Public Opinion Watch

Ruy Teixeira
Ruy Teixeira

(covering polls and related articles from the weeks of June 20-26, 2005)

In this edition of Public Opinion Watch:

Bush Continues to Weaken
• Hispanics Poised to Move Democratic

Bush Continues to Weaken

The latest ARG poll has Bush’s overall approval rating at 42 percent, joining several recent polls that have had his approval rating that low. A sub-40 approval rating from some public poll seems likely to appear fairly soon.

Some have argued, however, that a sub-40 Bush approval rating is unlikely to appear because his high support among Republican identifiers makes it difficult for his approval rating to drop much more that it already has. I don’t believe that is the case.

To begin with, though it depends on the poll, there is still considerable room for Bush’s approval to fall among independents. In the latest CBS News poll, his approval rating among independents is 37 percent. Given that his approval rating in that poll was 42 percent, if his approval were to fall to around 30 percent among independents and all else remained equal, his overall approval rating would fall to below 40 percent.

Just as important, the assumption that Bush’s approval rating among Republicans will remain steady is unwarranted. For example, if you compare his rating by party ID in the latest CBS poll to his rating by party ID in the late February CBS poll, his approval has fallen from just over 90 percent to 84 percent, a decline of 7 points. That’s almost as much as the analogous decline among Democrats (8 points) and actually more than the decline among independents (5 points).

You see a similar pattern in a number of other polls. Bush’s approval rating among Republicans has fallen in recent months from around 90 percent to around 85 percent. It is entirely possible it will decline further if the difficulties of the Bush administration continue to deepen. Certainly, there is no sound reason to suppose Republican identifiers will somehow be immune from overall political trends.

The ARG poll provides additional abundant evidence of Bush’s weakness, especially in the economic realm. Bush’s economic approval rating is down to 37 percent with 59 percent disapproval, his worst rating ever in this poll. Just 22 percent think the economy is getting better and 58 percent believe it is getting worse. Only 26 percent believe the national economy will be better in a year, compared to 50 percent who believe it will be worse.

In terms of rating the national economy today, 63 percent say it is bad, very bad or terrible, compared to 35 percent who say it is excellent, very good or good. Moreover, for the first time since considerably before the 2004 election, about as many say the economy is in a recession (44 percent) as say it isn’t (45 percent).

As for the household financial situation, more (52 percent) see their personal situation as bad to terrible than see it as excellent to good (46 percent). And exactly half say their personal situation is getting worse, while only 17 percent say it is getting better. Nor are people more positive about the future: looking ahead a year, essentially identical proportions expect their situation to get worse or better.

How about Iraq? Not much help there. According to a new Ipsos-AP poll, a majority of the public (53 percent) now believes that the U.S. made a mistake going to war there in March of 2003, compared to 42 percent who think we made the right decision. And, while the public does not favor immediate withdrawal (just 37 percent say they do in the Ipsos-AP poll), a new Harris poll finds a strong 63-33 majority favoring the more modest goal of “bringing most of our troops home in the next year,” rather than “keeping a large number of U.S. troops in Iraq until there is a stable government there.”

More generally, a new Zogby poll finds Bush languishing in net negative job approval territory in each and every area tested by the poll: overall (44 percent positive, 56 percent negative, for a -12 net); the war on terrorism (49, -1); the war in Iraq (39, -22); taxes (36, -26); foreign policy (36, -25); jobs and the economy (35, -30); education (33, -31); environment (30, -36); and Social Security and Medicare (27, -42).

OK, Bush is tanking….but how much is that likely to hurt the GOP in 2006? That’s still a long way away, but here’s something to think about. The new EMILY’s List report on a large-scale survey by Garin-Hart-Yang and The Feldman Group points out that working class (non-college-educated) white women (whom I have maintained was the key group that swung to Bush and the GOP in the 2004 election) now support the Democrats by 18 points in a prospective 2006 Congressional matchup. In 2004, this same group supported Bush by 18 points and House GOP candidates by 15 points.

Wow. That’s a huge swing. If anything like this holds up in 2006, the Republicans are in big trouble.

Hispanics Poised to Move Democratic

It’s early days, but Hispanic voters, despite the (real but typically exaggerated) GOP progress with these voters in the 2004 election, appear ready to move back toward the Democrats in 2006.

Let’s review the bidding. In 2004, Kerry carried the Hispanic vote 58-40 (using the figure from the combined NEP state exit polls, rather than the discredited national exit poll figure). Adjusting the NEP House vote data to reflect a 58-40 presidential split suggests that Democratic House candidates carried the Hispanic two-party vote by about 59-41.

So the Hispanic vote in 2004 was decisively Democratic, but less so than in recent years. For example, in 2000, Gore carried the Hispanic vote 62-35 and in 1998, 2000 and 2002, Democrats carried the Hispanic House vote by 63-37, 65-35 and 62-38, respectively.

These figures do indicate that 2004 was a good year for the GOP among Hispanic voters in comparison to the recent past. But the extent of GOP progress is sometimes exaggerated by using 1996 as a benchmark for comparison to both 2000 and 2004. This is problematic not only because Clinton’s 72-21 margin in 1996 was anomalously high, but also because Hispanics were defined differently in that year than they have been subsequently.

This difference is not widely understood and deserves some explanation. Here is the basic story: prior to 1998, the exit polls used a single race question (“Are you white, black, Hispanic/Latino…..”) to capture Hispanics, as opposed to a race question plus another question on whether the respondent is of Hispanic descent or not, which has been included on exit polls since 1998. The change in methodology allows the exit polls to capture more Hispanics, but since those Hispanics who do not identify themselves as Hispanic in the race question tend to be more conservative than those who do identify themselves as Hispanic in that question, it makes the expanded sample of Hispanics post-1998 more conservative than the pre-1998 samples.

Got that? Therefore, to compare pre-1998 Hispanic exit poll figures to post-1998 Hispanic exit poll figures is a little bit like comparing apples and oranges. A better comparison can be obtained by looking at just the Hispanics who self-identify in the race question, since that is common to all the exit polls.

With such a comparison, 1996 remains the high point, but the fall-off to 2000-04 is less severe. Indeed, the Hispanic presidential vote, defined in this way, has averaged 64-35 Democratic in these two elections, actually more strongly Democratic than in the two Reagan elections of 1980-84, when the Hispanic presidential vote averaged 61-35 Democratic=

And in the next election following Reagan’s relatively good performances among Hispanics–1988–the Hispanic presidential vote moved sharply Democratic to 69-30. Don’t be surprised if we see the same kind of trend in 2008.

Or 2006 for that matter. Consider the results of the new Democracy Corps poll of Hispanic voters. In that poll, Hispanic voters who express a preference for the 2006 congressional elections currently give the Democrats a 68-32 edge in the two-party vote. The rest of the poll tells us why Hispanic support for the Democrats has become so lopsided.

Hispanic voters give Republicans an average feeling thermometer (0=coldest; 100=warmest) score of 48 and Democrats an average score of 60. And 63 percent of these voters identify with or lean towards the Democratic Party compared to just 31 percent who identify with or lean towards the Republicans.

Hispanic voters were also asked which party they associate more with a number of positive characteristics. In no case did the Republicans have an advantage over the Democrats, even on national security-related items. Here are the characteristics with Democratic advantages in parentheses: accepting different cultures (44); addressing the concerns of the Hispanic community (41); for the middle class (39); support working families (35); cares about people (34); putting the public interest first (33); on your side (28); opportunity (26); for families (22); shares your values (18); freedom (15); prosperity (10); respecting religious faith (10); personal responsibility (8); know what they stand for (7); and can be trusted to keep America safe (4). When Republicans don’t even have an advantage on that last item, you know they’re not doing too well.

But perhaps Hispanics have more confidence in President Bush who, after all, did do relatively well among Hispanics in the 2004 election? Not by these data. Bush’s disadvantages relative to Democrats track pretty closely with his party’s disadvantages relative to Democrats, right down to “can be trusted to keep America safe,” where Hispanic voters give the Democrats a five point edge.

Doesn’t sound like Bush can be counted on to reverse his party’s currently poor image among Hispanics. And the rest of the poll suggests little in the way of issues (with the possible exception of education) that seem likely to boost the GOP.

Instead, the poll provides strong evidence of Hispanic voters’ dissatisfaction with the Iraq war and the economy and considerable interest in Democratic messages around issues like health care, economic opportunity and even stem cell research. Thus, Hispanic voters do appear poised to move Democratic in 2006, with very little pushing them toward the GOP. It’s up to Democrats to make sure this potential trend becomes a reality.

Ruy Teixeira is a joint fellow at the Center for American Progress and The Century Foundation.