Just in time for the latest, greatest Shitgibbon pursuit of all those not-good-people who got to vote for his opponent, Maggie Koerth-Baker brings the hammer down. She’s written an excellent long-read over at Five Thirty Eight on what went wrong in the ur-paper that has fed the right wing fantasy that a gazillion undocumented brown people threw the election to the popular-vote winner, but somehow failed to actually turn the result.
The nub of the problem lies with a common error in data-driven research, a failure to come to grips with the statistical properties — the weaknesses — of the underlying sample or set. As Koerth-Baker emphasizes this is both hardly unusual, and usually not quite as consequential as it was when and undergraduate, working with her professor, used found that, apparently, large numbers of non-citizens 14% of them — were registered to vote.
There was nothing wrong the calculations they used on the raw numbers in their data set — drawn from a large survey of voters called the Cooperative Congressional Election Study. The problem, though, was that they failed fully to handle the implications of the fact that the people they were interested in, non-citizens, were too small a fraction of the total sample to eliminate the impact of what are called measurement errors. Koerth-Baker writes:
Non-citizens who vote represent a tiny subpopulation of both non-citizens in general and of the larger community of American voters. Studying them means zeroing in on a very small percentage of a much larger sample. That massive imbalance in sample size makes it easier for something called measurement error to contaminate the data. Measurement error is simple: It’s what happens when people answer a survey or a poll incorrectly.1 If you’ve ever checked the wrong box on a form, you know how easy it can be to screw this stuff up. Scientists are certainly aware this happens. And they know that, most of the time, those errors aren’t big enough to have much impact on the outcome of a study. But what constitutes “big enough” will change when you’re focusing on a small segment of a bigger group. Suddenly, a few wrongly placed check marks that would otherwise be no big deal can matter a lot.
This is what critics of the original paper say happened to the claim that non-citizens are voting in election-shaping numbers:
Of the 32,800 people surveyed by CCES in 2008 and the 55,400 surveyed in 2010, 339 people and 489 people, respectively, identified themselves as non-citizens.2 Of those, Chattha found 38 people in 2008 who either reported voting or who could be verified through other sources as having voted. In 2010, there were just 13 of these people, all self-reported. It was a very small sample within a much, much larger one. If some of those people were misclassified, the results would run into trouble fast. Chattha and Richman tried to account for the measurement error on its own, but, like the rest of their field, they weren’t prepared for the way imbalanced sample ratios could make those errors more powerful. Stephen Ansolabehere and Brian Schaffner, the Harvard and University of Massachusetts Amherst professors who manage the CCES, would later say Chattha and Richman underestimated the importance of measurement error — and that mistake would challenge the validity of the paper.
Koerth-Baker argues that Chatta (the undergraduate) and Richman, the authors of the original paper are not really to blame for what came next — the appropriation of this result as a partisan weapon in the voter-suppression wars. She writes, likely correctly in my view, that political science and related fields are more prone to problems of methodology, and especially in handling the relatively new (to these disciplines) pitfalls of big, or even medium-data research. The piece goes on to look at how and why this kind of not-great research can have such potent political impact, long after professionals within the field have recognized problems and moved on. A sample of that analysis:
This isn’t the only time a single problematic research paper has had this kind of public afterlife, shambling about the internet and political talk shows long after its authors have tried to correct a public misinterpretation and its critics would have preferred it peacefully buried altogether. Even retracted papers — research effectively unpublished because of egregious mistakes, misconduct or major inaccuracies — sometimes continue to spread through the public consciousness, creating believers who use them to influence others and drive political discussion, said Daren Brabham, a professor of journalism at the University of Southern California who studies the interactions between online communities, media and policymaking. “It’s something scientists know,” he said, “but we don’t really talk about.”
These papers — I think of them as “zombie research” — can lead people to believe things that aren’t true, or, at least, that don’t line up with the preponderance of scientific evidence. When that happens — either because someone stumbled across a paper that felt deeply true and created a belief, or because someone went looking for a paper that would back up beliefs they already had — the undead are hard to kill.
There’s lots more at the link. Highly recommended. At the least, it will arm you for battle w. Facebook natterers screaming about non-existent voter fraud “emergency.”
Image: William Hogarth, The Humours of an Election: The Polling, 1754-55
Baud
Lies, damn lies, statistics, and Kobach.
TenguPhule
@Baud: Lies, Damn Lies, Donald Trump.
Brachiator
Tom Levenson
Thanks for this. The potential problems involving such small samples is so basic that you wonder how anyone could have run with this stuff.
NotMax
re: that last blockquote, two words –
Cold fusion.
Brachiator
Also, too, analysis and conclusions based on “self reporting” are fundamentally unreliable.
schrodingers_cat
Research that cannot be reproduced is worthless.
Mnemosyne
@Baud:
Yep. This particular zombie is going to be trotted out by Kobach and forced to dance for his “commission” to disenfranchise non-white Americans.
les
One nice thing about this fucking mess: with Kobach running the “committee,” there’s no “stupid or evil” question. He’s a really smart guy.
different-church-lady
Let’s not overthink this.
Gin & Tonic
@schrodingers_cat: It’s not worthless. It can have entertainment value.
Tom Levenson
@Gin & Tonic: Ah yes. JIR FTW. My long-time favorite, “Impure Mathematics”…or perhaps, “The Inheritance Pattern of Death”… or the money saving suggestion on reusing your electron micrograph for any article in any discipline…or…
Heh.
Wag
@les:
So we’ll be dealing with pure, unadulterated evil.
LurkerNoLonger
Of all of the innumerable reasons not to like Republican politicians and their enablers, the blindingly obvious steps they’ve taken to suppress the minority vote is number 1 in my mind. If they can win honestly, good for them. If they have to cheat to do it, they’re fucking scumbags.
debbie
I think James O’Keeffe is responsible for more illegal votes than actual illegal voters.
Miss Bianca
@Tom Levenson: My favorite article was an illustrated diagram of the operation to separate Siamese Marshmallow Peeps. But that might just be because of my Peep fixation.
Gin & Tonic
@Tom Levenson: I’ve been a fan since [mumble-mumble.]
Roger Moore
@NotMax:
And the vaccine/autism link. The key is when people desperately want to believe the research, they’ll happily continue to cite it no matter how often it’s been disproved or how many retractions are printed.
Brachiator
I am loving this 538 piece. Really enjoy the section where the reader can use a slider to see dynamic changes in results.
This is why, for example, Asian Americans (or Native Americans) are often omitted from state or national exit polling or election analysis.
les
@Wag:
You got it. K Kobach is at the bottom of more state level evil shit than you would believe. I’d bet he holds the record for immigrant bashing and voter suppression laws found unconstitutional, by a long way–all over the country.
Adam L Silverman
I just got back in, so this is OT, but in case it hasn’t come up, the guy who owns/runs the firm raided today used to work for the successor firm to Black, Manafort, and Stone…
Fair Economist
A simple proof there’s no significant vote fraud going on is that states keep records of who voted. If it were going on, it would be easy to find them – and Republicans have tried, but they can’t, because they don’t exist.
Kay
Well, his credibility is shot. He’ll be standing there like a dope while the Trumpsters say he said things he didn’t say and lie about voter fraud.
I don’t know how many people have to be brought down by Donald Trump before they learn to just stay away from him. He’s freaking poison.
I knew it, too. I knew there would be some dumb Democrat who would be flattered into allowing himself to be used. He thinks he’s “at the table”. He’s ON the table.
Baud
@Kay: Who is the Dem?
Kay
I’m ashamed that there are Democrats who would volunteer to lend this sham credibility. Jesus Christ. HOW MANY TIMES do they have to get screwed by these people?
Now the giant windbag will be out bellowing that both Parties endorse his crackpot conspiracy theories.
Good job. Bravo.
Kay
@Baud:
Maine. He says he can walk away if it’s a sham. Bullshit. Whatever nonsense they come up with Trump will lie and say this guy endorsed it. I don’t know why people think credibility has no value. It really does. ESPECIALLY for an election official, which is what he is. To volunteer to donate it to Donald Trump? Idiocy.
Baud
@Kay: The only non-gop male from Maine is Angus King, who is an independent.
Kay
@Baud:
Labor organizers call this kind of capture “at the tablism”. It always makes me laugh – it’s perfect :)
Baud
@Kay: I suppose some people think it’s better to keep an eye on things from the inside, but I tend to agree with you.
Kay
@Baud:
Six terms of good work shot after one photo op with Donald Trump. They just piss it away. “Sure- I’ll stand with the pathological racist liar and lend him my hard-earned credibility!”
Baud
@Kay: Oh, it’s a state guy. Yeah, they’ll walk all over him.
Patricia Kayden
@Baud: If it is Angus King, his constituents should be flooding him with calls, emails and personal visits to stop him from giving a bipartisan sheen to any findings that the Voter Fraud Commission comes up with. The whole purpose of this charade is to enact more voter suppression laws.
Corner Stone
@Kay: How fucking stupid do you have to be to be any where near Trump at this point? He will suck your credibility away at one stroke and leave you a dried out pile of shit.
Lizzy L
@Baud: If you can’t identify the mark at the table — it’s you. I wonder how long it will take for him to figure it out and resign.
? Martin
@Baud: There’s really nothing to keep tabs on from the inside. The leadership of the commission tells you that the outcome is already determined. Now its to find the 11 people in the country that will testify that fraud is widespread because ACORN, or whatever. That’ll be used to bolster whatever bullshit agenda they’re really after, and that will be used to justify whatever bullshit legislation they want to pass in order to suppress the vote in 2018.
The best course of action here is to fully discredit the effort by not participating in it.
Baud
@Lizzy L: I wonder who the mark at Balloon Juice is.
ET
From NBC:
One could only wonder why he actually thought that was a good idea.
Corner Stone
@Lizzy L: It doesn’t matter anymore for this person. He’s got the touch of shit on him now. There is no reclaiming your dignity once you give it to Trump.
Baud
@ET: Because it was so successful when he did it at the CIA?
Baud
@Corner Stone: We should print up and hand out “You’ve Been Trump’d” stickers.
efgoldman
@ET:
Sycophants and hangers-on have sucked up to him for 70+ years. And now that he’s “president”, there’s the additional institutional deference (I refuse to say “respect”) that’s built in to the job even if J Fred Muggs were in the office. So he’s convinced it’s due to him, he “earned” it, and all he has to do is show up.
debbie
@Baud:
Colbert beat you to that. He’s made up Trump’d hats.
Timurid
@Adam L Silverman:
I’ve heard some pretty hyperbolic stuff on Twitter about how this case could pull in a bunch of major GOP players.
I’ve also heard quite a bit about how this may have nothing to do with Trump or Russia and could be a simple fraud case in which the company was skimming funds raised for its clients (as was alleged in a 2013 lawsuit by a former client).
Patricia Kayden
@ET: Because the WH is claiming that rank and file FBI employees had lost confidence in Comey. Of course, this is just another lie.
Baud
@debbie: Funny.
Jeffro
@Kay: @Baud:
I get the concern, but I’m just dumb enough to think like this guy: “I’ll be on the commission and if I see something fishy, I’ll squawk, and if they cut me out of meetings or deny requests for info, I’ll squawk, and…”
the problem is, sometimes they figure out how to not make you squawk.
But if this guy is as much IDGAF as I am, hey who knows what might come of it all?
Roger Moore
@Baud:
One of the too-numerous cohort who reply to trolls.
? Martin
@Jeffro: The problem is that anyone who serves on that commission has just ruined their career. They can squawk all they want, the fact that they had such poor judgement to sign on in the first place undermines the credibility they believed they had going in.
Steeplejack
@efgoldman:
I wish!
Corner Stone
@Jeffro: Hey! Why ain’t I in that reply?
It’s not like I am going to be ignored here!
danielx
@Kay:
If there’s a sucker in the game and you don’t know who it is…..
Peale
@Corner Stone: your ID is out of date, so you’re only getting a provisional reply anyway.
efgoldman
@Patricia Kayden:
Right now, if anyone in the WH said today was Thursday, I’d look really closely at my calendar
Kropadope
Besides people leaning on surveys like this to claim the rampancy of voter fraud, another thing I often see brought up is that paid volunteers for voter registration drives will turn in forged registrations. Of course, they do this to the extra pay, so their scamming the drive organizers, not the electorate.
If no one shows up to vote claiming to be Mickey Mouse, having Mickey registered shouldn’t be that much of a problem.
TenguPhule
@efgoldman:
In parts of the world its already Friday.
Kropadope
@? Martin:
Maybe we should see what the commission does and his role before coming to a conclusion? If he winds up subverting an effort to fraudulently disenfranchise legitimate voters, good on him.
TenguPhule
@Baud:
Its always Baud. We already ordered pizza and pron with your credit cards.
Omnes Omnibus
@Kropadope: The other part of this is that there are people named Michael Mouse, Donald Duck, and the like. During the Scott Walker recall, a guy named Joe Stalin contacted the WI GAB to say that he was indeed Joe Stalin not a made up name and he wanted Walker recalled. His name was not removed from the petition.
japa21
@Kropadope: Interestingly enough, that was one of the complaints about ACORN. However, by law all registration forms have to be turned in, even if some are obviously fraudulent. ACORN would go through the forms prior to handing them in, put the suspected frauds in a separate bundle so the officials could make determinations, and then participation in legal action against those who collected them.
japa21
@Kropadope: I agree with you. It seems to me it is important to have someone on the inside looking at what is happening and being able to provide a counter argument. Also prevents any report from being labeled unanimous.
Kropadope
@Omnes Omnibus: I once knew a Donald Tuck.
Going back a ways, when they purged “felons” off the FL voters rolls before the 2000 election, they got rid of any close matches. It’s not enough to disenfranchise former convicts who paid their due and are now being taxed without representation, they have to disenfranchise anyone who sounds like a felon. You know, those people.
Steve in the ATL
@Baud:
Totally not you. Really.
Tim in SF
I’ve been reading this blog since the Schiavo days. It’s rare that I have to re-read a paragraph half a dozen times to figure out what it is supposed to say. And even after six reads, I still am not *completely* sure what this means.
Kropadope
@japa21:
/James O’Keefe
TenguPhule
@Steve in the ATL: Your pizza is in the mail.
Lizzy L
This is totally OT, but I’m putting it up here because it infuriated me. Mulvaney is a monster.
Oliver Willis reports:
TenguPhule
@Tim in SF: Balloon Juice: Come for the technical analysis, Stay because you can’t understand it.
Omnes Omnibus
@TenguPhule: Huh?
gene108
@Kropadope:
You don’t want the guy registering voters to make the decision that Mickey Mouse does or does not have the right to vote, by not turning in the registration card.
That is something to be left to the BOE.
TenguPhule
@Lizzy L: He’s going to be a literal dead man walking at this rate. Diabetes is no joke and plenty of those afflicted are much more mobile and better armed then last stage cancer patients.
TenguPhule
@Omnes Omnibus: Exactly.
SiubhanDuinne
@Adam L Silverman:
Frankly, this is the least surprising thing I’ve read all day. I didn’t predict it, but at some deep cellular level I just knew there had to be some kind of Manafort Stone connection. Thanks, Adam. Stay frosty and please keep finding this stuff for us.
Kropadope
@Lizzy L:
Nice to see that good Christian Republicans believe that people can make lifestyle changes and redeem themselves.
Sorry, I mistyped, meant can’t. Can’t find redemption.
Kropadope
@gene108:
Oh, of course not, and that would be its own pernicious form of fraud. But registration deadlines are typically months out from the elections, there should be plenty of time and opportunity to verify the authenticity of questionable registrations. That is, if the people so concerned about voter fraud were actually interested in stopping it, rather than posturing for the sake of putting up roadblocks for legit voters.
SiubhanDuinne
@Baud:
Does the FBI have a wall of anonymous martyrs that he could stand in front of and lie about something?
(Not that he needs a wall of heroes to lie, but it adds to the visuals. Might show up on a magazine cover, and then he’d have another picture to frame.)
Lizzy L
@TenguPhule: Fine with me. He’s a monster. My mother was an insulin-taking diabetic; I took care of her for her 10 final years. Threatening sick people pushes ALL my buttons.
Corner Stone
@Lizzy L: Mulvaney has already gone on record saying feeding hungry children is not useful. The dude is begging to get some kind of violent comeuppance.
jl
@Lizzy L: In interviews, these Trumpster cruds have been frequently dog whistling and hinting with pretty thin disguise, that most poor, working, and middle class people don’t deserve to have any health insurance. That is for the rich, and those with high class jobs. No one else deserves good health care. They are so eager for it, they can barely keep themselves with explicitly yelling and spelling it out with an ear-to-ear grin.
The only bright side I can see to it is that they won’t be able to hide their depravity with such poor self-discipline.
I don’t know if what Mulvaney is saying is much worse than Ryan happily and mindlessly burbling about how taking away lesser people’s health care is fulfillment of his beery frat rat kegger dreams.
TenguPhule
@Corner Stone:
Oh gods, the school lunch dickhead?!
What’s next, “only virgins can be raped”?
jl
@Corner Stone: These dudes are chomping at the bit to yell every detail of their plans to very people they plan to kill off. That is their problem. They won’t be able to hide their real intentions. They risk a huge public backlash, and I hope that they get one very soon.
TenguPhule
Today’s Edition of Evil Bastard Republicans (ht: Atrios)
The Republicans didn’t fall off the slippery slope, they jumped of their own free will.
jl
@Tim in SF: I’m not sure how to parse that extract either.
But the nub of the problem is that if the measurement error is about the same magnitude as the systematic effect you want to estimate, then what seem to be small things can blow up into huge mistakes in the estimates.
Applied to this issue, intuitively, the systematic effect you want to estimatete (prevalence of illegal votes by non-citizens) is directly related to the proportion of people who are undocumented, which is a relatively small proportion of the sample. So what seems to be a small measurement error produces an unexpectedly large mistake in your final estimates.
Maybe it will help to think of it in terms of the rates of false positive diagnoses for, say, a test for having disease, when the prevalence of people with the disease in the population is low.
Or, another intuitive example is the rate of false positives in IDing people who will commit terrorist acts from a screening algorithm, when the proportion of people who will ever commit a terrorist act is very low in the population.
TenguPhule
Via CNN (hat tip Digby), the shitapple didn’t separate from the tree.
Omnes Omnibus
@TenguPhule: Bill Sessions killed Vince Foster?
Jeffro
Carter Page on MSNBC right now, holy cow…talk about eyes-rolling-around-in-your-head crazy…
Jeffro
@Corner Stone:
He’s pretty high on the priority list, yes.
SiubhanDuinne
O.M.G. Have you seen the new New Yorker cover?
Here you go.
TenguPhule
@Omnes Omnibus: At this point I swear they’re just using a random phrase generator to come up with this shit.
Patricia Kayden
@TenguPhule: That’s cold but to be expected from Republicans. Nothing they do is a surprise anymore.
jl
And, apparently, Trump just said in an interview that, ‘yeah sure I fired Comey to stop the stupid Russian investigation because it’s all BS, and that’s OK since it’s all BS waste of time and money”
Whew boy…
The President Admits to Obstruction on National TV
http://talkingpointsmemo.com/edblog/the-president-admits-to-obstruction-on-national-tv
TenguPhule
@Patricia Kayden:
Would you believe they’re already saying that at least Trump isn’t as bad as Hitler?
Omnes Omnibus
@Patricia Kayden:
Shocking, but not surprising.
Jeffro
@jl: it was better and weirder (and even more 3rd person) than that…serves as impeachment material, late-night comedy gold, 25th amendment proceedings…
Not with ‘me’ and Russia…’Trump’ and Russia…
Patricia Kayden
@Lizzy L: Sounds like Mulvaney is perfectly ok with death panels — as long as it’s for people who he decides are responsible for their own medical conditions. It has never occurred to him that some poorer people eat unhealthily because of food deserts.
amk
@Lizzy L:
has the pos taken a dekko at the pos he sees every day?
debbie
@Lizzy L:
I think I miss David Stockman.
chris
@Corner Stone:
Well, you know it’s hard to feel truly blessed as a big wheel in a banana republic unless there are beggars in the streets. Christ, what an asshole!
Lizzy L
@Jeffro: He’s at the top of mine.
japa21
@debbie: Heck, I miss Alberto Gonzales. Hell, I even miss James Watt.
SiubhanDuinne
@Omnes Omnibus:
That really encapsulates my reaction to just about everything since 16 June 2015. Everything is horrifying, but when I run a particular episode through my mind (or past my friends), the reaction is always some variant of “Yeah, well? What else did you expect?”
Patricia Kayden
@jl: So how is this not an impeachable offense? Just strengthens the calls for a special investigator.
Omnes Omnibus
@japa21:
Let’s not go too far. Watt would fit in perfectly with this bunch.
TenguPhule
Fuck Texas’s Government. Seriously, Skullfuck them to death with a flaming chainsaw.
Break the law, get paid anyway. The Official Republican Motto.
ETA: Yes, they’re trying to ban abortion without having to pass a law to do so.
SiubhanDuinne
@debbie:
I miss the woodshed.
TenguPhule
@japa21:
Trump plans to rehire him as official Torture justifier.
SiubhanDuinne
@Omnes Omnibus:
Wait, Watt?
gene108
@Patricia Kayden:
You don’t need food deserts and poor people, some people are just prone to type2 diabetes, just like some folks have weak teeth and need more dental work than others, even though they take care of their teeth.
I’d love to see Mulvaney’s results from a physical. If he’s not the perfect example of physical health, he should be denied access to all medical care, because he’s a dick and an asshole.
efgoldman
@TenguPhule:
That (among other things) is what Tangerine Temper Tantrum is.
Lizzy L
@Patricia Kayden: I don’t give a flying fuck WHY people get diabetes. A civilized society does not impose some kind of morality test on citizens who need health care. (Which means this is not a civilized society. But you knew that.)
Mulvaney’s probably Catholic, like Ryan: what part of “That which you do to least of these, who are my brothers and sisters, you do to me” do these twin monsters not get?
Omnes Omnibus
@SiubhanDuinne: Roger, Roger.
Marcopolo
How’s about a new thread that prominently displays the new New Yorker cover which is lovely!
http://www.newyorker.com/wp-content/uploads/2017/05/CoverStory-Blitt-Comey-872×1200-1494533319.jpg
Kropadope
@TenguPhule:
Only if you’re the right sort of person breaking the right sort of law.
SiubhanDuinne
@gene108:
My read on him is, even if he is the perfect example of physical health, well tant pis, he is still a dick and an asshole. So deny away as far as I’m concerned.
danielx
@Tim in SF:
Statistically speaking, as a general rule the larger the sample the more accurate the estimate, up to a certain point. From what I recall from having done a lot of sampling and survey work in a previous life, the ‘certain point’ is around 2000 for a given population – after that it doesn’t matter whether the population you’re surveying is a hundred thousand, two million or a hundred million, anything over ~2000 is not going to improve the accuracy of the survey in terms of error rate, I forget why. Which is why you will see voter surveys at election time that say “a survey of 1,718 registered voters”, blah blah, “produced the following results at a 95% confidence level with an error rate of +/- 6%” or some such, which sounds very authoritative. Meaning that if you repeated the survey 100 times, 95 times you would get the same result +/- 6%. Which is a fairly substantial variance right there, that 6% figure. But okay, still not a bad sample of the overall registered voter population.* But when you get to individual subgroups of the overall registered voter population, your potential problems start multiplying quickly, especially for self-identified subgroups. Ideally you’d have a stratified random sample in which the composition of subgroups mirrors (as closely as possible) the percentages of those subgroups in the population of registered voters. But even then, assuming your sample size of ~2000, the error rate for the individual subgroups within that sample goes way up.
Let’s say you wanted to examine the voting behavior of left-handed Nicaraguan ski jumpers as a subgroup of your sample of ~2000. You don’t know how many left-handed Nicaraguan ski jumpers there are in the general population of registered voters, but your sample includes 52 people who have self-identified as such. Without knowing the overall population of that subgroup, your sample of 52 would produce an error rate of 13.59%, assuming the characteristic for which you are testing is 50% (say, 50% voting R, 50% D). Fifty percent is what you might call the worst case scenario and the most conservative estimate when you have no idea what the proportion may be for your subgroup***. However…in this example, what your subgroup sample of 52 left-handed Nicaraguan ski jumpers would produce would be a result that sez…..if you surveyed that same group 100 times, 95 times you would get a split of 50% R/50% D, +/- 13.59% – so 95 times out of 100 you’d get results ranging from 36.41 to 63.59% for Republican/Democratic voters, so essentially useless results for that population subgroup. To get the error lever down to say, +/- 5% for that subgroup, you’d need a minimum sample size of 384 left-handed Nicaraguan ski jumpers, so you begin to see the problem. If you want reasonably good estimates for all the various subgroups in which you’re interested, your sample size has to go up – way, way up, depending on which and how many subgroups for which you want your reasonably good estimates. And even then there’s going to some error. Larger sample size means more money and more time and still may not give you the level of accuracy you want.
You can probably hear the sound of imminent brain death….
*All this shit used to be a lot easier when people only had landlines.
**50% is the most conservative and the safest estimate when you have no idea of your subgroup’s leanings. If I was interested in NRA members, it’s safe to say that you could use a different proportion – assuming 80% will vote Republican, for example, which would give you a smaller required sample size for a given acceptable error level.
***If anybody is really interested in this stuff, there’s a handy-dandy online calculator (with explanatory notes) that will tell you how big a sample you need for a given population for a 95% or 99% confidence level and error level, etc etc.
japa21
@Omnes Omnibus: See, this bunch makes me lose all perspective. You are right.
japa21
@SiubhanDuinne: I think being a dick and an asshole qualify as pre-existing conditions.
Steve in the ATL
@SiubhanDuinne: this is why I am a subscriber. Well, this and the caption contest.
SiubhanDuinne
@japa21:
Good! Into the high-risk pool with him!