31% of Democrats are conservative! Party ID, Voting Intention, and the political makeup of samples

Ipsos has been getting a lot of questions lately about the political makeup of polls. This is normal towards the end of an election cycle – lots of people scrutinize the polls a lot more closely! We welcome the discussions, and it offers us an opportunity to help people better understand what makes up a quality political poll.

First let’s make some distinctions between question types. My colleague Chris has already explained the difference between all voting age Americans, Registered Voters, and Likely Voters. Equally important in political polling is being clear about the different types of questions we use to check the political balance of our samples. We’ve covered the basic ones below:

  • VOTING INTENTION: This is the ‘head-to-head’ or ‘horserace’ question that asks how people will vote. Ipsos asks: “If the 2012 Presidential Election were being held today and the candidates were [ROTATE] Barack Obama for president and Joe Biden for vice president, the Democrats, and Mitt Romney for president and Paul Ryan for vice president, the Republicans [END ROTATE], for whom would you vote?” Not all firms ask this in exactly the same way, but in general there is a broad consensus on approach to this question. It is this question which is usually filtered on Registered Voters, or as a further step on Likely Voters. This is also the question that poll aggregator sites like RealClearPolitics, Pollster.com, and FiveThirtyEight focus on for the national race when they track our data and other pollsters’ data.
  • PARTY IDENTIFICATION (ID): This question asks which party people most identify with, and how strongly they identify with it. We have explained this in great detail in another blog entry, but the crux of it is that Ipsos uses a ‘squeeze’ approach on people who identify as Independents or with no party, in order to better understand their likely voting behavior. This ‘squeeze’ approach gives Ipsos a smaller number of Independent voters than some other pollsters, however we publish the full breakdown of our question and so it is easy for interested parties to simply add the ‘leaners’ to the Independents in order to get a number comparable to other organizations’ metrics. It is critical to bear in mind that this question is self-reported and that there is no ‘correct’ target for this question – it changes constantly with the political ‘winds’, mainly as Independent and unaffiliated voters shift their allegiances. On balance, however, there are more people in the United States who identify as Democratic than Republican (this is balanced out, however, by the fact that Democrats are less likely than Republicans to vote – so when we apply filters by likelihood of voting, the party balance comes much closer).
  • Liberal-Conservative spectrum: Ipsos asks this question in this way: “Thinking about politics and government, do you consider yourself…?” (Very conservative; Moderately conservative; Lean conservative; Lean liberal; Moderately liberal; Very liberal; Don’t know/Refuse). This is a less frequently used question, but many pollsters (Ipsos and Gallup included) ask it as part of our demographic battery as a ‘check’ on the sample makeup for each of our polls. While this question may seem similar to the Party ID question, that is not the case at all: far more people identify as ‘Conservative’ than ‘Republican’, for example.

These three political metrics are all important to capture when carrying out political polling work, but they do not measure the same things. For example, while more Americans identify as Democrats than Republicans, more also identify as Conservative than Liberal. This is because there are lots of Democrats who identify as Moderate or Conservative (often called ‘Blue Dog Democrats’). The data from our September Telephone Poll breaks down as follows:

LIBERAL-CONSERVATIVE

ALL

Registered Voters

Likely Voters

Very Conservative

15%

16%

17%

Moderately Conservative

30%

31%

33%

Lean Conservative

9%

8%

8%

Lean Liberal

7%

7%

6%

Moderately Liberal

18%

17%

17%

Very Liberal

10%

10%

10%

(DK/NS)

11%

10%

8%

Total Conservative

54%

55%

58%

Total Liberal

35%

34%

33%

So Ipsos’ data also reflects the fact that there are more people who identify as Conservative than Liberal, by a fairly large margin. However, this cannot be directly mapped onto the Party ID question. In fact, when we look at this Liberal-Conservative question by Party ID, we find that 31% of Democrats identify as ‘conservative’, and 10% of Republicans identify as ‘liberal’.

In addition, we apply filters to the data as part of our effort to more accurately capture the opinions of those who will actually vote on Election Day, and this also impacts on the figures. You can see from the table below that Obama’s lead over Romney narrows as we move from all voters, to registered voters, to likely voters.

VOTING INTENTION

ALL

Registered Voters

Likely Voters

Barack Obama

50%

47%

48%

Mitt Romney

40%

43%

45%

Wouldn’t vote (VOL)

2%

2%

0%

None/Other (VOL)

4%

4%

3%

Don’t know/Refused (VOL)

4%

4%

4%

OBAMA LEAD OVER ROMNEY

+10

+4

+3

For the Voting Intention question, Ipsos’ goal is to replicate as closely as possible what might happen on Election Day, which means we think that the ‘Likely Voter’ column is the most accurate. This column also shows the election to be closest between the two candidates.

Party ID is the third political variable that is critical to any election analysis:

PARTY ID

ALL

Registered Voters

Likely Voters

Strongly Democrat

23%

24%

28%

Moderately Democrat

8%

9%

8%

Lean Democrat

15%

14%

12%

Strongly Republican

14%

17%

18%

Moderately Republican

10%

11%

11%

Lean Republican

14%

14%

15%

Independent (No lean)

15%

12%

8%

Total Democrats

47%

47%

48%

Total Republicans

38%

42%

44%

From the numbers above, many other pollsters would group the ‘Lean Democrat’ and ‘Lean Republican’ groups with the Independents, which in this case would move the Independents from 12% of Registered Voters to 40% of Registered Voters (which brings it in line with the broader polling market average for this metric.) While Ipsos measures this question in a different way from other pollsters, we value transparency and so make the full breakdown above available in every poll, so that people can calculate Independents however they prefer.

This table also demonstrates that there are more people in the United States who identify as Democratic rather than Republican; but, as mentioned above, Democrats are also less likely to vote, and so when we filter on Registered Voters, the gap between Democrats and Republicans narrows.

The final thing to consider in looking at political polls is whether you are looking at weighted or unweighted data. Most pollsters report sample sizes and subgroup sizes based on unweighted figures. This is considered best practice in the polling world because it reflects the actual number of interviews conducted (rather than a base size inflated or contracted by the effects of weighting). However, when assessing poll quality, the absolute number of interviews conducted with people who identify as Democrats and Republicans is less important than the way that the percentages in the tables I’ve included above break down. This is because the process of weighting the data balances out any imperfections in the sample makeup: if too many people from the East Coast, for example, were included in the sample, the weighting process will balance this out and ‘down-weight’ those interviews. It is the final, weighted percentages that matter, rather than the raw and unweighted data.

These dynamics are important to understand when trying to assess the validity and quality of a poll. It is easy to confuse the measures as well, given the different types of questions and the filtering that is required in order to get to the ‘Likely Voters’ who will actually turn up on election day. Pollsters aren’t fortune tellers; we can only work with what our data gives us. As my first boss Sir Robert Worcester often says, “Polls take a snapshot in time. They don’t predict events – but pollsters sometimes do!”

Ipsos is a nonpartisan polling firm; we strive to be as accurate as possible, and we are using all the tools at our disposal. We have a strong record for accuracy in this country and elsewhere. First and foremost, our goal is to accurately measure and predict the 2012 election outcome. Fingers crossed!

Comments are closed.