The Washington Post reports on the latest Post/ABC poll, which came out today:
President Bush’s approval rating has surged in recent weeks, reversing what had been an extended period of decline, with Americans now expressing renewed optimism about the future of democracy in Iraq, the campaign against terrorism and the U.S. economy, according to the latest Washington Post-ABC News Poll.
Bush’s overall approval rating rose to 47 percent, up from 39 percent in early November, with 52 percent saying they disapprove of how he is handling his job. His approval rating on Iraq jumped 10 percentage points since early November to 46 percent, while his rating on the economy rose 11 percentage points to 47 percent. A clear majority, 56 percent, said they approve of the way Bush is handling the fight against terrorism — a traditional strong point in his reputation that nonetheless had flagged to 48 percent in the November poll.
This was a “random adults” poll, which means that it understates the President’s support among voters. The ABC site contains more information about the poll’s internals than the Post’s. ABC says that Republican self-identification is up 6% since their most recent survey, to 33%. They also say that 87 percent of Republicans approve of Bush’s job performance, while 15 percent of Democrats and 38 percent of independents approve. We have lots of mathematicians among our readers (Dafydd ab Hugh, for instance); I’m guessing some of them will write in and tell us what percentage in the poll were Democrats, and what percentage were independents. My crude calculation is that respondents were split almost exactly equally among Republicans, Democrats and independents.
It’s remarkable how much it helps to fight back against the Democrats’ calumnies. Events have contributed to the President’s resurgence too, of course.
UPDATE: Dafydd, on cue, writes:
There are three variables (R, D, and I), but only one equation. Thus, without constraints, the problem cannot be solved. However, taking the null hypothesis constraint (R=D=I=0.33333…), the equation balances; so it’s definitely possible that, as you suggested, they polled equal numbers of each.
I also tried a couple of other scenarios: assuming independents were 50% and the Republicans and Democrats were each 25% yielded an answer much too low; while assuming R=D=0.40, while I=0.20, yielded an answer somewhat high.
So it appears that the best guess is that the numbers were more or less equal of each of the three groups.
Which means, I think, that my crude trial and error approach wasn’t too far off the mark. And if the ABC/Post poll contained equal numbers of Republicans, Democrats and independents, it should be fairly accurate.