A few comments on our post of the new SurveyUSA Texas poll raised two questions worthy of further discussion.
First, reader s.b. notes:
[W]ith an automated survey, if its in English, they aren't sampling spanish only or mostly spanish speakers. I think it skews these results.
Some pollsters (such as Gallup) offer voters the opportunity to complete the survey in Spanish when they encounter Spanish speaking respondents. Most pollsters, however, will simply end the interview in these instances. I asked SurveyUSA's Jay Leve about their procedure in Texas and he notes that while they do have the facility to offer respondents the option to complete a survey in either English or Spanish (and have done so in mayoral elections in New York and Los Angeles and some congressional districts), they did not offer a Spanish interview for their Texas poll.
However, before leaping to conclusions about the SurveyUSA results, keep in mind that none only one of the other Texas pollsters report using bilingual interviewing for any of their surveys [Correction: interviews for the Washington Post/ABC News poll "were conducted in English and Spanish"]. Three of the other pollsters -- Rasmussen Reports, PPP and IVR polls -- also interview with an automated methodology rather than live interviewers.
And before leaping to conclusions about all the Texas polls, we might want to know just how many Latino voters in Texas speak only Spanish. I have not done survey work in Texas, but my memory from conversations with pollsters that do is that the percentage that will actually complete an interview in Spanish when offered is typically in the low single digits.
Second, several commenters have speculated about the small changes in the demographic composition of the last two SurveyUSA Texas polls. For example, "Mike in CA" points out:
Hispanic turnout at 28% sounds just about right. The last SUSA survey had it at 32% which was way too high. It seems SUSA has scaled back their Hispanic estimates, so they must have a reason. Additionally, the boosted AA to 23%, from 18%. Seems reasonable considering the extraordinary increases in early voting turnout from Houston and Dallas [emphasis added].
That's not quite right. Keep in mind that SurveyUSA's approach to likely voter modeling is comparable to that used by Iowa's Ann Selzer, in that they do not make arbitrary assumptions about the demographic composition of the likely electorate. As SurveyUSA's Jay Leve explains, they "weight the overall universe of Texas adults to U.S. census" demographic estimates, then they select "likely voters" based on screen questions and allow their demographics to "fall where they may." So some of the demographic variation from survey to survey is random, but large and statistically statistically significant variation should reflect real changes in the relative enthusiasm of voters. Leve goes into more detail in the email that I have reproduced after the jump, which also includes the full text of the questions they use to select likely voters.
Comments emailed by SurveyUSA's Jay Leve:
SurveyUSA makes no “forced” assumptions about the size or composition of any likely voter demographic group. We weight the overall universe of Texas adults to U.S. census, and let the sub-groups fall where they may.
What that means is: if, using the identical methodology from one tracking poll to another, the composition of Hispanics drops from 32% to 28%, it did so occurring naturally. In the same way, if the composition of seniors drops, it did so occurring naturally. If the number of likely voters increases 31% of adults on 2/19/08 to 35% of adults on 2/25/08, it did so occurring naturally. (same exact instrument used for both surveys).
Any one of these changes may indeed have happened by chance alone – because of random sampling – but when a couple of groups that could be aligned with support for one candidate appear to shrink simultaneously, it is possible to infer some diminished enthusiasm for that candidate. (Composition of Hispanics declined from 32% to 28%, and composition of voters age 50+ declined from 46% to 42%, from 2/18/08 SurveyUSA TX poll to 2/25/08 SurveyUSA TX poll).
In the same way, though on a separate point: SurveyUSA found in 2/25/08 survey that of 2,000 adults in Texas, 704 were likely to vote in the Democratic Primary, 484 were likely to vote in the Republican Primary. If you do ballpark math, you might (incorrectly) infer that Democrats are 45% more plentiful in TX than Republicans. They are not. But Democrats are more likely to tell SurveyUSA they have already voted, or are likely to vote, in the Texas Primary. That enthusiasm, or lack of enthusiasm, among certain groups, filters through to the final primary poll composition numbers.
The same exact survey instrument that produced 704 likely Democratic Primary voters and 484 likely Republican Primary voters, was also used to produce the following general election results (released today 2/26/08), which are based on registered TX voters, not likely primary voters. The number of registered Republicans and number of Registered democrats is approximately equal. (39% Republican, 38% Democrat).
In summary: in almost every case, by design, the water in a SurveyUSA poll seeks its own level, and as enthusiasm for a particular cause or candidate manifests itself, it flows through to our results (comment continues below image).
The exact questions used to determine likely primary voters in TX were:
Copyright 2008 SurveyUSA.
1) Are you registered to vote in the state of Texas?
Yes, press 1.
No, 2. [branch to demographics]
Not sure? 3. [branch to demographics]
2) On March 4 … [On Tuesday …] [Tomorrow …] Texas will hold a presidential primary election. Early voting is possible. Have you already voted?
Yes, press 1.
No, 2. [branch to Q4]
3) Did you vote in the Republican primary? Or the Democratic primary?
If you ...
Voted in the Republican primary, press 1. [branch to GOP voting question]
Voted in the Democratic primary, 2. [branch to Democratic voting question]
Can't remember? 3. [branch to demos]
4) On a scale of 1 to 10, how likely are you to vote…where 10 means you are certain to vote … and 1 means you are certain NOT to vote?
[highest-interest voters continue; others are branched to demographics]
5) If you are …
Certain to vote in the...Republican Primary, press 1. [branch to GOP voting question]
Certain to vote in the Democratic Primary, 2. [branch to Dem voting question]
Not yet certain which Primary you will vote in, 3. [branch to demos]
