Over the years whenever
you read an article online from a publication, whether it from a right or left-wing
media, a political organization, the mainstream media and even some political comedians,
you often here a certain phrasing. This has to do with so many of the hot
button issues of today’s society: gun control, universal health care, income inequality,
legalization of certain drugs, and of course abortion and the death penalty.
So often when these
arguments are phrased by so many they come with the same subhead: “A vast
majority of Americans believe that…” Particularly
when it comes to so many of the progressive or Democratic platform ideas, it’s
argued the same way. “An overwhelming number of Americans, a vast majority of
Americans, most Americans,” pick your phrasing.
The intelligent person –
who as we all know these narratives are argued for – knows that there is a critical
word missing from that phrase after Americans. That word is ‘polled’. Most of
us do understand the difference - no
organization has to power to interview every single person in America. The
problem is a sin of omission is still a sin. We know all too well that polling
has always been an inexact science and we also know that phrasing can make a
difference in any number of issues. But somehow not only the left, but the
media and so many other supposedly smart people, have decided that polling on
so many issues should essentially be carved in stone and considered policy when
it comes for every single legislator, state and national, to consider their agenda.
The left, in
particular, loves to lean this argument when they argue that Republicans and conservatives
with their agenda and policy are going against the will of the people. They
will point to any poll that they can fight (even Fox News if it suits their
purpose) to demonstrate yet again that they are correct and the Republicans are
wrong.
But just as the devil
can quote Scripture if its serves his purpose, anybody can quote polling to serve
theirs. The left loves to leans on the
idea of polling as fact, when in fact it is nothing more than statistics. The
media argues that it is de facto the voice of the people when the average poll
is somewhere between 1000 and 1500 surveyed. And remember that number is based
on the ones who responded to the poll. We never know how many people might have
other opinions but who just don’t have the time and energy to waste on a phone
call. And that, of course, is all the many ways a survey can be weighted to
make it say what the pollster or the person polling wants the message to be.
Once again the most
accurate source to deal with how we regard polling is The West Wing. Aaron
Sorkin knew how important polls were to politics; one of the most beloved
recurring characters during the show’s tenure was Joey Lucas, played by Marlee
Matlin. The Bartlet Administration and Josh Lyman in particular trusted her
intellect and ability immensely. But even she admitted once to Josh that she ‘didn’t
know what those numbers said.’
And the administration
itself knew just how ridiculous the numbers could be. In the aftermath of the
attempt on Bartlet’s life in Season 2, the staff was marveling over Bartlet’s
81 percent approval rating. They acknowledge they’re soft. CJ tells the staff: “A
week ago, the job approval’s 51. We get shot at it and it’s 81.” Sam follows up
with:
“When asked ‘whose
approach on important national problems do you think generally best, President
Bartlet or the Republican leaders of Congress,’ Bartlet’s gets 61 %
Bartlet responds. “Well
19 percent of the country has clearly made up their minds about me; 20 percent
just feel sorry for me.”
In an episode in Season
4, Josh is appalled to know that they have lost the vote of a Senator on foreign
aid because of a poll in the New York Times that says that ‘68 percent of
Americans think we spend too much on foreign aid, and 59 percent say that
amount should be cut.” Josh takes this in and repeats over and over throughout
the episode. At one point he’s told he really loves that figure and is asked
why. Josh explodes:
“Because nine percent
of Americans think we spend too much and it shouldn’t be cut! Nine
percent of Americans polled could not wrap their heads around the question!”
I have always loved that
particular line more and more as the years go by. Because if that line doesn’t prove
that so much of what politicians, the media and lobbyists consider gospel is
based on the basic inability of phrasing, I don’t know what will.
This brings me to the
larger point of this ‘post-truth America’ when it says that so much of the
country is entitled to their own truth but not their own facts. The problem
comes that so much of what the media and the political industrial system
consider facts are really statistics and most of those statistics come from
surveys done by specific organizations of a sampling of Americans.
Now I need to be clear
I do believe in the accuracy of polling at other levels: congressional and
Senatorial races, Presidential primary polling and election polling. I don’t
believe it’s anywhere near perfect and I know that, if anything, it can be
weighted even more based on whose doing the polling, but over the last decade I’ve
found it mostly accurate.
The problem I have
comes when so many people – and this is mostly a Democratic and progressive
issue – want to create a national legislative agenda based on polling. This is
the rule rather than the exception on every swing issue as long as I’ve been following
it and if you can’t find the flaw in trying to argue the popularity for an
issue of national importance based on a polling of a few thousand people (at
most) then you’ve basically deciding to ignore the importance of math.
Not the least of these
problems has been how both sides then to view something as a winning issue. In
recent years Democrats in particular love to glam on to the idea that so much
of their policy, from raising the minimum wage to gun control to education, is
favored by a ‘majority of Americans’. And they’re not lying in many cases: the
polling often fits the dictionary definition of a majority, which is 51 percent.
51 percent, last I checked, is a failing grade everywhere but politics and even
in this limited construct, it barely counts as the teeming masses.
Speaking for myself, I have
always considered a safer number to be something like 75 to 80 percent. I might
be willing to go as low as 60-65 percent because that’s a landslide in today’s
politics. But if you’re going to tell me that ‘a vast majority of Americans
favor affirmative action’ and you list as evidence a poll that says 54 percent do, I am going to
want to know more before I say that you have the will of the people.
Then there’s the problem
of trying to argue that so many surveys taken are in favor of this. Now I’d be
more willing to take this seriously if rather than one giant conglomerate poll –
and as we know, the larger the sample size, the more inaccurate these polls can
be come – then, I don’t know, a poll for all 50 states. I am all too aware
these states are not all the same size so perhaps we should try to get a proportionate
number for each one. The only way we can truly know how the nation feels on any
major issue is taking a sampling from every state. I imagine that people in
Idaho and Utah are going to feel very different about gay marriage and abortion
then people in California and New Jersey. I also know that in blue states and
red states, you will not get 100 percent agreement on anything and the
important part is to figure out where the nuances are in state-to-state.
But who am I kidding?
The left wouldn’t set foot in any of these states other than to mock them as backwards
and undeserving. I’ve read so many articles in so many sources – not all of
them political – mocking and berating every single aspect of a state in the
South, the Midwest, the West, basically everywhere that isn’t either on the coasts
or a swing state as backwards, full of white trash and MAGA extremists, all of
whom deserve to suffer the horrible lives they have because they are dumb
enough to either vote for Republicans or not live in a blue state. I have no
doubt that there are African-Americans, women, LGBTQ+ and maybe even Democrats
in all of these states but as far as the left goes they don’t count because
they live in the 2500 sections of the country that Trump won where (according
to Salon) ‘nobody lives’ rather than the 500 sections that Biden won where ‘everybody
lives’. I can’t imagine any progressive organization worth a damn has even tried
to do phone calls polling in Wyoming or Arkansas, much less tried to run a
voting drive there.
The only breakdowns in
polling that the left cares about is, of course, political affiliation. And
there they only really care about what Democrats think. The Republicans are
lost to them and the independents are untrustworthy because they haven’t fully
committed to the leftist cause. It’s nice if some of them agree with the views
they have, but they don’t really care one way or the other; they’re certainly
not going to make an effort to win them over in the next election.
The media thinks that
polling by party affiliation is the best way to get a national perspective. It’s
the worst because it plays in to just how statistics work. There’s an old joke
about statisticians doing archery practice that you probably all know but here
it is anyway:
“Three statisticians
are practicing archery. The first, after much work, shoots his arrow ten feet
to the left of the target.
The second, after
similar calculation, shoots his arrow ten feet to the right of the target.
The third starts
jumping up and down: “Bullseye!”
That’s what weighted
statistics are, and its especially true when it comes to politics and the kind
of people you survey. Say Anderson Cooper went on CNN and said a recent poll
said that 53 percent of all Americans believe that we show forego any trial of
Donald Trump and we should just burn him at the stake in front of a nationally
televised audience while the Biden family sings campfire songs in front of it. Cooper
could then say that this is a point of view held by a bipartisan coalition of
Americans – 98 percent of all Democrats favor it as well as 8 percent of all
Republicans. It averages out that way, but it’s not a true perspective of the
landscape and anybody who tried to use that survey as part of their argument
would be a fool – though I could see some people trying.
This is an extreme
example I grant you but it is still of the same kind of model that is being
used by pollsters across the country, echoed by the media and reported by
late-night comics to argue an agenda as fact.
Like everything else in
the world a poll is meaningless without context. And context is one of those
things that is harder to find. USA Today can preach that a vast majority of
Americans believe in some form of gun control. What they mean is usually that,
out of 1000 to 1100 people surveyed 55-60 percent believe this. They might be
able to give certain details on the question of their political affiliation but
that’s meaningless without knowing anything else.
No comments:
Post a Comment