It happens almost every election season: the return of the Poll Analysis Concern Trolls. Well…they’re baaaaaaaak!!!
This season we have HA’s newest amateur right-wing propagandist, “Bob”, who is vewy, vewy concerned about the methods and polls I use for the election analyses. And we have the return our most esteemed amateur right-wing propagandist (to put it kindly), currently under the name, “Smilin'” (before that, ironically self-named, “GetFactsFirst”) . If you are interested, you can follow some of their election analysis concern trolling here, here and here.
I don’t want to totally belittle our Concern Trolls. The do play some useful roles here, like contributing to the raucous back-and-forth in the comment threads. And, for me, providing new opportunities to pontificate about polls, probabilities, statistics, simulations, bias, etc—topics that I enjoy in my professional life as well as in my hobby of collecting and analyzing electoral polls.
I also want to acknowledge them for inspiring a new occasional feature for this election season: The Electoral Pundit Contest. It is sort of like Lee’s Birds Eye View contest, but dealing with polls and stuff. The challenge is given below, but first allow me to pontificate….
This first contest was inspired by Bob and Smilin’s discussion of “outliers” in polls. It really bothers them that I don’t assess whether polls are “outliers.” And their latest “target” is a new Pennsylvania poll from Franklin and Marshall college (also known as The Keystone Poll). It shows Obama leading Romney 48% to 36% with 17% selecting neither.
What triggers their “concern” is the partisan make-up of the poll: “Respondents 50% D, 37% R, 10% I.”
Smilin’ puts it:
Why would Darryl include a poll that uses 50% Dems? Seems like there are several “outlier” polls like this that have zero credibility because of their underlying assumptions.
Is this poll an outlier? We could approach this from a probabilistic point of view by asking the question: if the sample of 412 registered voters was truly a random sample of PA voters, what is the probability of drawing a result as “extreme” as 50% Ds and 37% Rs and 10% I?
To make this easier, let’s ignore the “I” category, so the question becomes: if the sample of [207 Ds + 154 Rs =] 361 registered “partisan” voters was truly a random sample of PA voters, what is the probability of drawing a result as “extreme” as [50%/(50% + 37%) =] 57.5% Ds and [37%/(50% + 37%) =] 42.5% Rs?
A proper test would require us to know the “truth” about the probability of drawing a D versus an R in the population. Suppose the “true” probability is 54% for drawing a Democrat and 46% for drawing a Republican (ignoring folks who are Independent). We could then ask: for a sample of 361 partisans and a true probability of 54%, how probable is it to draw at least 207 Ds?
There is an exact answer to this question that can be found from the Binomial Distribution. The answer is about 11%.
In other words, if we did a bunch of polls with truly random samples of 361 registered voters each (assuming truthful answers, etc.) and with the true proportion of Democrats of 54%, we would, just by chance, draw a Democratic sample of 57.5% or greater about one out of every nine such polls. Hence, this particular evidence is not very strong, under our assumptions, that the poll is an outlier.
Whether partisan make-up or whether we look at the percentage “voting” for each candidate, there isn’t usually strong evidence for outliers. For example, let’s look at all polls for PA in the 2012 Obama—Romney race:
The vertical lines show the plausible range of “true” proportions, given the poll proportion and the sample size.
Two points. First, the plausible range of the most recent Franklin and Marshall poll largely overlaps all recent polls. The best evidence of an outlier comes from the previous Franklin and Marshall poll that just barely overlaps a Susquehanna poll (yellow). But both polls plausibly overlap their neighbors. So…which one should go? Or are they both perfectly valid, but happened to legitimately draw samples at each end of the spectrum? The rule for my analysis is to assume the difference is sampling variability, and include both polls. Since the election analyses typically have 60 or more polls, this sampling variability will, more or less, cancel out.
The second point is that the most variable polls are the smallest polls. The most current Franklin and Marshall poll is tiny. (In fact, you can get a rough idea of the sample sizes of polls from the plausible range—the Quinnipiac polls (cyan) all have samples over 1,100.) Because of the mechanics of the simulation analyses, larger polls (with smaller sampling error) have greater influence on the analysis.
Contest: There are three parts.
(1) In the above discussion, I had used 54% as an example for the “true” proportion of Ds versus Rs in Pennsylvania. Your task is to provide your best estimate of the true proportion of Democratic, Republican and “independent” (or other) voters in Pennsylvania. Use any resource and estimation technique you wish. Since partisan composition could change daily, let’s pin it down to June 4th (the last day of the Franklin and Marshall poll) as our target day.
(2) Assess the difference between your best estimate (part 1) and the partisan composition of the Franklin and Marshall poll (this is simple subtraction). The difference may be surprising.
(3) What is the cause for the “surprising” difference?
Good luck!
You mean I’m not a one-tune shill for the unions? That would be news to Liberal Scientist who…
Thank you for including a discussion of not just poll results, but what might go into a poll. Maybe flesh it out and go into differences between polls of adults (the Pew thing that has been mentioned this week on HA in the comments), registered voters, and likely voters.
A subsequent post can do into differences resulting from how people are contacted: exit poll, whether cell phone numbers are used, etc.
It’s more fun than talking about what someone did 20 years or 40 years ago, anyway.
In 2008, PA went for Obama over McCain, 54/44. That’s 10 points.
In 2010, PA went for Toomey over Sestak, 51/49. That’s two points in the opposite direction.
In 2010 PA governor race, Corbett (R) beat Onorato (D), 54.5 to 45.5. That’s nine points in the opposite direction.
Ergo, Darryl, with respect to this specific poll, my point is that given that even when Dems had about every possible star aligned for them, the Dem winner only had a 10 point margin, and in the most recent statewide elections in which results show that the winner was a Republican in both of the majors, with vote margins of 2 and 9 percent, it is very, very, very unlikely that in this current environment Dems have a 13 point advantage over Republicans.
That is my sole point, actually.
Unless, of course, there are an awful lot of Republicans out there in PA who really think that it was creepy that Romney played pranks in trooper uniform 40 years ago and because of it will vote Dem this year. Maybe it’s the radon.
Or perhaps there are an awful lot of Dems out there in PA with plenty of time to answer polls but no time to fill out an absentee ballot or go stand in line come election time. And if that’s the case, then we really should focus on likely voter polls to exclude the lazy scum who won’t participate when it’s time.
To be fair, I did read that PA has more college-educated whites than whites without college. That demographic could help explain some of the Dem advantage in PA.
But still……..
Ba-Ba-Boom Bob @2: As they say in the investment business, past performance is no guarantee of future performance. Now I’ll leave it to your brilliance to extrapolate that truism to voting behavior.
I’m not a statistician (although I did get an “A” in my college statistics course), but it’s not hard for me to imagine why a current poll might be a more reliable indicator of current voter sentiment than past election results:
1. Change of heart;
2. People moving in and out of the state;
3. Old voters dying and new ones joining the voter rolls;
4. Variations in voter turnout rates;
5. Demographic changes;
6. Media influences;
7. Changes in local economies;
8. Shifts in issue priorities; etc.
But then there are these factors, too:
1A. A growing distrust of pollsters and unwillingness to cooperate with polls, so you’re only getting a profile of a subset of the population, i.e. those people willing to cooperation with polls;
2A. Increased incidence of lying to pollsters by partisans who wish to conceal their real intentions for various reasons;
3A. Poll respondents who honestly don’t know their own minds and may say one thing today but do something else on election day;
4A. Low information voters who haven’t a clue what’s going on in the world and probably will make a knee-jerk voting decision in the Voting Booth on Election Day who will tell a pollster whatever comes off the tip of their tongue.
If I recall correctly, pre-election polls haven’t been all that close to the actual result on Election Day recently, which suggests that something is breaking down in the polling process. I suspect it’s not the methodology, but rather the respondents, that have deteriorated. See 1A – 4A above.
The takeaway, it seems to me, is that we shouldn’t get too infatuated with polling data — especially at this early stage of the 2012 campaign.
When buying stocks, I try to predict the future by looking at today’s fundamentals. The fundamentals of the 2012 presidential race — what the issues are, who the candidates are, what their records are, what the mood of the electorate is — leads me to believe this is not 2008 or 2010 and we may see wholesale slaughter in downticket races but I really question whether Romney has what it takes to make voters feel they have an alternative to Obama.
@3 So you’re admitting that educated people don’t vote Republican?
This is the first presidential election since ’36 in which Pop Rabbit (RIP; sniffle) won’t be voting. That will skew the election results in Wisconsin, as some young bunny (or, more likely, several) will take his place who have no idea what it was like to have lived through two World Wars, the Depression, and the Cold War. In one of my last conversations with Pop R. before his passing last year, I was shocked to find out how far he had strayed off the FDR-New Deal reservation in his later years, which I attributed to feeble-mindedness. I don’t know if he would have voted for Romney, but I do know that if he would have, that’s a vote Romney isn’t going to get (notwithstanding snarky troll comments about dead voters in Milwaukee, because even if the snarks’ premise is correct and Pop Rabbit votes again this fall, according to their own thesis it won’t be Romney who gets that vote).
@ 5
That might be one way of putting it.
Democrat voters have a bimodal distribution. The most educated and the least educated vote Democratic.
The GOP voters are clustered more in the middle although the average GOP voter is more educated than the average Democrat voter.
Would you be OK with me saying that stupid people don’t vote Republican, too? Or would you object to that, RR?
@5, @6 @ 7 not really on topic.
@7 I wouldn’t feel comfortable with any of your assertions unless backed by empirical data, competently analyzed.
Darryl-
Good work…sincerely.
It is interesting and actually fun to hash over the underlying assumptions and source data used in these polls. But like you said, it’s only trying to hone in on an accurate assessment of where things stand today, not trying to predict the November outcome. I suppose one could using trend analysis and other tools to try to take Daily Analysis over time and try to project it to November, but that requires lots of assumptions too.
Question Darryl–
Let’s say you had 20 poll results and that 80% of them (16) were between +2 and -2 and you had 4 consistent “outliers” that were both on the same side by say +20. Those 4 polls would have quite an impact on using your methodology Darryl….and it would not seem to generate a reasonable snapshot. I know that’s quite an extreme and probably hasn’t happened in your case, but it helps to see it can happen at some level..and in a close race could give you some false confidence.
I appreciate that you lay out your assumptions and sources of data Darryl. Too many folks just cherrypick to try and prove a pre-determined point & you cannot see how they came to the result they did.
You are forthright Darryl.
Do I agree with methodology including “outliers”? It’s really not a matter of that.
It’s a fun exercise though.
@8 If you whine loud enough to Darryl maybe he’ll do something about it.
@2: not even remotely on topic.
The question has nothing to do with voting results. Darryl wants to know about registration patterns.
I’m waiting for the current spreadsheet of voter registration by party to finish downloading from the Pennsylvania Department of State’s website. In the meantime, though, I can school Bob on what the party registration patterns were in November 2011 (PDF):
Republican 37.1%, Democratic 50.7%, all others 12.2%
Sounds like the Keystone Poll was spot-on.
Well, now that the spreadsheet (XLSX) is ready, I’ll give more contemporary numbers. As requested, the June 4 numbers:
Republican 37.3%, Democratic 50.2%, other 12.5%
As noted earlier, the poll distribution is spot-on.
@ 12
You didn’t read my comment in @ 2.
If the registered ones don’t vote, polls of registered voters aren’t accurate insofar as election results are concerned, are they?
You point out a 13% Dem registration advantage in 2011 but I pointed out a 2% and 9% GOP vote tally advantage only a year before. Of these two perspectives, which is more relevant to the 2012 election?
@ 9
Next time do your own homework.
See:
http://pewresearch.org/pubs/20.....ite-voters
Second table, most of the way down. Highest and lowest education categories are populated primarily by Dems, the middle one is by GOP. This one link doesn’t address average education but the highest earners are GOP so you can figure it out, knowing that higher education levels generally correlate with higher income. At least until the current Higher Education Bubble, anyway.
Asked, answered. And all because you shot your mouth off.
13:
When one states that one is polling registered voters, the distribution of registered voters is the only relevant comparison.
@ 12
We’re essentially talking about two different things. You’re addressing the accuracy of a poll relative to party registration levels.
I’m addressing relevance of a poll when the registration levels are so badly disconnected with ballot box results.
Ask yourself: Do you really think Obama will be +13 or anything close to it in PA this year? Coal, Catholics, etc. Really?
@ 15
Then you’re setting yourself up for an awfully big downside surprise in fewer than five months.
17:
Tell that to the folks at Franklin & Marshall. They’re the ones who found it appropriate to poll registered voters (perhaps because they haven’t yet worked through their “likely voter” modeling, so it would be unscientific to poll in any manner other than “registered voters” at this time).
BTW, I’d say that the current poll results are one of the high points for Romney. He’ll peak again around the time of the RNC.
@ 18
Both candidates will likely get a post-convention bounce. Fairly typical.
Don’t get me wrong: I do not object to inclusion of registered voter polls. I simply believe that in some situations an obvious disconnect between one set of objective data (registration) and another (election results) should prompt discussion, as is occurring.
BTW any idea what the registration patterns in PA were at the time of the ’10 election? Just curious.
This might give us all some clue as to why it is Bob struggles so with scientific analysis of polling data. This statement would seem to imply that for Bob, if he likes the story the numbers seem to tell, no further analysis is required.
So let’s assume Bob has some empirical data to back up this claim. Let’s also assume that the data reflects measured number of years of formal education. Without even acknowledging it to himself, Bob is assuming that any “one year” of formal education is equivalent to any other “one year”. Probably doesn’t pass the common sense test. Since even Bob knows how a mean average is calculated, should we really conclude that as far as Bob is concerned one Democrat with 12 years of formal education plus one Democrat with 16 years of formal education is equivalent to two Republicans with 14 years of formal education? Averages out the same. But I’ll take one H.S. grad plus one college grad over his two college dropouts. Numbers can tell a story. But what story they tell depends an awful lot on what numbers we choose and how we use them to tell the story. Edward MacNeal’s book Mathsemantics is full of good examples.
@ 20
If you have a link to trends in educational levels in each party over time, that’s great. Please provide.
Was what I provided in in @14 inconsistent with what I said in @7? Was my source in @ 14 inappropriately selected?
I wasn’t trying to be super-accurate. The most educated in our society and the least advantaged in our society have in recent past consistently voted Dem. The middle ground is primarily GOP. My link confirms that.
I am aware of ‘average’ and how it may vary from one year to the next. I am under no obligation to micro assess data for your reading pleasure.
Jesus, this is what I mean about off-point @8. Can we get back to the fucking point of the thread, please?
@19:
see for yourself
Two timepoints per year, back to 1998.
It’s a whole lot better than you’ll ever be able to find for Washington, since we don’t even register by party.
@ 22
Thanks. Looks unchanged although I only looked at May 2010 to gauge the November 2011 results.
I guess we’re left with:
1. The poll accurately reflects registered voters.
2. A bunch of registered Democrats probably didn’t vote in 2010, throwing the elections to the GOP. Either that or they voted GOP rather than for their own party.
They might decide to show up this time. Or, less likely, if they voted GOP in ’10 they might still back Obama this year.
The former is more likely.
This beats the hell out of an open thread.
# 4: RR may have a point, regarding the potential disconnect between opinion polls and the election results.
I am on the “Do not call” list, so it is illegal to call me unless it falls within one of the exceptions. One of the exceptions, of course, is political calls. Starting about a week before each election, I get a flurry of telephone calls disguised as polls which are, at best, “push-polling”. I’ve also seen cell-phone calls disguised as “market surveys”.
I’ve hung up each time I realize it’s not a real call.
I’m wondering how many others are like that. In the past, I’d be willing to go through their polls, but starting two years ago it became clear that some folks are gaming the system, at which point I choose to no longer participate.
One point: Franklin & Marshall College is located smack in the middle of Pennsylvania, which is between the Democratic strongholds at either end of the state. Lansing isn’t that far from Harrisburg, Newt Gingrich’s childhood home, and the area is very conservative. There is hardly a tendency toward any “liberal bias” there.
As for “timepoints”, I tend to think that each election is unique, not a variation of any other, or compromise between them. Not only do events change in the world around them, but the candidates change and the electorate’s response to campaign tactics change. What worked last time usually doesn’t work the next time – at least not nearly as well.
That being said, between any generic black Democratic male running against a generic white Republican male, the race would be closer than in 2008. We don’t have the economic crisis to the same extent we did then, and the black vote isn’t going to vote with quite the same unanimous turnout it did in 2008. But it’s not 2010 either. And it’s not generic candidates we are measuring – we are measuring a real President against a specific challenger that has been unable to acquire any solid support within his own party other than “…at least he’s not a black Democrat….”.
@ 26
the black vote isn’t going to vote with quite the same unanimous turnout it did in 2008
I disagree. Very, very reliable Democrat voting demographic.
The polls will say what they say, but I think Obama will win in November. The math involved in polling is interesting, however.
N in Seattle,
Nice…you got part 1! Do you care to finish parts 2 and 3 for the win?
@12 Wow, 7 months of intense national campaigning have moved the GOPers’ registration numbers by 2/10ths of a percentage point. At that rate, they might be in striking distance of winning Pennsylvania in time for the 2044 election.
@28 I recently saw an article on MSNBC that basically said Romney doesn’t have a strategy to win the popular vote, all of his campaign scenarios that possibly get him to 270 electoral votes would fall short of a popular majority. That looks like a weak candidate to me. It also appears the smart money on Wall Street is assuming Obama will be re-elected. Unless something changes, I really think the GOP strategists will focus on retaining the House and capturing the Senate, plus expanding their grip on state-level offices. It wouldn’t surprise me if a lot of GOP shakers and movers have written off Romney.
@ 31
Obama’s strategy during the 2008 primary was to collect delegates by focusing on caucus states in which delegates could be concentrated by persuasion tactics, and not allocated according to popular vote. He wasn’t trying to win the most votes. He was trying to win the most delegates.
Did that look like a weak candidate to you, clown?
@29:
If you insist…
Differences (sample – best estimate, to integer percentage):
Definitions differ slightly in the last line (“independent” isn’t the same as “neither R nor D”).
Cause of “surprising” difference:
@32:
Apples to oranges.
Party nomination strategy and general election strategy are completely unrelated.
It looks like a candidate of a party that is sailing into demographic headwinds.
It looks like the candidate of a party that benefits from/depends on disenfranchising as many voters as possible.
It looks like the candidate of a party that is relying on fundamentally undemocratic features in the system (every state gets two votes just for being a state, regardless of the number of citizens there, just like the Senate) to undermine the will of the majority.
@ 35
If ONLY we didn’t have to adhere to that dastardly Constitution.
If ONLY Obama could camp out in CA and NY and be assured of victory because the majority of America’s territory wasn’t able to catch his attention on their numerous important issues such as farming and energy production.
Drat!
Liberal Lab Tech Beaker Cleaner apparently doesnt know the difference between a democracy and a representative democracy.
and yes, that darned constitution has always been a thorn in the side of the progressives…
WRT to original challenge posed by Darryl, which we are not discussing AT ALL so far:
This article states that Pennsylvania as of Jan 2012 had 4.1 registered Democrats to 3 million registered Republicans.
So the poll in question looks pretty far from an outlier …
Bob? Any comment?
Apologies to N in Seattle … missed your earlier post due to all the off-topic partisan bloviation.
No prob, Rael.
Repetition brings emphasis. At least, when minds are actually open to factual information (unlike the many deliberately opaque trolls hereabouts).
Please let the class know when you actually have some factual information.
Thanks in advance,
The Man
asshole @41:
Go tell the Secretary of State of the Commonwealth of Pennsylvania that the material I discuss in comments 12, 23, and 33 isn’t factual.
Preferably in person. And I suggest you walk to Harrisburg to do so.
Keep flailing putz, its comical to watch…
And maxie and bobbie, as expected, are utterly unacquainted with the concept of one person, one vote.
You and bobbie counter that the Electoral College is constitutional. Duh.
What you studiously avoid is the inherent undemocratic nature of both the EC and the Senate. Vestiges of slavery that in the present era give the same Senate representation to the 568,000 people in Wyoming as the 37,691,000 in California, and skew the EC such that there are 189000 citizens/EC vote in Wyoming, but 672000 citizens per vote in California.
You have no response because there is no defense of this. And it’s this very unjust system that Republicans have been deriving power from, and Rmoney is depending on in what will be his unsuccessful bid to be President.
Could your head be further up your ass?
“America’s territory” doesn’t vote.
American citizens do, to the degree that they can get past the obstacles put in place by Republicans to prevent people of color, young people, the elderly and others to vote in the first place.
Moreover, without the EC, every vote is equally powerful, and a strategy to win CA and NY only is as idiotic…as bub. In bub’s world the EC is the only thing that gets Affirmative Action celebrity Kenyan Socialist Obama to pay attention to Iowa, because he is so… different… from Iowans.
@ 44, 45
Which uniform gift do you prefer to receive first, popular vote or instant-runoff voting?
Neither of them are remotely possible.
It’s sad that Obama has to cross his fingers and hope he holds places like Nevada and Iowa when his real money comes from the coasts, but those are the rules of the game. Instead of wishing Al Gore could have been elected because he took more national votes than GWB43, work within the system. If the system says you need delegates and you can disproportionately win delegates by focusing on caucuses rather than elections in the primary season, then that’s what you do. It’s what Obama did in 2008 and it’s what Ron Paul did this year (stay tuned – apparently the Paulians will have a fairly major presence at the RNC convention this year).
Wanting it to be different is silly because you work within the current system.
The Founders had it right. Small, less populous states shouldn’t have their interests negated by larger ones on all issues merely because of population (this is the defense you said does not exist). It’s also why there will never be an amendment changing it: the smaller states will never vote to give up the leverage the Founders gave them and you’ll never get through ratification.
@ 38
I would re-iterate @2 and @16.
I have no issue with the accuracy with which the pollsters matched their sample with the party distribution of registered voters by party in the state. None at all. Seems spot-on.
But if the thread title includes the term ‘pundit’, and it seems readily apparent that the 2010 vote results did not reflect the party distribution of those registered to vote in the state, isn’t part of punditry looking past the demographics and at the actual recent results so that we can discuss why they are so different?
Since you asked me a question, I’ll return one to you, as well as to Darryl: Do you think that, if the election were held during the period in which the poll was taken, PA would have voted to elect Obama over Romney by a spread anywhere close to 12 points?
2010 was an off-year election with an angry/motivated right wing and somewhat complacent left wing.
2010 is the most *recent* election, true, but generalizing into the future based on that is probably no more accurate than doing so from 2008 (the most recent Presidential election we’ve had) would be.
If we were to generalize into the future based on the most recent Presidential election we’d have to predict a Democratic sweep … which doesn’t seem very likely.
@48
Fully agree with your first sentence. How much different do you think 2012 will be from 2010? The Tea Party is still going strong and Dems are in serious disarray.
I suspect we won’t see a 2008 again in our lifetimes. Anointing someone with no experience will be looked back on as a Teachable Moment, and a damned embarrassing one, in our history.
I guarantee voter turnout will be higher in 2012 than it was in 2010. Presidential election years ALWAYS produce higher turnout. That creates less opportunity for a motivated but disgruntled minority (Tea Partiers & right wingers in general) to sway the election.
2008 was an unusual year, yes.
2010 was a backlash to that year.
I expect 2012 will largely revert to “normalcy” with some characteristics of a secondary backlash to 2010.
Bob @ 49
I agree Obama has been a huge embarrassment…and even more embarrassing was the MSM failing to vet him and cheerlead him into office. It was disgraceful. Now Obama has a voting record and a record of actions including embarrassing appointments, debt increase of another $6 Trillion when he promised to cut it in half and high und employment. Afghanistan was a huge disaster and he left Iraq way too soon. He did kill some Al Quada leaders which was a good thing. But our Foreign Policy is a nightmare, especially with Iran and N. Korea. We have Solyndra, Fast & Furious and High Security leaks that had to come out from the White House somewhere.
Specific actions and consequences ought to be debated and vetted in the free press. But they’ve already shown to be in the tank for Obama again.
Let’s have the debates!!
@ 50
“Normalcy” hasn’t been seen since pre-election day, 2000.
2004 was characterized by a vocal although ultimately unsuccessful objection to GWB43, through the anti-war sentiment over Iraq although in general he was very much disliked by the left. Had a less inept candidate survived the Dem primaries, he probably would have been voted out of office. Recall The Guardian starting a reader letter-writing movement in an effort to sway voters in Ohio, which might have backfired big-time.
2008 was a hugely successful statement against the GOP and for ‘change’, whatever the latter was supposed to have meant.
2012 will be only slightly less massive in the upcoming effort to deny Obama a second term. Those on the right most opposed to what they have seen since 2008 are certainly disgruntled, absolutely motivated, and since 2010 are now organized and very, very well financed.
There will be nothing ‘normal’ about 2012.
Yes, voter turnout is always higher in a presidential year.
Low information voters… The DUMMOCRAPT friends. Low information voters…
DUMMOCRAPTS have no idea what is job growth. They hear the weekly jobs number and think it’s great. They never hear the previous week’s downgraded number the next week. From Gallup
I guess 60,000 jobs a month is a great thing if you are DUMMOCRAPT! Hence the higher poll numbers for Obummer across America. They reject anything about Obummer’s history. Low information voters will sway this election.