There are a lot more lefties and center-lefties in America than polls show. Here are three reasons why:
1. Self-identification does not mesh with policy preference
Ideological self-identification does not accurately place Americans into categories representing coherent worldviews. This is a point that the Pew survey itself makes, though it has received less attention than the topline numbers:
Still, ideological labels do not always predict opinions about key policy issues. For example, about half of Americans who describe their political views as conservative say that all (24%) or some (27%) of the tax cuts passed under George W. Bush should be repealed. More than four-in-ten conservatives (43%) say that abortion should be legal in some or all cases. On the other hand, nearly half of self-described liberals (49%) favor more offshore drilling for oil and gas in U.S. waters.
Just because someone describes him or herself as a liberal, moderate or a conservative, does not mean that person actually is a liberal, a moderate or a conservative. Data supports this conclusion as well. The long-term trends of the Harris poll shows that ideological self-identification has changed little since 1978, with the percentage of "conservative" self-identifiers remaining between 34%-38% in all but two of the last 30 years, and with the percentage of "liberal" self-identifiers remaining between 17%-20% in all but one year. However, political scientist James Stimson shows that despite this apparently static national ideological mood, the national inclination toward public policy has actually shifted dramatically and frequently during the same time period:
Although it is more expedient in terms of public polling to simply ask people if they are liberals, moderates or conservatives, a more accurate reading would involve a battery of questions about general worldviews and specific policy inclinations. For example, the more accurate ideological reading of the electorate in the 2008 exit poll came not from the ideological self-identification question, but instead from the following classic public vs. private sector question:
View of Government
Should do more: 51%
Should do less: 43%
Although it leaves out cultural issues, that is about as straightforward an ideological positioning question as one can possibly ask. The result revealed a slightly center-left nation, quite different from the ideological self-identification question.
2. The Rise of Progressives
The last ten years has witnessed a marked rise in the use of a different left and center-left ideological term in America: progressive. This rise has been so pronounced and widespread that "progressive" is now the most favorably viewed ideological term in America, surpassing both "moderate" and "conservative."
July 2007 Survey By Rasmussen Reports
| Label |
Positive View |
Negative View |
Net Favorable |
| Progressive |
35% |
18% |
+17% |
| Moderate |
29% |
12% |
+17% |
| Conservative |
32% |
20% |
+12% |
| Liberal |
20% |
30% |
-10% |
There are now two competing ideological labels for the center-left and left in America, liberal and progressive. The chic term, "progressive," is viewed far more favorably than "liberal," but it is not offered as a choice in public opinion surveys or exit polls. Given the enormous difference in favorability ratings, it isn't even remotely a stretch to argue that more people would identify as a "progressive" than as a "liberal" if the term were offered in public opinion surveys. The absence of "progressive" in such surveys thus creates a false sense of a national ideological advantage for conservatives, as being a "progressive" is viewed more favorably than being a "conservative."
3. The Liberal Bradley Effect
The third way ideological self-identification polling numbers skew toward conservatives is what I term the "Liberal Bradley Effect." In terms of ideological self-identification, the gap between liberals and conservatives is always much smaller in Election Day exit polls than it is in national telephone polls. This indicates that people might feel less social pressure to identify as either conservative or moderate on Election Day than they do over the phone and in the abstract.
For example, in 2000, the conservative posted a 9% advantage on liberal in the exit poll, despite holding a 20% advantage that year according to Pew and a 17% advantage that year in Harris. In 2004, the exit poll gap was 13%, even though it was 18% in both Pew in Harris. In 2006, the exit poll ideological gap was 12%, even though Harris pegged it at 18%. There are consistently more liberals, and fewer conservatives, on Election Day than in abstract telephone surveys. Given the negative public views of liberals noted in polling data above, it isn't a stretch to argue that this gap is the result of a "Liberal Bradley Effect," where many people do not want to admit they are liberal, and many others claim they are conservative, in order to maintain public appearances.
Conclusion
The direction of the national ideological shift was better described by a far more detailed survey undertaken by Pew two years ago. The survey, with topline results alone more than 100 pages in length and supported by more than twenty years of data, identified a leftward ideological drift in America:
Increased public support for the social safety net, signs of growing public concern about income inequality, and a diminished appetite for assertive national security policies have improved the political landscape for the Democrats as the 2008 presidential campaign gets underway.
At the same time, many of the key trends that nurtured the Republican resurgence in the mid-1990s have moderated, according to Pew's longitudinal measures of the public's basic political, social and economic values. The proportion of Americans who support traditional social values has edged downward since 1994, while the proportion of Americans expressing strong personal religious commitment also has declined modestly.
Pew will release another one of these detailed "core values" surveys early in 2009. At that point, we will be able to determine if this leftward drift has continued, accelerated, or reversed. Until that time, try to avoid looking at ideological self-identification numbers, as they are a particularly poor measurement of the national ideological mood. |
STATE BLOGS
