Yesterday, I posted about a couple of polls taken on the coming race for governor in Texas. A September Texas Lyceum Poll had shown Republican Greg Abbott with an 8 point lead over Democrat Wendy Davis. The University of Texas / Texas Tribune Poll, taken at the end of October, showed Abbott with only a 5 point lead. While both polls showed the likely Republican candidate with a lead, it is not an unsurmountable lead.
Now a third poll on that statewide race has been released, and it is not as kind to Democratic hopes as the two previous polls. It is the Public Policy Polling survey -- taken between November 1st and 4th of 500 Texas voters, with a 4.4 point margin of error. The results of that poll are illustrated in the chart above -- and it has Abbott with a 15 point lead over Davis (50% to 35%).
Why is this poll so different from the others. Why does this poll show a 15 point difference, when the UT/TT Poll (taken only days earlier) shows a 5 point difference? The difference could easily lie in the size of the pool of respondents in each poll. While the UT/TT Poll surveyed more than 1,200 Texas voters, The PPP survey polled only 500 Texas voters. That means this new poll has a significantly larger margin of error.
And it would have an even larger margin of error when it is broken down into demographic groups. For instance, the PPP survey showed Hispanic voters favoring Abbott by 5 points. Frankly that is hard to believe, and I doubt any statewide Republican can win a majority of the Hispanic vote. This survey had only 19% of the total 500 voters as Hispanics, meaning that less than 100 Hispanics were surveyed (a tiny number considering the large number of Hispanics in the state, and a number that would have a very large margin of error).
This does not necessarily mean the PPP survey is wrong. It might not be, but it is obviously not as accurate as it could have been if a larger sample had been used. Democrats have never believed this race would be an easy one to win. They know they will have to give until it hurts and work their butts off if they are going to win it. However, I still believe it can be won. And I look forward to seeing the results of future polls.