CNN and The Gallup Organization - Leading You By the Nose

CNN, USA Today, and The Gallup Organization:
Leading You By the Nose
by Jeff Koopersmith

Tuesday, November 25th, 1997 (APJP) -- CNN, USA Today and The Gallup Organization are giving the White House an early Thanksgiving Turkey this week publishing a farcical survey of Americans which supposedly tells us that "most" of us want Janet Reno to appoint an independent counsel to investigate whether Bill Clinton and Al Gore are suitable for Club Fed.

Let's take a look at the results.

As usual, Gallup, CNN and USA Today paint an incomplete and therefore inaccurate picture. Gallup polled 1,019 voters over 3 days on several unrelated issues. What questions they asked, how they were asked, and to whom they were ask is not clear. But let's assume that the Gallup organization, which, to my mind, lives mostly on its founder's name, randomly selected respondents in a statistically valid manner and that no news broke over the 3 day period .during which the survey was conducted which might skew the results. After all, you know us Americans, we change our opinions as often as a chameleon changes color in a microwave.

In short, three days can be a long time. A serious poll might question the 1,000 people over a few hour period, not a few days. Certainly, so much news of Iraq was flowing at the time that nothing reliable could result from questioning Americans bombarded by political pap from a variety of spin doctors and irresponsible newscasters.

Of course conducting a poll of this size over a few hours is more expensive to do since you need a lot more telephone interviewers working at the same time. But heck, who cares, just get me some figures.

Bill Clinton - Doesn't care what Gallup says, he's got Mark Penn, a RELIABLE pollster.

It immediately strikes one as odd that, according to Gallup, 60% of Americans -- hopefully identifying themselves as voters -- think that the Attorney General should appoint an independent counsel to investigate Clinton and Gore, yet if she chooses not to, that's okay too.

What does that tell you? Well, for starters, the question was most likely improperly formed. For instance, if I ask you whether you would eat fertilizer, 98% of you would answer no. However, if I asked you whether you would eat fertilizer because it has been proven that such a diet will cure cancer 98% would say yes.

Here, what appears to be a schizophrenic reaction to a political poll, could really be a truthful set of responses to a poorly framed series of questions.

I think that's the case. However, whether the poll was purposely poorly constructed or not is a matter for you to decide.

One thing's for sure, Gallup should have asked the second question first. I'm not sure they didn't, but I suspect not. If they posed the question, "Would you trust Janet Reno to decide whether or not to appoint an independent counsel for Clinton/Gore?," it appears that around 60% would say yes - We trust Janet. If Gallup then asked whether each respondent thought a counsel should be appointed despite Reno's opinion , it's an odds on bet that far fewer than 60% would have said yes to that question as well.

Why? Because they've just been reminded that they trust the Attorney General to "do the right thing" and to now say they would appoint a prosecutor would seem at least a bit reaching.

So, you see, even the order in which questions are asked is important.

Al Gore - a victim of Gallup?

Another thing to look at is the what might be the hard core anti Clinton/Gore element within the tested population. You can see from the poll results that respondents were also asked the main reason that Janet Reno would not investigate Clinton and Gore. 57% answered that the Attorney General's choice would be based on the facts, 36% said "to protect Clinton." In Gore's case, 53% said the facts would sway her and 36% -- the same percentage as Clinton, said "to protect Gore. It doesn't take a genius to figure out that those 36% are dyed in the wool Clinton/Gore opponents.

But what if Gallup had given respondents more choices? For instance, what would have happened with the hard core 36% White House Opposition if this choice had been offered:

"If Janet Reno does not choose to have Bill Clinton/Al Gore investigated would you trust her judgment?" -- followed by -- "Would you appoint a special counsel to investigate Clinton and Gore?" The answers would be totally different.

Janet Reno: She'll do what
she thinks is right, and
Americans know it.
Something else interesting to me is USA Today's and CNN's failure to underscore that even this skewed poll showed a ten percent drop in those who think an independent counsel should be appointed. In October, based on the exact same question, the number was 66%, today it's supposedly 60%. Are Americans realizing that the campaign finance "investigation" is a political sham, or are they just tired of one independent counsel after another coming up with zilch? Your guess is as good as mine, of Gallup's for that matter.

CNN/USA TODAY/GALLUP also tested American opinion on Iraq. Here again we see this "split mindedness" on the part of respondents. But it really isn't split mindedness at all, just a poorly framed questionnaire.

Gallup says Americans are "skeptical" on Iraq policy and that a majority of Americans felt that only military force will prevent Iraq from developing "weapons of mass destruction" yet 54% of them approve of the way the situation has been resolved thus far.

Huh?

How can that be? We haven't used military force in Iraq as yet this year. Does this mean that Americans think it's not important to prevent Saddam from developing these weapons? Or perhaps they think nothing really could stop their development in the first place. Or, maybe they just don't care.

But again we see at least what looks like the split brain, but what is in actuality a poor question format and perhaps questions again asked in the wrong order.

My main concern with the CNN/GALLUP poll was asking this next question to begin with, and then compounding this insult by broadcasting the results worldwide:

Does "X" Mark the Spot for Americans? I doubt it.
Gallup actually asked whether Americans were for or against ASSASSINATING Saddam. I thought this was a nation of laws, and it is against international and U.S. law to order the assassination of anyone let alone a head of state. Regardless of how you view Saddam Hussein, even asking about assassination is a display of total disregard for the law, not only of man, but of God for you Promise Keepers.

A whopping 39% of Americans think we ought to bump off Hussein, with, thankfully, 55% saying no according to Gallup. CNN points out that during Desert Storm 60% thought we should murder him.

Just the tone of the Gallup questions is outrageous. Should the US assassinate Saddam? Should the U.S. assassinate Saddam only after he's been removed from power? Should the U.S. have "removed" Saddam from Power before ending the war?

What are these nitwits talking about? I repeat Mr. Gallup, IT IS ILLEGAL TO ASSASSINATE SADDAM. We don't have the capability to "remove" Saddam from power. It's that simple.

But did Gallup explain any of this to their respondents?

No.

So, you see, despite the fact that "The Gallup Organization" gets a heap of money for conducting these polls, the results are weak at best, and totally misleading at worst. I'd choose worst.

Americans are not stupid. If a pollster explained reality to them before asking what are just out and out ridiculous questions, we would not respond with schizophrenic answers.

The bottom line...

Don't trust pollsters unless their name is Penn and Schoen. This isn't a plug, although I've never used another polling firm. But I assure you that neither Mark Penn or Doug Schoen would allow me to frame questions like this without pointing out the folly of the result. Good pollsters, in my opinion, should not permit their clients to hype poll results based on fantasy.

By the way, I asked Sarah Van Allen of Gallup in Princeton if I could see the questions and what's called the cross-tabs of this poll. She never answered. That's not the first time. Here's a copy of my e-mail.

Subject: Sarah, could I see the questions on The CNN/Reno poll?
From: American Politics Magazine <ampol@americanpolitics.com>
Reply-To: editors@apj.us Organization: American Politics
To: Sarah_Van_Allen@gallup.com
Dear Sarah:
Hi. Wondering if we could take a look at the cross tabs on the CNN/GALLUP poll re Reno appointing a special counsel?
If not, how about the exact questions posed?
Thanks
Managing Editor
American Politics Journal

I should have added. "Sarah, don't publish polls like these. It's misleading, it's dangerous and it's, well ... Un-American!"

So there.

PS: Never, ever trust a corporation with the word "Organization" in it's name. Think about it.