Today surveys routinely ask people if they are liberal, moderate, or conservative. This wasn't the case until about 1970--there are questions going back to the late 1930s, but they are not common. I was looking at a Gallup survey from 1950 which asked "do you consider yourself to be a conservative or a liberal in your political views?" 26% said conservative, 30% said liberal, 14% said neither, 23% no opinion, and 8% "no code or no data." I treated "no code or no data" as missing. There is a question of how to interpret "neither" and "no opinion". People in those categories might have been unable to choose because they saw themselves as in between, or because they weren't interested in politics or had no idea what the terms meant. I looked at the relationship between self-rated ideology and vote in 1948:
Dewey (R) Truman (D) Didn't vote
Conservative 54% 27% 11%
Liberal 24% 54% 11%
Neither 31% 43% 18%
No opinion 24% 38% 29%
So in terms of party support, both "neither" and "no opinion" were in between liberals and conservatives; both are also somewhat less likely to vote. In both of these respects, they are like people who choose "moderate" in contemporary surveys. So I treated them as a middle groups and "no code or no data" as missing.
The GSS has a question asking people to put themselves on a 7-point scale from very liberal to very conservative, with moderate in the middle. I collapse that two three categories (liberal side, moderate, conservative side) in order to compare contemporary (2010-18) patterns with those from 1950. Political ideology is more strongly connected to vote today--92% of liberals and only 26% of conservatives reported voting for Obama in 2012 (among people who voted). This is not surprising--there used to be liberal Republicans (Dewey was sometimes regarded as one) and conservative Democrats (especially in the South).
Next, I'll give the percent liberal minus the percent conservative for several demographic groups: gender, race, education, and size of place, in 1950 and recent years.
1950 2010-18
Men +6 -9
Women +2 -4
A small gender gap, which reversed direction.
Black +17 +8
White +4 -11
Blacks more liberal at both times.
no diploma +3 -6
HS 0 -10*
Some college +13
College grad +7 +1
More educated people a bit more liberal at both times.
less than 3,000 -10 -16
3,000-10,000 -6 -15
10,000-100,000 -2 -6
100,000-500,000 +9 +3
500,000+ +30 +7
A big difference at both times, maybe larger in 1950.
Age also made a difference at both times, with older people more conservative. The correlations were of similar size (-.11 recently and -.14 in 1950).
The two most noteworthy points are (1) educational differences in ideology haven't changed much, even though educational differences in party support have changed a lot (2) the urban/rural difference in ideology was at least as large in 1950 as in recent years, although the urban/rural difference in party support seems to have grown.
The 1950 survey also contained a number of questions about political issues; I'll discuss the connection of ideology to these in my next post.
*includes people with some college
[Data from the Roper Center for Public Opinion Research]
No comments:
Post a Comment