The Texas Voter ID Law and the 2014 Election: A Study of Texas’s 23rd Congressional District Principal Investigators Governor Bill Hobby Mark P. Jones, Rice University’s Baker Institute for Public Policy Jim Granato, University of Houston Hobby Center for Public Policy Renée Cross, University of Houston Hobby Center for Public Policy Research Team Members Cong Huang, David Llanos, Chris Mainka, Kwok-Wai Wan, Ching-Hsing Wang, Isaiah Warner August 2015 Executive Summary This study examines the impact of Texas’s current voter photo identification regulation in the November 2014 election using a case study of voter behavior, preferences and attitudes in U.S. Congressional District 23 (CD-23). CD-23 is widely considered to be the only one of the state’s 36 U.S. House districts where both the Democratic Party and Republican Party candidates have a realistic chance of victory in November. The 2014 election saw CD-23’s then freshman incumbent representative (Pete Gallego, D-Alpine) narrowly defeated by the district’s current representative (Will Hurd, R-San Antonio). A survey of 400 CD-23 registered voters who did not vote in the November 2014 election indicates that for 5.8% of these Texans the principal reason given for why they did not vote was because they did not possess any of the seven forms of photo identification required by the state to cast a vote in person. More than twice that many (12.8%) agreed that their lack of any one of these seven photo IDs was a reason they did not vote. However, when further queried about the different forms of photo identification in their possession, the survey revealed that a much lower proportion (2.7%) of CD-23 nonvoters in fact lacked one of the seven needed to vote in person. In all, while 12.8% and 5.8% of these non-voters cited a lack of a photo ID as a reason or the principal reason they did not vote, only 1.0% and 0.5% of the respondents both respectively attributed their non-voting in part or primarily to a lack of photo ID and actually did not possess an approved form of photo ID. This study suggests that the most significant impact of the Texas voter photo ID law on voter participation in CD-23 in November 2014 was to discourage turnout among registered voters who did indeed possess an approved form of photo ID, but through some combination of misunderstanding, doubt or lack of knowledge, believed that they did not possess the necessary photo identification. The disjuncture between the proportion of voters who listed a lack of an ID as a reason or the principal reason they did not vote and the proportion of these individuals who actually did not have an ID highlights the potential for a future voter education campaign to clearly explain the types of photo identification required to cast a vote in person in Texas. This education campaign however needs to be designed based on a comprehensive study targeted at understanding the sources and causes of confusion which resulted in so many Texans believing they did not have a required form of photo ID when in reality they actually did. This study also examines the potential impact of the Texas voter photo ID law on the outcome of the 2014 election in CD-23. It suggests that the presence of the law kept far more Gallego than Hurd supporters away from the polls last fall. 1 The 2011 Texas Voter Identification Law In 2011, the Texas Legislature passed Senate Bill 14 (SB 14) that created a new requirement for voters to show photo identification when voting in person.1 Initially, the U.S. Department of Justice (DOJ) argued that Texas’s voter photo ID law disproportionately placed an undue burden on minority voters and thus rejected the Texas law, a decision upheld by the U.S. District Court for the District of Columbia on August 30, 2012. However, on June 25, 2013, the U.S. Supreme Court decision in Shelby County v. Holder removed for the present time the requirement that Texas seek DOJ approval for election law related changes, and subsequently the 2011 voter ID law immediately took effect. While a U.S. District Judge struck down Texas’s voter photo ID law on October 9, 2014, a panel for the U.S. 5th Circuit Court of Appeals issued a preliminary injunction against the ruling of the U.S. District Court, which then was confirmed 6-3 by the U.S. Supreme Court on October 18, 2014. As a result, Texas’s photo ID law was in force for the November 2014 election, requiring a state-approved form of photo identification in order to vote in person. The approved forms of photo identification are as follows:        Texas driver license issued by the Texas Department of Public Safety (DPS); Texas Election Identification Certificate (EIC) issued by DPS; Texas personal identification (ID) card issued by DPS; Texas concealed handgun license (CHL) issued by DPS; United States military identification card containing the person’s photograph; United States citizenship certificate containing the person’s photograph; United States Passport. Photo identification is not required to cast an absentee/mail ballot in Texas, but noexcuse absentee voting is limited to those 65 and older. All other voters must either be disabled, in jail (but otherwise eligible to vote), or out of town on election day and during the entire early voting period (October 20 to October 31 in 2014). Prior studies have examined the impact of voter ID requirements on voter registration and turnout for the population in general and for ethnic and racial minorities in particular (Ansolabehere 2009; Hood and Bullock 2008; Mycoff 2009), the relationship between socioeconomic status and access to valid photo identification (Atkeson et al. 2010; Barreto et al. 2009) and whether voter ID requirements advantage one party over the other (Barreto et al. 2009; Gomez 2008). Texas provides an additional opportunity to investigate the effects of voter photo ID requirements on individual voting behavior. The 271,005 CD-23 registered voters who did not participate in the November 2014 election were the target population of this study. Pete Gallego lost to Will Hurd by 2,422 votes (49.8% to 47.7%) in a district where 118,004 registered voters cast a ballot (115,429 in the CD-23 race) and 271,005 did not. Therefore, if the voter ID law affected the outcome of any federal race in 2014 in Texas, it would have been in CD-23. 1 Texas enacted a voter identification requirement in 1971. However, it did not require a photo. 2 Survey Research Design 1. Survey: A telephone survey was conducted among 400 registered voters in CD-23 who did not cast a ballot in the November 2014 election. These non-voters (identified through the official voting records) were randomly selected from a list of registered voters who did not participate in last fall’s contest. The survey was conducted in both English and Spanish by the Hobby Center for Public Policy’s Survey Research Institute between March 2 and March 28, 2015, with 52.2% of respondents interviewed in English and 47.8% in Spanish. The survey’s margin of error was +/- 5%. In this analysis, the data are weighted using the raking method to insure an accurate representation of the respondent universe (registered voters in CD-23 who did not vote in the November 2014 election) in regard to gender, age and ethnicity/race. The raking method allows us to improve the relation between the sample and the population by adjusting the sampling weights of our cases so that the marginal totals of the adjusted weights on specified characteristics are consistent with the corresponding totals for the actual population (Battaglia et al. 2004). 2. Question Wording: To design the questionnaire, we primarily utilized survey questions about electoral behavior and attitudes employed in prior academic studies. Specifically, the survey instrument was divided into five major sections:  Survey Questions I: Respondents were asked a series of questions about the reasons why they did not vote.  Survey Questions II: Respondents were asked if they had voted in the November 2014 election, what their vote choice would have been in the races for governor, lieutenant governor and U.S. Congress (CD-23).  Survey Questions III: Respondents were asked which of the seven forms of acceptable photo ID they possessed and if they did not, they were asked if they had these IDs in the past.  Survey Questions IV: Respondents were asked their opinion on voter identification legislation, both in general and in regard to specific issues.  Survey Questions V: A series of socio-demographic and partisan identification questions were asked. 3 Congressional District 23 CD-23 covers a larger geographic terrain (see Figure 1) than any other congressional districts in the country other than those whose boundaries are those of the entire state. It is comprised of all or part of 29 counties stretching from large sections of Bexar County (San Antonio) in the east, down to Maverick County (Eagle Pass) on the border and out to the western edge of the state of Texas in the southern portion of El Paso County. Bexar County contains 42% of the district’s registered voters, followed in number by El Paso County (11%), Maverick County (7%), Medina County (7%), Val Verde County (7%) and Uvalde County (4%), with the remaining 22% of the registered voters spread across the other 23 counties. Ten of these counties do not possess a DPS office (where Texans they must go to obtain a driver license, Texas ID or EIC), while in many other counties, the DPS office is more than an hour away by car for significant portions of the population and open only one to three days a week. Figure 1. A Map of Texas’s Congressional District 23 Source: http://www.fyi.legis.state.tx.us/fyiwebdocs/PDF/congress/dist23/m1.pdf 4 Non-Voting in Congressional District 23 in 2014 CD-23 residents who were registered to vote in the November 4, 2014 election but did not exercise their right of suffrage were asked why they did not vote. First, they were asked about their level of agreement (strongly agree, agree, neither agree or disagree, disagree, strongly disagree) with six common reasons why people do not vote along with a seventh reason related to the lack of the required photo identification: 1. 2. 3. 4. 5. 6. 7. “You or a family member was ill.” “You had transportation problems.” “You were too busy, with conflicting work, family or school schedules.” “You didn’t like the candidates or the issues.” “You weren’t interested and felt your vote wouldn’t make a difference.” “You were out of town.” “You did not possess any of the state approved forms of photo identification needed to cast a vote in person.” After this battery of questions was asked, respondents were queried regarding which single one of these reasons best explained why they did not vote in the November 2014 election. Table 1 provides the proportion of the non-voters who either strongly agreed or agreed with the seven statements above. Since the respondents could strongly agree or agree with multiple reasons for why they did not vote, the column percentages sum to more than 100%. The most common reason for not voting was that the respondent was too busy with their conflicting work, family or school schedules; almost two-fifths (38.5%) of respondents strongly agreed or agreed with this statement. The second most common reason for non-voting was that the respondent was not interested in the electoral process or felt that their vote would not make a difference, with over one-third (34.1%) strongly agreeing or agreeing with this reason. The third most popular reason was that the respondent did not like the candidates or issues, with more than a quarter (26.9%) agreeing that this was a reason. Among these seven potential reasons for why someone did not vote in the November 2014 election, the one with which respondents had the least agreement (12.8%) was that they did not possess any of the state approved forms of photo identification needed to cast a vote in person. 5 Table 1. Reasons for Not Voting in CD-23 in the November 2014 Election Percent Strongly Agreeing or Agreeing This Was A Reason Reason for Not Voting Too Busy With Conflicting Work, Family or School Schedules 38.5% Not Interested or Felt Vote Wouldn’t Make Difference 34.1% Didn’t Like the Candidates or the Issues 26.9% They or a Family Member Was Ill 23.2% Out of Town 20.3% Transportation Problems 16.2% Did Not Possess Any of the State Approved Forms of Voter ID Needed to Cast a Vote in Person 12.8% When asked about the principal reason why they did not turn out to vote in the November 2014 election, the most common response (see Table 2) by CD-23 nonvoters was that they were too busy with conflicting work, family or school schedules (25.9%), followed by the response that they were not interested or felt their vote wouldn’t make a difference (18.5%), and that they or a family member was ill (17.0%). The two least frequent responses were that they had transportation problems (7.2%) and that they did not possess one of the required forms of state approved photo identification (5.8%). Table 2. Principal Reason for Not Voting in CD-23 in the November 2014 Election Reason for Not Voting Percent Listing This As The Principal Reason Too Busy With Conflicting Work, Family or School Schedules 25.9% Not Interested or Felt Vote Wouldn’t Make Difference 18.5% They or a Family Member Was Ill 17.0% Out of Town 14.6% Didn’t Like the Candidates or the Issues 11.0% Transportation Problems 7.2% Did Not Possess Any of the State Approved Forms of Voter ID Needed to Cast a Vote in Person 5.8% 6 In sum, 12.8% of the respondents either strongly agreed or agreed that their lack of a state-approved form of photo identification was a reason why they did not cast a vote, while 5.8% listed lack of photo identification as the principal reason they did not vote. Respondents were also asked which of the seven forms of state-approved photo identification they possessed. The analysis below excludes any information pertaining to registered voters who possess an Election Identification Certificate (EIC). As of November 2014, only 340 EICs had been issued in the entire state of Texas (Wilder 2014), meaning that most likely fewer than 10 of the almost 390,000 CD-23 registered voters had one in their possession. The results of the survey indicate that most of the CD-23 non-voters (89.9%) possessed a valid Texas driver license, with an unexpired U.S. Passport (59.5%) as the next most common form of approved photo identification. Other than an EIC, the least frequently held form of photo identification was a concealed handgun license (CHL), which was possessed by 5.3% of the CD-23 non-voters. Nearly all — 97.3% — of the respondents possessed at least one of the state-approved forms of photo identification needed to cast a vote in person. In other words, the percentage of non-voters in CD-23 who did not have one of the state approved forms of photo identification was 2.7%. At the same time, 6.4% of the respondents indicated that there was a member of their household who did not possess one of the seven forms of photo identification needed to vote in person in Texas. While 2.7% of the respondents did not possess any of the seven valid forms of photo identification, only 1.0% did not possess a photo ID and agreed that a lack of this photo ID was a reason why they did not vote. Only 0.5% both did not have a valid ID and listed this deficit as the principal reason why they did not vote in the November 2014 election. Misinformed Registered Voters A quick summary of the above analysis underscores that 12.8% of the respondents agreed that their lack of a required photo ID was a reason they did not vote, with 5.8% claiming that it was the principal reason that they did not vote. However, when asked which of the seven specific acceptable forms of photo identification that they possessed, only 2.7% of the respondents reported not possessing a single form. Less than half (1.0%) and a fifth (0.5%) of the 2.7% respondents who listed a lack of ID as a reason or the principal reason they did not vote really did not possess a valid photo ID. That is, most of the non-voters who stated they did not vote due principally or at least in part to the fact that they did not have one of the required forms of photo identification actually did possess at least one of the seven state-approved forms of photo ID. Our study suggests that while one effect of the 2011 Texas voter photo ID law was to prevent registered voters who wanted to cast a ballot from exercising their right of 7 suffrage, the proportion of non-voters who experienced this difficulty was relatively small (i.e., between 0.5% and 1.0% of non-voters). The most prominent impact of the legislation was that due entirely to a misunderstanding or a general lack of information of the photo identification requirements under the law, somewhere between one out of every 10 and one out of every 20 non-voters in CD-23 did not participate in the general election process in 2014. The aforementioned findings suggest that the very modest public outreach campaign carried out in 2014 by the Texas Secretary of State and some county election officials to educate registered voters about the state’s new photo identification law was suboptimal, at least in CD-23. They also suggest however that a well-designed and wellfunded public voter education campaign which clearly explains the forms of photo identification that can be used by Texas registered voters to cast a ballot in person could go a long way towards ameliorating most of the adverse impact of the state’s voter photo ID law on voter turnout. To be both credible and effective however such a voter education campaign would first require a comprehensive study to determine the principal sources and causes of the mistaken belief among many Texans that they did not possess a valid form of photo ID. Ethnicity/Race and Non-Voting in CD-23 Table 3 provides the ethnic/racial distribution for five populations in CD-23: the voting age population (based on U.S. Census data), non-voters, non-voters who strongly agreed or agreed that a lack of a photo identification was one reason they did not vote in the November 2014 election, non-voters who stated that a lack of a photo identification was the principal reason they did not vote in the November 2014 election, and non-voters who did not possess one of the seven forms of photo identification needed to vote in person in Texas. Table 3. Ethnic/Racial Distribution for Five Populations in CD-23 Ethnicity/Race Voting Age Population Non Voters No ID One Reason Didn’t Vote No ID Principal Reason Didn’t Vote Actually Did Not Possess a Valid ID Latino/Hispanic 65.8% 73.4% 80.8% 77.2% 76.9% Anglo/White 28.9% 21.5% 12.9% 13.0% 10.6% African American 3.6% 2.3% 2.3% 4.0% 0.0% Other 1.7% 2.8% 4.0% 5.8% 12.6% 8 CD-23 is a Latino majority district, with Hispanics accounting for 65.8% of the district’s voting age population (based on self-identification in the 2010 U.S. Census) and 58.5% of registered voters (i.e., 58.5% of the registered voters have Hispanic surnames). Anglos (i.e., White non-Hispanics) represent 28.9% of the voting age population in CD23 and African Americans represent 3.6%. The non-voting registered voter population in CD-23 is almost three-quarters Hispanic (73.4%), a fifth Anglo (21.5%) and 2.3% African American. Overall, with all other things held constant, Latinos were more likely than Anglos to be non-voters. Latinos were also more likely to agree that their lack of photo identification was a reason they did not vote and that they did not (in reality) possess one of the required forms of photo identification. However, these ethnic/racial differences are not statistically significant with one exception; Latino non-voters were significantly more likely than Anglo non-voters to strongly agree or agree that a lack of photo ID was a reason that they did not cast a ballot in the November 4 contest. Voter Photo ID and Elections Opponents of the Texas voter photo ID law criticize the legislation for two principal reasons. The first is that they believe it unconstitutionally restricts the suffrage rights of many Texans. The second is that they believe it was designed to benefit Republican candidates at the polls by reducing the number of Democratic-leaning voters. In the preceding pages we detailed the extent to which non-voting could be explained as a consequence (whole or partial) of the Texas voter ID law. Next, we look at the impact of non-voting on partisan political outcomes. The survey asked the 400 non-voters who they would have voted for (had they voted in November 2014) in the two most high profile statewide races (governor and lieutenant governor) as well as in the CD-23 election. According to the Texas Secretary of State’s November 2014 gubernatorial election results within CD-23, the Republican Party candidate Greg Abbott won 57.1% of the actual vote cast, the Democratic Party candidate Wendy Davis 40.8%, the Libertarian Party candidate Kathie Glass 1.5% and the Green Party candidate Brandon Parmer 0.6%. 2 Among non-voters however, Davis was the preferred candidate (35.8%), followed by Abbott (30.3%), Glass (1.7%) and Parmer (1.1%). Nearly one-third of respondents (31.1%) would not have cast a vote for any of the candidates, reflecting in part the lack of interest in and awareness of electoral politics among many non-voters in CD-23. 2 Statewide, Abbott won 59.3%, Davis 38.9%, Glass 1.4% and Parmer 0.4%. 9 Table 4. The 2014 Gubernatorial Race in CD-23: Five Populations3 Non Voters No ID One Reason Didn’t Vote No ID Principal Reason Didn’t Vote Actually Did Not Possess a Valid ID 57.1% 30.3% 28.0% 34.3% 4.8% Davis (D) 40.8% 35.8% 27.3% 30.0% 0.0% Glass (L) 1.5% 1.7% 4.0% 5.8% 21.1% Parmer (G) 0.6% 1.1% 5.8% 13.0% 0.0% NA 31.1% 34.8% 16.9% 74.1% All Voters (actual vote results) Abbott (R) Candidate (Party) None Among those non-voters who agreed that their lack of a photo ID was a reason they did not vote and those who listed it as the principal reason they did not vote, Abbott was favored over Davis (28.0% vs. 27.3%; 34.3% vs. 30.0%). Finally, by far the most common response (74.1%) among the 2.7% of voters who actually did not have one of the seven required forms of photo ID is that they would not have cast a vote for any of the four gubernatorial candidates. Table 5 provides the same five categories used for the gubernatorial race for the election to determine the state’s next lieutenant governor. In CD-23, the Republican Party candidate Dan Patrick won 54.3% of the actual vote cast, the Democratic Party candidate Leticia Van de Putte 43.2%, the Libertarian Party candidate Robert Butler 2.1%, and the Green Party candidate Chandrakantha Courtney 0.5%.4 Among all nonvoters surveyed, Van De Putte was narrowly favored over Patrick, 30.5% to 29.7%, with more than a third (35.9%) of the respondents not indicating a preference for any of the four candidates. 3 The data for All Voters (Actual Vote Results) were obtained from the Texas Secretary of State’s November 2014 election results (2015). 4 Statewide, Patrick won 58.1%, Van de Putte 38.7%, Butler 2.6% and Courtney 0.6%. 10 Table 5. The 2014 Lieutenant Governor Race in CD-23: Five Populations5 Non Voters No ID One Reason Didn’t Vote No ID Principal Reason Didn’t Vote Actually Did Not Possess a Valid ID 54.2% 29.7% 20.8% 19.8% 0.0% LVDP (D) 43.2% 30.5% 29.2% 36.5% 0.0% Butler (L) 2.1% 1.4% 1.4% 0.0% 0.0% Courtney (G) 0.5% 2.5% 5.5% 12.3% 12.6% NA 35.9% 43.1% 31.4% 87.4% Candidate (Party) All Voters (actual vote results) Patrick (R) None Among those non-voters who agreed that their lack of a photo ID was a reason they did not vote and those who listed it as the principal reason they did not vote, Van de Putte was favored over Patrick (29.2% vs. 20.8%; 36.5% vs. 19.8%). By far the most common response (87.4%) among the 2.7% of voters who actually did not have one of the seven required forms of identification is that they would not have cast a vote for any of the four lieutenant governor candidates. In the 2014 CD-23 election, the Republican Party candidate Will Hurd (49.8%) defeated the Democratic Party candidate Pete Gallego (47.7%), with the Libertarian Party candidate Ruben Corvalan garnering 2.5% of the vote (see Table 6). Among non-voters Gallego (40.5%) was preferred by more than twice as many respondents as Hurd (17.1%), with two-fifths of the non-voters (40.7%) not having a preference for any of the three candidates in the race. 5 The data for All Voters (Actual Vote Results) were obtained from the Texas Secretary of State’s November 2014 election results (2015). 11 Table 6. The 2014 CD-23 Race: Five Populations6 Non Voters No ID One Reason Didn’t Vote No ID Principal Reason Didn’t Vote Actually Did Not Possess a Valid ID 49.8% 17.1% 9.9% 9.1% 4.8% Gallego (D) 47.7% 40.5% 39.7% 54.2% 12.6% Corvalan (L) 2.5% 1.6% 1.4% 0.0% 0.0% NA 40.7% 49.0% 36.7% 82.6% All Voters (actual vote results) Hurd (R) Candidate (Party) None Among those respondents who agreed that their lack of a photo ID was a reason they did not vote and those who listed it as the principal reason they did not vote, between four and five times more of these non-voters would have cast a ballot for Gallego than for Hurd (39.7% to 9.9%; 54.2% to 9.1%). As was the case for the gubernatorial and lieutenant governor elections, the most common response (82.6%) among the 2.7% of voters who actually did not have one of the seven required forms of identification is that they would not have cast a vote for any of the three congressional candidates. In the actual CD-23 election, Hurd bested Gallego by a little more than 2,400 votes: 57,459 to 55,037. Corvalan won 2,933 total votes. At the same time, 271,005 registered voters did not turn out to vote in CD-23. According to our survey, 12.8% of these nonvoters stated that their belief that they did not possess one of the seven required forms of photo ID to vote in person was a reason they did not vote, and 5.8% cited it was the principal reason they did not vote. The survey responses of non-voters suggest that in the CD-23 race, Gallego was far more adversely affected by the impact of the voter ID law on voter participation than was Hurd. Four times as many non-voters (39.7% to 9.9%) who agreed that the photo ID law was one of the reasons they did not participate would have voted for Gallego instead of Hurd had they voted. And five times as many non-voters who listed the photo ID law as the principal reason they did not participate would have voted for Gallego rather than for Hurd. In sum, while the results of this survey do not allow us to conclude that Gallego would have been re-elected in the absence of the voter photo ID law, they do indicate that the law did have a disproportionate impact on his supporters, and therefore may have possibly cost him the election. The negative impact of the voter ID law on Gallego’s electoral performance was however overwhelmingly the product of a lack of knowledge and/or confusion regarding the requirements of the law among the potential Gallego voters who did not turn out 6 The data for All Voters (Actual Vote Results) were obtained from the Texas Secretary of State’s November 2014 election results (2015). 12 because they mistakenly believed they did not possess one of the required forms of photo ID, when in fact they did have at least one. Among the small number of registered voters who really did not have an acceptable photo ID, Gallego’s support was minimal, since almost all of these citizens did not have a preference for any of the three candidates in the race nor any salient identification with either the Democratic Party or the Republican Party. Conclusion This case study of 400 CD-23 registered voters who did not vote in the November 2014 election is a snapshot of their perceptions, attitudes and behaviors. While 5.8% of these non-voters stated that the primary reason they did not vote was because they did not possess any of the state required forms of photo identification and 12.8% agreed that their lack of one of the required forms of a photo ID was a reason they did not vote, only 2.7% of them in reality lacked one of the seven forms of photo ID needed to vote in person. These findings suggest that the most significant impact of the current Texas voter ID law is confusion and subsequently depressed voter turnout. This is potentially a critical consequence in highly competitive elections such as the 2014 congressional race in CD-23. This study also shows that the Latino non-voters were significantly more likely than Anglo non-voters to strongly agree or agree that a lack of photo ID was a reason that they did not cast a ballot in the 2014 general election, which is especially relevant in a district in which roughly three-fifths of the registered voters are Latino. Just what can be done to understand and possibly improve voter participation with the voter photo ID law in effect? The following outlines three possibilities: 1. A comprehensive non-partisan study of the education efforts surrounding the Texas voter ID law during the latter half of 2013 and during 2014 could be conducted. First, determine what was done (and when) in CD-23 by the Secretary of State’s office as well as by individual counties and stakeholders such as the Democratic and Republican Parties and non-governmental organizations such as the League of Women Voters. The study could also include a comparative analysis of what other states (with similar demographics and new voter ID laws) have done. Second, a survey could determine the sources and causes of the mistaken belief among many registered voters that they did not possess a valid form of photo ID. This knowledge is essential to designing an effective education strategy. 2. Extending this case study as well as the study of the voter photo ID education endeavors (both past and current) beyond CD-23 will enrich the knowledge of what is occurring throughout Texas. Will we see similar patterns in other congressional districts, especially those with a large number of Latino registered voters? For example, consider CD-23 and CD-29, a Latino majority district in Houston. Are education efforts more effective in urban areas, where density and resources such as volunteers and easy internet accessibility can be more 13 conducive to outreach? Are education outreach efforts tailored for different constituencies; and if not, can they be better designed to reach more people? 3. A survey such as the one utilized for this study records people’s opinions on a given day. An annual tracking survey records people’s opinions on one day and a subsequent survey will record other people’s opinions on another day. In contrast, a longitudinal or panel study tracks the same people’s attitudes and behavior over time. A longitudinal study of voters in Texas could test cause and effect in voter participation by determining a baseline, an intervention (such as education about the voter photo ID law), and any subsequent difference in behavior (such as increased voter turnout). 14 References Ansolabehere, Stephen. 2009. “Effects of Identification Requirements on Voting: Evidence from the Experiences of Voters on Election Day.” PS: Political Science & Politics 42(1): 127-130. Atkeson, Lonna Rae, Lisa Ann Bryant, Thad E. Hall, Kyle L. Saunders, and R. Michael Alvarez. 2010. “A New Barrier to Participation: Heterogeneous Application of Voter Identification Policies.” Electoral Studies 29: 66-73. Barreto, Matt A., Stephen A. Nuño, and Gabriel R. Sanchez. 2009. “The Disproportionate Impact of Voter-ID Requirements on the Electorate – New Evidence from Indiana.” PS: Political Science & Politics 42(1): 111-116. Battaglia, Michael P., David Izrael, David C. Hoaglin, and Martin R. Frankel. 2004. “Tips and Tricks for Raking Survey Data (A.K.A. Sample Balancing).” Paper presented at the annual meeting of the American Association for Public Opinion Research, Pointe Hilton Tapatio Cliffs, Phoenix, Arizona. Accessed from: https://www.amstat.org/sections/srms/proceedings/y2004/files/Jsm2004-000074.pdf. Gomez, Brad T. 2008. “Uneven Hurdles: The Effect of Voter Identification Requirements on Voter Turnout.” Accessed from: http://myweb.fsu.edu/bgomez/Gomez_VoterID_2008.pdf. Hood Ⅲ, M. V. and Charles S. Bullock Ⅲ. 2008. “Worth a Thousand Words? An Analysis of Georgia’s Voter Identification Statute.” American Politics Research 36(4): 555-579. Mycoff, Jason D., Michael W. Wagner, and David C. Wilson. 2009. “The Empirical Effects of Voter-ID Laws: Present or Absent?” PS: Political Science & Politics 42(1): 121-126. The Office of the Secretary of State. 2015. Vote Results. Accessed from: http://elections.sos.state.tx.us/index.htm. Wilder, Forrest. 2014. “Texas Has Issued Seven Times More Auctioneers’ Licenses than Voter ID Cards,” The Texas Observer, November 3, 2014. 15