An open letter to Erick Erickon over at RedState, bless their little hearts.
Dear Erick,
You moron.
I'm trying to figure it out, this conservative meme surrounding the supposed "skewing" of the polls to advantage Democrats. I actually understand polling and have a research job, and I understand that most Americans do not. Scientific polling can be counter-intuitive and complicated to the layman -- "what do you mean they only asked a handful of people and can now predict what others they did not ask would have said, if they had?" -- so I get that the average conservative might be confused.
But I'm really struggling to understand why you and others seem to think it serves the larger conservative movement -- if there is such a thing -- to embrace this complexity and throw its collective hands up in mock frustration.
In fact, I'm not sure it does. I think you're blowing it, Erick.
I think this is going to come back to bite you.
Your recent RedState article Media Malpractice lays the case out neatly enough. In fairness, you haven't been the most hysterical on this issue, and you "accept that Romney is a bit behind" broadly speaking (which passes for moderation today, I guess), yet you still insist on joining the pearl-clutchers convinced their man is doing better than we say he is based on some sort of junk science poll-bashing nonsense.
To do this, you single out a single Washington Post/ABC poll that characterizes the national race as "tight" but goes on to correctly note that:
The new survey also highlights an emerging dynamic: the disparity between the state of the race nationally and in battleground states, where campaigning and advertising by the two candidates have been most intense and where the election will be decided.
Now to review: right now President Obama leads in national polling by 2-3 points, yet if the election were held today he would likely win the Electoral College by over 140 EVs. That's how we decide elections -- not a national popular vote -- so the distinction is critical (as you know). So the race is both tight, and not tight.
But back to how you're being a moron.
Despite the fact that there is ample other polling being done expressly to gauge swing state voter preference, WaPo oddly chose to slide and dice its own national poll as it pertains to swing states, and then draw conclusions from it. That's weird. You sum up your issue with this thusly, and actually make a fair point:
In short, the Washington Post conducted a national poll and discovered that 161 of those surveyed lived in swing states. Charitably — and Cohen does not break it down — that would work out to 20 voters polled in each of 8 swing states. TWENTY!!! So there’s an 11 point gap among 161 people with a margin of error of around 8%. But the Washington Post reported this as a fact and only subsequently admitted
That margin is significant at the 80 percent confidence level, not a standard, conservative 95 percent threshold, which we take as added evidence of the state-of-play in state polling, and the crucial link between those and the national numbers (49 percent for Obama, 47 percent for Romney). No such dissection of a national poll — no matter how many interviews — could be anything more than that. They are simply not designed that way.
So two things:
First, yeah -- there's no doubt that this dissecting of data is poor statistical analysis and that, taken alone, the lower confidence level assertion of Obama's swing state lead would be highly suspect. It's crappy work, and they probably know it.
But second, critically, here's the thing -- dozens of other polls have found the same thing. Obama is ahead in the swing states -- all of them. That's what the best data analysis tells us.
If you want to harp on WaPo for trying to reach a state-level conclusion using a national-level tool not designed to inform state-level sentiment, that's cool. Do that. Get all technical.
But you use this fair critique to then jump on the "pollsters are in bed with the Democrats!" meme -- which 1) muddies your entire point, and 2) is dumb:
Most conservatives view most members of the press as willing collaborators with the Obama Administration. Many independents also distrust the media. It is stuff like this that fuels the rage against a press that routinely refuses to acknowledge that it really does, collectively, lean left of center.
There. You see that? You just lost it. You just took an easy-to-assail example of poor data analysis and tried to use it to fit a misguided and wrong-headed predetermined view of the situation.
This wasn't a bad poll -- the methodology is sound, it's not an outlier, and WaPo has done this for a long time. And the swing state findings -- however oddly reached -- aren't outliers either -- they fit with polling done over the last few weeks by many others. So you're hysterical for nothing.
Now, RedState is for adolescents and Erick Erickson is a hack, you say, so why should any of this surprise anyone or matter at all?
Because:
After the election is over, and President Obama has been soundly reelected, we're going to have a national conversation about What Just Happened. Did Mitt just suck? Did a few too many jobs get created? Was it all a communist plot? And so on. We're going to have this conversation in coffee shops and around water coolers and on opinion pages and at Thanksgiving dinner tables and on Facebook.
And every conservative who believes the completely legitimate election was in fact illegitimate because of faulty polling will be disqualified from believably participating in this conversation. They will instead be ridiculed. And rightly so.
Your work buddy who insists that if only those bullshit polls hadn't discouraged so many voters, Romney would have won out -- he's disqualified.
Your brother-in-law who tells you that the rumors of Mickey Mouse voting for Obama 200 times in Volusia County, and if only they'd had better voter ID laws Romney would have won -- he's disqualified.
Your tea partying neighbor who yells about some thing he heard on Rush about how they oversampled Democrats and if only the entire media wasn't liberals from Harvard we could be a proud nation again that didn't apologize for itself and wasn't run by Muslims and gays -- that dude is definitely disqualified.
The people who insist that bad/skewed/coordinated polling caused this result, rather than substantive issues, will lose any standing they still have in whatever communities they communicate. Pollsters are just not going to credibly be blamed for this. They just aren't.
In short: trying to hang your hat on discrediting the entire concept of scientific polling as a post-election believability strategy? That's not going to work.
By feeding your followers this stuff, you're setting them up for a very awkward and uncomfortable future. They have it bad enough as it is, trying to explain the Ryan tax plan (which Ryan himself can't even explain) and how global warming is a myth and how we're going to drill our way to energy independence and how immigrants are destroying the nation and how vaginas kill sperm they don't like, etc. Already you've given them so many odd positions to try to defend. So many areas where common sense works against them. Where the truth shows its liberal slip.
But this is going to be a bridge too far.
It's one thing to flail around and try to blame others for your party's mistakes. It's one thing to try to disown the reasons you're actually losing -- distrust of women and minorities, unpopular social positions, anti-science thinking, unrealistic economic ideas, unwavering support of the rich at the expense of everyone else, incoherent foreign policy, etc. -- and instead lash out at the pollsters. It's one thing to blame the messengers.
But to do it the way you're doing sets up your side for years of further -- and growing -- irrelevance. You're further fringe-iying people already trying to eke out a bare existence on the political fringe. You're giving them yet another typically trusted, universally understood area of How We Talk About What Is Happening In Our Country -- using transparent, statistically valid polling -- to now inherently distrust, and turning it into an issue. A cause for blame. And you're needlessly delaying the internal conversation that conservatives are going to eventually have to have with themselves in order to return to any sort of political relevance in the country on a national level. You're burying your head in the sand, ranting about sampling rates and party ID. You're making it harder on yourselves.
That's not going to help anyone. It's not going to serve the country, or help us the next time a complex issue comes up and gauging public opinion would aid the conversation or the policy-making. It's not going to prop you up, as a conservative commentator farther and farther out on the cliff, pandering to an ever-shrinking number of true believers.
And it's certainly not going to help conservatives as they try to pick up the pieces and go forward after the 2012 shellacking they're about to receive.
Re-think your strategy. Own your coming loss. And it's going to be a big one.
hugs,
save the clock tower