How has the image of Vice President Dick Cheney fared after the South Texas shooting incident? Several polls conducted since the incident have attempted to answer that question.
In one instance, I think there could be some misleading inferences made on the basis of a comparison of Cheney's job approval rating with President George W. Bush's. Let me explain.
A TIME magazine poll, conducted Feb. 15-16, showed Cheney with a job approval rating of 29%. The same poll had Bush's job approval rating at 40%.
Some news reports, including TIME's own report of its data, emphasized that Cheney's approval rating did not fall significantly between the TIME poll's late November, early December poll (when it was 32%) and the mid-February (post-shooting incident) poll. That is important and valid information. Both Cheney job approval questions were asked using the same format and in the same way, and the trend comparisons are certainly legitimate. This evidence suggests that the incident probably did no lasting harm to Cheney's image (other trend data suggest the same thing).
But there is some danger if observers try to compare Cheney's job approval rating directly to Bush's, and then imply that the 11-percentage point difference between the two shows that the veep is less highly evaluated than his boss. Here's an example from CNN's Situation Room from Feb. 20:
[WOLF] BLITZER: I suspect we're over this. But one thing that the vice president is not over with, which is his job approval number. In this new "TIME" magazine poll, he's got 29 percent job approval. The president had, what, 39 or 40 percent. Twenty-nine percent, that's really low job approval.
Now while Blitzer didn't make the comparison explicitly, his comments seem to suggest that the 29% for Cheney is "really low" by virtue of the fact that it is 11points lower than Bush's.
But any inference that Cheney is underperforming Bush in terms of his job approval rating -- based on the TIME poll data -- is not legitimate.
Here's why.
First, the questions asked about Cheney and Bush are different. The precise question wording used by TIME about Cheney is "Do you approve or disapprove of the job that Vice President Dick Cheney has done in the Bush Administration?" The precise TIME poll question asked about Bush was: "In general, do you approve or disapprove of the way President Bush is handling his job as president?"
Subtle, but real differences.
This leads to the second point.
The "don't knows" on the Cheney job approval question are extraordinarily high at 29%. This is the highest "don't know" for the vice president of any recent job approval rating for Cheney that I'm acquainted with. Gallup's last asking of the job approval rating for Cheney, for example, produced a 12% "don't know" in January.
But it would appear that there is something about the way the TIME poll words its question that causes the higher "don't know." According to pollingreport.com's compilation of previous trends from the TIME poll, the Cheney job approval was asked differently prior to last fall, using the more standard: "Do you approve or disapprove of the way Dick Cheney is handling his job as vice president?" wording. The three times the question was asked by TIME using that format, the "don't knows" ranged between 11% and 18%. (These three polls apparently used registered voters, which arguably could have the effect of lowering the "don't knows" slightly, but I don't think enough to account for the difference.) The two more recent times the TIME poll has asked the question using the new wording, the "don't knows" were 23% and 29%.
Why does the high "don't know" percentage matter? Because a high "don't know" percentage drives down the percentages on both the approval and disapproval side of the ledger.
Suppose for a moment that a public official is known by only 30% of the population. In other words, a 70% "don't know." Suppose, however, that everyone who knows that person likes him. He would then have a 30% approval and a 0% disapproval rating, coupled with the 70% "don't know."
Now suppose there is another public official who is known by 90% of the population, or in other words a 10% "don't know." Suppose only a third of the people who know him like him, while two-thirds disapprove. This personality, too, would have a 30% approval rating, but a 60% disapproval rating. If only the approval ratings for the two personalities were compared, however, they would appear to be identical. And that would be misleading.
The "don't know" percentages usually aren't a problem for well-known figures because they are universally high. When one compares approval ratings between two individuals with significantly different "don't knows," however, it can be a problem.
A quick mathematical calculation using the TIME magazine data shows that Cheney's approval rating among those who know enough about him to have an opinion (70%) is 41%, while Bush's approval rating among those who know enough about him to have an opinion (94%) is 43%. That's not a statistically significant difference. In other words, Cheney and Bush have virtually identical approval ratings among those who have an opinion.
So I think the best conclusion about Cheney at the moment is that he is just about as well-liked as his boss. That's supported not only by the TIME data, but also by our Gallup Poll trends. It is incorrect to imply that a comparison of Cheney's 29% approval rating in the TIME poll with Bush's 40% approval rating shows that Cheney is not as well-liked as his boss.