• Text smaller
  • Text bigger

The primary weapon in the media’s pro-Clinton campaign has been the polls.
We are reminded every ten minutes, it seems, that the president is hugely
popular. That means the GOP is radically out of touch, and that the views
of the American people are embodied in the opinions of Betty Friedan, Jesse
Jackson and Barney Frank.

Not that Congress should shun the unpopular. The press celebrates
politicians when they forge ahead to, say, raise taxes when public opinion
opposes it. This is called “responsible statesmanship.” On the other hand,
cutting federal arts subsidies or foreign aid that 9 out of 10 Americans
oppose is called pandering.

Why isn’t impeaching a perjurious executive an example of courage trumping expediency? Because, to the media elite, the will of the people is to be invoked when convenient and suppressed when necessary. If polls are
trumpeted loudly and often, you can be sure the message is for all
political dissidents to fall in line.

Grant that Clinton is more popular than he should be, especially given his
war crimes. Grant, too, that much of this has to do with the growing
economy. When people’s lives are improving, they understandably care less
about the minutiae of politics because politicians play a smaller role in
their lives. Even granting the conventional polling data, Clinton’s
supposed popularity reflects more indifference than cheerleading.

But let’s ask a more fundamental question: how much can polls about
political controversies be trusted? To answer that question requires
thinking about the methods used to conduct them. The most closely guarded
secret of polling these days is that fewer and fewer people are willing to
participate. Fully two-thirds of the calls placed to people’s homes result
in hangups.

If you’ve received a call from a polling firm, you know why people are
reluctant. There is nothing in it for you. It feels like an invasion of
privacy. You have no way of verifying the veracity of the caller. If your
political opinions are politically incorrect — that is, if you disagree with
the White House and CBS — you are far less likely to talk. An official
pollster might as well be from the Justice Department, for all the citizen

Hence, participants tend to have conventional opinions they feel safe in
spouting off to a perfect stranger on the phone. Most people are unwilling
to express an un-PC opinion at a cocktail party, much less to a pushy
character interrupting their dinner.

Pre-election polls provide a good test of all this. And they are less and
less able to predict actual results. The more unconventional an opinion
is — for example that a pro wrestler nicknamed “The Body” ought to be
governor — the less polls are able to discover. Political outliers, even if
they are in the majority, fly under the polling radar screen.

The question of whether a president ought to be removed from office falls
into the potentially dangerous category. If the person agreeing to the
poll senses that he will be regarded as a kook for saying the president ought to be tried and convicted, on the margin he will say what he is supposed to
say and not say what he is not supposed to say.

This thesis is easily tested. Find a medium that represents something of a
cross section of the population, where people can express their political
opinions without fear of reprisal or consternation. Compare answers on that
medium to the results of the typical phone poll. As it happens, in the last
year, massive internet news sites like CNN and MSNBC have become such

Phone polls show 65 to 70 percent (of 500 compliant people) favoring a
censure resolution in the Senate (the very thing the media are clamoring
for). But web polls show exactly the opposite. Between 65 and 70 percent of
participants (tens of thousands of willing clickers) want a full-blown
trial, and half say Clinton should resign immediately. In addition, the
results of these web polls fit with most people’s experience and the
knowledge they have of their neighbors’ opinions.

Now I know that these polls are regarded as mere entertainment. Sites
always have this caveat: they “are not scientifically valid surveys.” But
this is nonsense. In what sense is a phone poll of 500 self-selected people
browbeaten into saying what they are supposed to say more “scientific” than
an internet poll soliciting the opinions of tens or hundreds of thousands
of separate and anonymous mouse clickers?

Moreover, if there is any bias among web news users, it would tilt
leftwards: demographically they fit the characteristics of people with
relatively liberal opinions (more graduate degrees, more upper class, more
urban). Neither are internet users more libertarian in their politics, as
the old stereotype would have it. For example, polls asking about
government space gizmos routinely garner 75 to 85 percent support.

Even regular polls show a hardcore of one-third of the public that wants
to see Clinton ousted immediately. Add to that those who don’t answer, those who don’t reveal the truth, and those for whom the entire political game is utterly sickening, and you’ve reached the two-thirds mark and then some.

No, it is hard-core Clinton supporters — the Friedan-Jackson-Frank nexus — who are in the minority. The Republicans will be making a huge error if they
follow the media line about polls, which is designed to mask the disgust
most Americans have with Washington and everything associated with it.


  • Text smaller
  • Text bigger
Note: Read our discussion guidelines before commenting.