Today, we took great interest in the Tweet below by Carroll Doherty at the Pew Research Center, highlighting increasing conservative identification among Republican voters over the timespan between John McCain's first presidential campaign in 2000 and today. Pew's data show conservative idenfitication in the GOP increasing by 12 points, from 56 percent to 68 percent. 

In 2000, when John McCain first ran for president, 56% of Reps identified as conservatives, 32% as moderates. Last year, it was 68% conservative/26% moderate. Over the same period, the share of Dems identifying as liberal as increased 18 pts.https://t.co/vtFDM7euMu pic.twitter.com/CcNG0kWnPU

— Carroll Doherty (@CarrollDoherty) August 28, 2018

The Pew data got us wondering about whether these trends manifest themselves in Texas, so we pulled together polling data from over 30 University of Texas / Texas Tribune polls to see if and how ideological identification in Texas has changed since 2008 (the inaugural year of our data). The data series is represented in the graphics below.

Looking at the changes and consistencies over time led us to three hot takes, with more reflection and analysis pending.

1) While Republican identifiers have grown more conservative as a group nationally, Texas Republicans' ideological self-identification has held remarkably consistent over the last 10 years. There are explicable ebbs and flows to the rate of conservative identification among Texas Republicans. Conservative identification surged around the 2010 Tea Party electoral wave, then dipped heading into the 2012 Election, when Mitt Romney stood on the precipice of losing to Barack Obama, and dipped again in the immediate aftermath of the Trump Election. Nonetheless, moderate identification has stayed in a narrow band, and in what should be a surprise to no one, conservative identification among Republicans has hovered consistently above 80 percent, and never gone below 70 percent. In the wake of his death, it's worth noting that John McCain experienced this in Texas in 2008. Though McCain was already the presumptive nominee by the time of the 2008 Texas presidential primary, he won barely half of the votes in the that election. The runner-up was conservative, Christian stalwart Mike Huckabee, who received over 500,000 votes (McCain got just over 697,000). As we've written elsewhere, if national conservatives have embraced Trump's most reactionary rhetorical appeals on immigration and race relations, the market for these appeals was well developed in the Texas GOP long before Trump used them to become the national party's figurehead. 

2) Liberal self-identication among Democrats as a group has increased sharply since Donald Trump became president. It's long been a baseline assumption that, whether you're looking at public opinion or the profiles of candidates and elected officials, Texas Democrats tend to look more conservative than their northern counterparts. For most of the time series this is true, with the plurality choice among Texas Democrats oscilating between "liberal" and "moderate" for most of the period the data covers. But since October 2017, liberal identification has taken a commanding lead among the identification choice of Texas Democrats, reaching as high as 65 percent – 19 percentage points higher than the mean percentage (46 percent) between June 2008 and February 2017, right after Trump's inauguration. Whether this increase is durable and whether it reflects more liberal policy preferences rather being another manifestation of intense opposition to the president can't be readily determined at the moment. But it does illuminate why Democratic U.S. Senate candidate Beto O'Rourke's seeming embrace of a campaign agenda that makes him look far more like a national Democratic candidate than previous high profile contenders has been so warmly greeted by attentive Democrats. While critics (and, of course, the Cruz campaign and allied forces) have pondered whether O'Rourke has moved too far to the left for Texas, it may well be that O'Rourke is meeting his own market in an effort to maximize Democratic mobilization. The election will in part provide a crucial measure of which of these two positions was accurate. The obvious risk for the O'Rourke campaign is that both are correct, but the GOP advantage is too large to overcome this cycle, even if the candidate is ideologically in tune with a base that can't be turned out in sufficient numbers to overcome the existing GOP advantage in Texas.

3) There have been hints of Democratic change in other survey items. From time to time, the UT/TT poll asks partisans about the ideological orientation of their party in state government. Republicans are asked whether their elected officials are too conservative, conservative enough, or not conservative enough, and Democrats are asked if theirs are too liberal, liberal enough, or not liberal enough. The origins of this item, ironically, come from the observation that ideological infighting among Republican elites may not emanate from the Republican electorate, and polling data illustrates this point. Overall, only 10 percent of Republicans say that Republican elected officials in Texas are too conservative, while 30 percent say that they are not conservative enough. The remaining 50 percent are happy with the Republican Party's ideological orientation. But it's also revealed that among Texas Democrats, a slim pluarlity (37 percent), say that their elected officials aren't liberal enough, with slightly less saying they're liberal enough, and only 10 percent saying they are too liberal.