics #PA2017 people.umass.edu/schaffne/sc…
4
81
109
some data on potential VP picks from recent CES surveys. 🧵
tldr: Beshear, Shapiro and Cooper have all demonstrated the ability to position themselves as more moderate than the party and win over independents and even some republicans
(h/t to @carolinelsoler for the graphs!)
6
29
114
91,455
“young people are stupid as hell” 🤔
Hopefully, the @kanyewest crap is a publicity stunt. He doesn't need ballot access- he's a celeb w mad $ & @KimKardashian, so he poses a threat to Biden as a write-in bc young people are stupid as hell. Not a lot will do it- but it wouldn't take much.
Hollywood should shame them
3
8
101
Happy to announce that the 2018 CCES Common Content dataset is now posted on @dataverseorg. This is a preliminary release of data for a nationally representative sample of 60,000 American adults interviewed around the 2018 midterm elections. Enjoy! dataverse.harvard.edu/datase…
4
30
106
We often don't spend enough time on description in political science. A conversation at my #APSA2018 panel yesterday reminded me of a descriptive finding that I don't show enough -- how men and women compare on hostile sexism across educational attainment. Data: 2016 CCES.
6
55
103
Here is our pre-election release of national House vote results from the 2022 Cooperative Election Study survey. This is based on interviews conducted by @YouGovAmerica with over 60k American adults during the past month. We show Democrats +2 among LVs.
cooperativeelectionstudy.shi…
4
26
101
hard to imagine it doesn’t immediately become more popular if only because democrats will become much more supportive now that they are no longer getting mixed signals from democratic leaders
2
23
101
"This tool is broken, it won't let me tax wealth at a rate higher than 5%!"
- @SeanMcElwee, any minute now
2
1
92
🚨🚨 announcing the preliminary release of the 2022 Cooperative Election Study (CES) survey data. interviews with 60,000 American adults conducted by @YouGovAmerica before and after the 2022 election. lots to explore so dig in! dataverse.harvard.edu/datase…
1
26
102
38,084
Could increased levels of political knowledge among mathematicians put an end to naive stories like this? Political scientists say 'Yes' nitter.app/gabbybirkman/status/88…
3
24
92
Happy to announce that my new @CUP_PoliSci Elements book was published today.
I explore how Trump's rhetoric encourages some Americans to express more prejudice than they would otherwise. You can access it online for free until Nov 2 & here's a thread...
cambridge.org/core/elements/…
5
26
91
Need to talk about the #Midterms2018 in your classes, with your colleagues, or just around the dinner table? Come armed with evidence from the CCES! I'm sharing the graphics I posted on Twitter earlier this week in one easy to download slide deck. Enjoy! docs.google.com/presentation…
12
36
89
What we can learn from the Obama-Trump voters that Democrats won back in 2018, and what it means for the party’s strategy in 2020. My latest with @SeanMcElwee nytimes.com/2020/01/06/opini…
10
34
100
I just added age by gender to the 2020 CES analysis tool. Why? Because my mother-in-law requested it.
bfschaffner.shinyapps.io/CES…
12
3
86
as of today, you can now order Hometown Inequality! I recommend ordering this many amazon.com/gp/aw/d/110872537…
9
5
89
Stereotypes about Muslims were strongly associated with support for Trump in 2016. Sadly, EO likely appeals to his strongest supporters. 1/x
10
83
86
you can now get our new book, Hometown Inequality, on your kindle. or pre-order your paperback copy for delivery next month.
amazon.com/dp/B08BKXHNKQ/ref…
9
21
82
We just published the final 2020 CES dataset, which includes vote validation and full guide to the data.
The data includes 61,000 respondents (39,198 validated voters), including 51,551 who responded to both the pre- and post-election questionnaire.
dataverse.harvard.edu/datase…
3
20
97
Just released! The 2016 CCES Common Content pre/post election data. Massive sample of 64,600 American adults dx.doi.org/10.7910/DVN/GDF6Z…
3
62
87
One note on '08: Data suggests 9% of Obama voters defected that year. In \16, 97% of Clinton prim voters also voted for her in general. 4/n
3
31
74
me and @carolinelsoler wrote a thing for @goodauth on how the noise about black voters abandoning the democratic party is mostly a mirage generated by the survey responses of black adults who don't vote.
goodauthority.org/news/are-b…
2
33
93
23,961
Slide from talk I'm giving at NYU Friday about the 2016 election. More evidence that sexism/racism (not economics) cost Clinton votes.
4
57
85
This plot shows primary voters, partisans who aren't primary voters, and true Independents on an ideological scale based on their positions on 36 issue questions from the 2016 CCES. Key point: primary voters are actually quite ideologically diverse...
9
54
86
Results from a nationally representative election poll fielded by our @TuftsUniversity Polling the 2020 Election seminar.
Nat'l Likely voters (10/20):
Biden 52%
Trump 45%
Interactive crosstabs: bfschaffner.shinyapps.io/tuf…
Story: tischcollege.tufts.edu/tufts…
Key findings in the THREAD
9
25
80
happy to announce that in 2024 the CES will not only be validating turnout for all 60k respondents, but will also include validation of college degrees (from college transcripts), income (from tax returns), and age (from birth records). excited to see how people use this data!
9
3
89
12,232
reminder that this election is a toss up largely because Democratic party elites were strong/willing enough to push their unpopular candidate out of the race while Republican elites have been unable to rid themselves of their unpopular candidate for 3 consecutive cycles
8
8
83
12,552
sharing a couple of graphs from this week's public opinion lecture...basically, with all that happened during the past four years, almost nobody changed their partisanship
8
20
79
My Tufts PS119 ("Polling the 2018 Elections") class conducted a pre-election survey of American adults. I'm going to use this thread to Tweet out some of the interesting results they have found so far. First off, here is our topline. 1/x
6
26
91
i usually use "hair color" as an example to explain what a nominal variable is, but only this week did i finally realize that i could use my bald head to add an explanation of what missing data is
4
93
7,401
working on my #SPSA2020 paper and can't stop staring at this graph, which shows effect of a 1 std dev increase in hostile or benevolent sexism on a person's rating (0 to 100 favorability scale) of Warren, Harris, Sanders, Biden, Romney, & Trump (controlling for party/ideology)
6
34
83