Some surprising tea party numbers

[Alcibiades]

I'm working on a little project about the tea party movement, so I wanted some numbers on tea party strength in the various states, but couldn't find any good data. The best I could find was a wikipedia list of the various tea party protests

http://en.wikipedia.org/wiki/List_of_Tea_Party_protests,_2010

My initial hypothesis was that this is basically a regional movement. You see these folks with signs that basically extol the same ideas held by southern politicians from John C. Calhoun to George Wallace, so I had the idea that this was some sort of dead-end movement that arose in reaction to the election of Barack Obama, with the idea that they would take to the streets not simply as the result of the election of the first black man to the presidency (though that's certainly true), but that the legacy of southern culture has created an ideology antithetical to the ideals of the modern Democratic Party, and that they would pretty much have taken to the streets in response to any Democrat. I grouped the states into three groups: slave states in 1860, free states in 1860, and territories, also including AK and HI, even though these last two were not even territories in 1860 (they are empty cells anyway). This would be interesting for many reasons, not the least of which being that we don't usually expect temporally remote independent variables to exert much influence on the dependent variable. The competing hypothesis is the one that seems to be assumed by the MSM: that the TPM is a broad-based movement disaffected by politics as usual, and that there is no regional bias in its activities.

I found I needed a new hypothesis, that my hypothesis and the null hypothesis are incorrect on their face.


Here are the sums of the reported tea party protests in 2009 by state. Since I wanted a measure that does not simply measure actual tea party protests, which is not available in any event, but the attendance reported by the TPM, I used the highest end estimates (which were wildly overblown in many, many cases). In those instances where no number was provided or some other descriptor was provided (i.e., "small"), I just used 50. Whenever the number was simply given as "thousands," I used 5,000. The dependent variable here is not really actual tea party movement attendance, so much as it is the attendance in the most fevered imaginings of the tea party movement.


Slave states
TX 43940
MO 2700
AR 0
MS 550
AL 250
GA 14070
FL 27123
TN 11650
SC 4925
NC 1835
VA 1115
WV 80
MD 2850
DE 0
DC 80050
LA 700
KY 5500

Free states
CA 22300
IA 0
WI 15000
IL 4700
MI 1250
IN 2740
OH 57655
NY 22600
VT 0
ME 0
NH 0
RI 1000
MA 500
CT 3000
NJ 3500
PA 14075
MN 2100

Territories
WA 8250
ID 5000
MT 275
NV 1000
UT 100
NM 0
AZ 13500
OK 10060
NB 340
KS 700
ND 1500
SD 3000
CO 5550
WY 0
HI 25
OR 1000
AK 0
total 398058

Notice anything? Obviously, both though my original hypothesis and the media hypothesis seem disconfirmed, though mine may still look pretty good when I convert these numbers into a per capita basis, which I will do momentarily.

I have an idea, which I will post later. I want to see what you think caused differences in reported tea party attendance in 2009.

[Warpy]

1. The slave state: fee state hypothesis mostly holds up

and the major outliers are Ohio, where the recession has hit especially hard and Arkansas, where it's hit but not nearly as hard. Maryland, Delaware, and DC were all free, by the way, and DC is not a state but a city full of Republican appointees and lobbyists.

I'm delighted to see my own state of NM listed as 0 members. They tried to get it going here but it fizzled after a few acrimonious house meetings and attempts at more public meetings in local bar and grill places. NM was considered a "free" territory and in fact drubbed slave state Texas at the Battle of Glorieta Pass during the Civil War.

Thanks for assembling what data were available.

[Alcibiades]

2. I'm not done yet

This is claimed attendance, not actual membership. I actually would expect that the membership would be very different from attendance, if there were such a thing as membership.

DC was slave in 1860, but slavery was abolished there in 1862. The cut point of 1860 is admittedly rather arbitrary, though DC won't factor much in the final analysis, since I'm assuming the (claimed) 80,000 people who came to tea party events there in 2009 were not from DC.

I get your point about the economy, but if it's responsible for Tea Party turnout in Ohio, then why not Michigan? The macroeconomic conditions are similar, but why is Ohio such an outlier?

[Warpy]

8. Ohio saw much more immigration from Appalachia

than Michigan did, although both states did see it.

Personally, I think another factor might be the number and degree of viciousness of right wing hate radio stations in any given area.

I didn't hear much of it in Arkansas during my last drive across the country. I heard a hell of a lot of it in Oklahoma and most of Dixie.

[Alcibiades]

10. Low turnout in AR fits with the amended theory

It's fairly small in the electoral college, and would probably not vote Obama in 2012.

[Alcibiades]

3. Here is the attendance per capita, using 2009 population

We find that it was claimed that one out of every 564 Texans attended a tea party event in 2009, versus, say, one out of every 6417 Louisianans. (Not exactly this, of course, as we cannot really say that these folks all came from the states in which the events were held, and it is quite likely that many of these folks are repeat protesters, anyway. Since the numbers are so phoney baloney anyway, though, it hardly makes a difference).

TX 564.003231679563
MO 2217.62222222222
AR ERR
MS 5367.26545454546
AL 18834.832
GA 698.593532338309
FL 683.477823249641
TN 540.450987124464
SC 926.140507614213
NC 5112.19836512262
VA 7069.58744394619
WV 22747.2125
MD 1999.81684210526
DE ERR
DC 7.49103060587133
LA 6417.25142857143
KY 784.384181818182
CA 1657.47372197309
IA ERR
WI 376.984933333333
IL 2746.89553191489
MI 7975.7816
IN 2344.20182481752
OH 200.20197727864
NY 864.666061946903
VT ERR
ME ERR
NH ERR
RI 1053.209
MA 13187.174
CT 1172.76266666667
NJ 2487.92542857143
PA 895.542948490231
MN 2507.72095238095
WA 807.781212121212
ID 309.1602
MT 3545.41454545455
NV 2643.085
UT 27845.72
NM ERR
AZ 488.576148148148
OK 366.505964214712
NB 5284.17352941177
KS 4026.78142857143
ND 431.229333333333
SD 270.794333333333
CO 905.36
WY ERR
HI 51807.12
OR 3825.657
AK ERR
total 771.260846409317 (US population/total claimed attendance)

[valerief]

4. Do your per capita numbers exclude certain groups like children or the incarcerated?

Not sure what's available to you.

[Alcibiades]

6. It's the Census Bureau population estimate

So it's total population of persons.

[valerief]

7. Oh, because I was thinking that non-voters could be excluded and that

would make the TPM claims even more improbable.

[Alcibiades]

9. It's not even to prove that their estimates are improbable

According to these figures, all highball numbers, less than 400k protested in 2009, not even accounting for the fact that many of these events were attended by the same people over and over again (if you have ever been to any sort of political rally you know what I mean). What I was trying to see if this was purely a white, southern movement. It's not, but it would seem that there is very high turnout in some states, states that only make sense if we assume it is the result of national mobilization.

If I had good numbers, I would expect a high correlation between turnout in places such as FL and OH and spending by groups such as Freedomworks. It's a complex hypothesis, I know, but the Tea Party Movement is both a grassroots movement and a national GOP astroturf movement.

[valerief]

11. I see! Thanks. nt

[Alcibiades]

5. Here's the new hypothesis

Claimed Tea Party attendance is a function of electoral votes and how close the race was in 2008. States that had a close race in 2008 would have higher attendance, states that were not close would not. States that are too small to bother with also would not.

In other words, the Tea Party events in 2009 and 2010 were a GOTV effort on the part of the GOP. They didn't have to spend a lot of resources to get good attendance in places like Texas and Tennessee, but national efforts to bolster the movement were centered on Florida and Ohio. This does say some interesting things:

* Attendance in North Carolina was low because there was no national effort to mobilize. This says that the national GOP thinks 2008 was a fluke, and that they don't have to do anything in NC.

* They think Wisconsin is in play? The number in WI may be more fake than others, based on one rally that the Sheriff of Milwaukee spoke at, when he was also the person who gave the number I used, which was derided by many there.

[Alcibiades]

12. 54.5% of protesters only in TX FL OH NY CA

These states account for 173,618 attendees (not counting the DC protests, which presumably drew from a bigger pool than nay of the state events).

[Roland99]

13. Kick