Analyzing the Marijuana Reform Movement in Texas as a Participant and an Observer: A Case Study Approach

Project Overview and Context

For more than four decades, the United States has fought a drug war based on the false premise that harsh punishments for using and selling drugs deter these activities. Over 40 years of evidence demonstrates that the drug war has done little to reduce the supply or demand for drugs. Instead, it has contributed to mass incarceration and increasing racial disparities in the criminal justice system; it has torn apart families and communities; it has made it difficult or impossible for some citizens to obtain housing, education, and employment; and it has cost taxpayers billions of dollars per year.

Increasingly policymakers and the public are reckoning with the futility of the war on drugs, reflected in the slow shift in drug policy focus from punishment to treatment. One of the most visible signs that the drug war is waning is the state-by-state repeal of marijuana prohibition. Currently nine states have legalized marijuana for adult use, 29 have legalized it for medical use, and 14 others have decriminalized its use.

Yet marijuana offenses continue to account for the majority of drug-related arrests (Drug Policy Alliance, 2017). This is at odds with how the public views marijuana use. In 2016, 44% of Americans reported having used marijuana in their lifetime and a slight majority (51.8%) of 18- to 25-year-olds reported having used it at least once (National Survey on Drug Use and Health, 2017). In 2017, a Gallup poll found that 64% of Americans agreed that marijuana should be legal (McCarthy, 2017). Marijuana is not as harmful to users or to society as other drugs that are legal, particularly alcohol and tobacco, and there is a growing wealth of data that point to the potential medical benefits of the plant (Neill & Martin, 2017). Most of the harms that come from marijuana are not from its use but from the penalties associated with its use.

I came to Houston, Texas, in 2014 to work as a postdoctoral fellow in drug policy at Rice University’s Baker Institute for Public Policy. At that time, 23 states had legalized marijuana for medical use and two states had legalized it for adult use. But none of these states were in the South, and I assumed that Texas was still years away from meaningful changes to drug laws. So I was surprised when I found a large, enthusiastic, and well-organized movement for marijuana reform that was preparing for the 2015 state legislative session.

The mission of the Baker Institute’s drug policy program is to pursue research and pragmatic policy solutions focused on reducing the harms caused by drug use. I was eager to contribute to this mission, and I began forming relationships with local and state policymakers and other actors in the drug policy arena. I went to meetings of the Houston National Organization for Reform of Marijuana Laws (NORML) and Republicans Against Marijuana Prohibition (RAMP) and was eventually introduced to the leaders of the state coalition for marijuana reform, Texans for Responsible Marijuana Policy (TRMP).

My position as a drug policy researcher at Rice University made it easy for me to establish contacts within the marijuana reform movement. Activists were excited to have a nonpartisan think tank on their side; they perceived our institution’s support as a boost to their credibility in the eyes of skeptical legislatures and members of the public. I participated in strategy meetings and lobbying days at the capitol, wrote material summarizing the evidence on the impact of marijuana reform in other states, and in April 2015, I testified in favor of a bill to reduce marijuana possession of 1 ounce or less from a Class B misdemeanor to a civil penalty.

My participation in the mobilization for marijuana reform in Texas that took place in late 2014 and the first half of 2015 offered me an opportunity to study closely how grassroots organizations try to influence policy decisions and the challenges that such organizations face.

Research Design

The concept for this project1 evolved with my involvement in the marijuana reform movement. My primary motive for participating was to have an impact on legislative outcomes, but as my participation in the movement progressed, I felt that the movement’s activities were worth studying. This was an inductive approach to research, as I allowed my observations to guide my hypotheses and theoretical development.

At this stage in the research process, I did not have a fully developed theoretical framework. A framework is important because it “connects the researcher to existing knowledge … and forces [the researcher] to address questions of why and how” about the phenomenon being observed, rather than just describing that phenomenon (University of Southern California, 2018). That said, it is also the case that because qualitative research often begins inductively, a pre-determined framework is not necessary and may not be ideal. Some argue that in qualitative studies, researchers should focus on “‘bracketing’ (i.e., masking or trying to forget) their a priori theories so that it does not influence the collection of data or any meanings assigned to data during an investigation” (Lederman & Lederman, 2015, p. 596). The theoretical framework is useful for providing context for the research question and making decisions about how to answer the question, but it does not have to be the project starting point, especially not for qualitative research.

Once I decided I wanted to study the reform movement’s impact, I began taking detailed notes following my interactions with movement members and my participation in events. My initial participation and early observations of the movement, combined with an extensive review of the social movement and public policy literature, led to the decision to use John Kingdon’s (1995) concept of multiple streams in the policy process to frame the study.

Kingdon argues that policy-making is nonlinear and is comprised of “three primary processes, or streams, [of] problem recognition, identification of policy proposals, and politics” that develop independently and may become “coupled” (Neill Harris & Morris, 2017, p. 97). These processes may come together for a variety of reasons, such as crisis events or major changes in political administrations; this coupling can create an opportunity for policy change. Social movements often seek to influence all three areas and will try to capitalize on major focusing events, but the impact they are able to have varies across different stages of the process. Kingdon’s framework is a useful conceptual tool for understanding social movement impact on policy action and is not often applied to the study of grassroots activism for criminal justice policies.

The purpose of this study from a theoretical perspective was to gain a deeper understanding of the relationship between social movements and policy action as seen in the context of Kingdon’s multiple streams model. The practical purpose was to identify ways in which the marijuana reform movement in Texas could be more effective in future legislative sessions. To that end, the underlying research questions were, why has the marijuana reform movement in Texas failed to achieve policy success, despite its strong organization and resource mobilization? Can this failure be better understood by framing the problem in terms of a policy streams argument?

A case study was the best way to approach this topic because issue framing and political context were key to understanding movement impact, themes not well captured with quantitative data. My involvement in the reform movement naturally steered the research toward a participant-observer design supplemented with interview data. Participant observation enables “researchers to learn about the activities of the people under study in the natural setting through observing and participating in those activities” (Kawulich, 2005). This method of study helps the researcher form relationships with individuals in the population of interest, allows for the collection of different kinds of data, and guides the formulation of interview questions.

Participant observation dovetails nicely with interview research because the researcher’s participation and observations inform whom to interview and what questions to ask, and the researcher’s own observations can verify information obtained through interviews, while the interviews also serve as a check on validity of the researcher’s interpretations of events. Observational and interview data were supplemented with review of relevant documents, including organization material, legislative decisions, and media coverage to provide a fuller picture of social movement activity in its political context. Using multiple data sources, or data triangulation, improves the validity and reliability of research findings, and can improve understanding of the research topic (Patton, 2001).

Research Practicalities

Due to my participation in the reform movement, I had relationships with most key participants. A mixture of purposive and snowball sampling was used. Purposive sampling relies on the judgment of the researcher to determine the appropriate sample. This requires preexisting knowledge of the population in question, which participant observation provides. Despite having contacts with key actors in the movement, it was likely that there were individuals important to the movement of whom I was not aware. To account for this, I also relied on snowball sampling by asking interview respondents for referrals to other people I should speak with about marijuana reform. There are biases built in to these sampling strategies, as they rely on the judgment of the researcher and the initial interview respondents, but for qualitative analysis of a small population, these techniques can be quite effective (O’Sullivan, Rassel, & Berner, 2008).

I waited until the legislative session ended in June 2015 to begin collecting interview data because I felt that an accurate perspective of organizational influence required knowledge of policy outcomes. Interviews with organization leaders, activists, and other participants prior to the end of session could have resulted in skewed perceptions of the impact of advocacy activity. Based on my observations of movement activity, I identified individuals whom I regarded as key informants; these were typically individuals who served in leadership roles or as points of contact in the reform coalition. I attempted to interview these individuals first and to allow findings from these interviews to guide future interviews (Neill & Morris, 2017).

Most interview requests were initiated via email, with follow-up phone calls when necessary. I explained the purpose of my research and offered each individual the interview method of their choice—in person, over the phone, or email response to questions. All but two of the 40 people I interviewed opted to talk over the phone.

All interviews were confidential. Whereas some respondents were happy to speak on record, others were concerned about the possibility of their employers finding out about their marijuana activism, or about media outlets or other activists learning of their criticisms of the movement. Due to these concerns, I opted to make all interviews confidential so that participants could feel more secure about sharing honest feedback.

Some persistence was required to secure interviews with state legislators. Interviews with these respondents can be understood as “elite” interviews, because respondents were chosen specifically for their positions as state legislators (Hochschild, 2009). As part of the population directly responsible for deciding the fate of marijuana reform legislation, their insights were critical to the research questions. I followed up on unanswered emails with phone calls and this increased the number of positive responses I received. I found that most of the legislators who consented to be interviewed were very busy and for that reason had not responded to my initial requests. To the extent that there was any hesitation, this was due to concerns that their comments would be made public, making it especially important to stress interview confidentiality to this group. It was also important to know all available information regarding the respondents’ stances on and actions toward marijuana reform prior to the interview, so that I could be informed and not “waste” respondents’ time (Hochschild, 2009). I conducted my interviews with state legislators and their aides last or nearly last to ensure that I had as much information as possible to ask highly relevant and informed questions.

Method in Action

Data collection included observation of and participation in activist activities and legislative hearings, review of organizational documents, legislative activity, and media coverage of marijuana reform, and 40 interviews with marijuana activists and opponents, legislators, and legislative aides.

Interview respondents were categorized by their position on marijuana reform and their role in the reform movement. While the majority of respondents (31) were proponents of full legalization, 10 supported medical marijuana legalization and decriminalization of possession but not full legalization, seven supported medical marijuana legalization only, three supported legalization of only a limited, low-THC (tetrahydrocannabinol) form of cannabis oil, and three opposed all reforms. Interview respondents included 23 active coalition members, six individual supporters not active in the coalition, eight legislators or their chiefs of staff, seven active or retired criminal justice actors, seven patients or patient caregivers, and three individuals representing policy organizations that were not marijuana specific. Roles in the movement were not mutually exclusive; for example, several respondents who were active coalition members were also criminal justice actors, patients, or patient caregivers. The high representation of coalition members in the interview sample was desirable because the study aimed to understand the impact of movement efforts, making actors in that movement a central data source.

In total, I requested interviews with 50 individuals. Of the 10 who declined or did not respond to my request, nine were legislators or their aides. None of the legislators I contacted who had been vocal critics of marijuana reform agreed to be interviewed. This was unfortunate because it meant the study would not include the perspectives of lawmakers who remained opposed to reform despite movement efforts. However, three individuals who supported full marijuana prohibition did agree to interviews. Although this is a small sample, the respondents represented law enforcement and conservative lobbying organizations that were the most notable opponents of marijuana reform during the legislative session. This gave me reason to believe that they provided a sufficient perspective regarding the case against marijuana reform that was made to state legislators.

Interviews were conducted in person or on the phone between May and December 2015. Prior to the start of each interview, I reminded participants that their responses were confidential and asked their permission to record the interview, to which all consented. Interview questions were developed from previous knowledge of the marijuana policy context in the state, an initial literature review, and early discussions with central participants in the movement.

Interviews were semi-structured and lasted from 20 min to 2 hr. Respondents were asked about a variety of issues relevant to the marijuana policy field and the study of social movement activity, including the impetus for their involvement, organizational strategies, the role of messaging in reform efforts, the role played by the media, the political context surrounding reform, and the successes, challenges and lessons to take away from the 2015 legislative session. (Neill & Morris, 2017, p. 102)

Interview questions were the same or similar for all respondents, with some variation in questions for legislators/chiefs of staff and reform opponents. For example, it would have made little sense to ask reform opponents about organization strategies used by the marijuana advocacy coalition; instead, opponents were asked about their own tactics for persuading lawmakers to oppose reform and their perceptions of the marijuana reform movement in the context of their opposition. Legislators and their chiefs of staff were asked more specific questions about their perceptions of the support for reform among their colleagues and how the movement was viewed from the perspective of elected officials.

Interview data were organized using template analysis, which involves creating codes based on themes found in the data. This analytic approach is readily adaptable to different research settings, allows for flexibility in data interpretation, and is well-suited to comparing different perspectives within a given context (King, 2004). Codes describe the content of text which they identify. For example, several respondents discussed the importance of framing their arguments for reform in terms that appealed to their different audiences. Data related to this point were coded as “issue construction.” As themes were identified, they were organized in relation to one another, a technique called hierarchical coding. Hierarchical coding involves grouping together related themes that emerge from the data to create “master themes” (King, 2004, p. 257). This is a useful organizational tool because it allows for different levels of specificity in analysis, depending on research needs. Themes that emerged from interviews were compared with those identified during participant observations.

Interview data were initially organized according to naturally emergent themes, a process that was instrumental to analyzing the data within the context of the multiple streams framework. The multiple streams framework has several elements—problem recognition, policy solutions, the political environment, and opportunities for policy change, or policy windows. It was clear that the movement’s greatest success came in getting marijuana prohibition recognized as a problem, evidenced by the 11 marijuana-related reform bills introduced in the 2015 legislative session. The conclusion drawn from participant observations, analysis of media coverage, and interview data was that message framing, delivery, and content were central to getting marijuana prohibition recognized as a problem.

The reform movement was less successful in having its desired policy reforms adopted. This was apparent at the end of the legislative session, when the only bill signed into law was to allow access to a very limited, low-THC strain of cannabis oil for use by Texans with intractable epilepsy. This bill was advocated for by Compassionate Access for Epilepsy (CAFÉ Texas), a group that worked diligently to separate itself from the other groups advocating for broader marijuana reform. Indeed, members of both groups agreed that the cannabis oil legalized by the bill was not equivalent to marijuana.

The failure of broader reforms led the analysis to focus on why a window of opportunity for reform did not open. The sentiment among the majority of respondents, including state legislators, was that marijuana reform failed to advance through the Texas legislature due to continued political resistance. Many respondents noted that despite their efforts to educate elected officials, “ignorance” and “stigma” against marijuana remained strong. Despite support for reform among some Republicans, the party leadership in the governor’s office had voiced its opposition, making it politically risky for Republicans to support. Many Republicans also had the persistent fear that supporting marijuana reform would make them vulnerable to a more conservative primary challenger (Neill Harris & Morris, 2017).

Although the conservative political environment in Texas made marijuana reform an uphill battle, interviews with participants revealed that some felt the movement could have done a better job communicating its message. Advocates concerned primarily with medical access felt the coalition did not effectively separate the issues of medical and adult use, and reported receiving feedback from lawmakers that this was the case. The concern among elected officials that medical legalization was a “slippery slope” to full legalization was stated repeatedly, and medical marijuana activists felt that the coalition did not assuage these concerns effectively. Although there were two marijuana coalition groups, TRMP, whose stated mission was to end marijuana prohibition, and Texans for Medical Freedom, whose mission was focused specifically on medical access, it was difficult in practice to tell the difference between the two because they had the same lobbyist and significant overlap in organizational membership. The apparent closeness of the two groups may have confirmed the fears of conservative lawmakers that support for medical marijuana could be construed as support for full legalization (Neill & Morris, 2017).

Some medical marijuana activists expressed a desire for notable pro-marijuana groups to refrain from advocating for medical access because they felt this advocacy confused the message. Although some full legalization advocates acknowledged the need to better separate the issues of medical and adult-use access, they felt strongly that they could continue to advocate for both. This rift highlighted a central challenge for social movements: the inability to control membership. Movements “cannot prevent individuals or groups from participating, even if that participation is detrimental to the cause” (Neill & Morris, 2017, p. 118). Because medical marijuana advocates cannot prevent other groups from advocating for medical access, their ability to control the frame of their issue is limited.

The coupling of interview data with participant observation was critical to understanding the disagreement over movement strategy as it related to medical and adult-use marijuana advocacy. If I had not seen movement efforts in action, I would have been limited in my ability to evaluate the feedback from interview respondents. For example, one memorable moment at the House Committee on Criminal Jurisprudence’s hearing for the bill to reduce marijuana possession from a misdemeanor to a civil penalty came when a woman wearing small, bright green shorts testified in favor of the bill. Her testimony focused on her positive experiences participating in drum circles with homeless people in Austin. There are no dress code requirements for offering legislative testimony, and citizens are free to express themselves as they choose, but movement leaders felt that in their battle against decades of negative stigma, this was not an ideal spokesperson. One coalition leader told me that she had encouraged the woman to only sign her name in support of the bill rather than offering testimony, but was not able to persuade her. This woman was not the reason that the civil penalties reform did not pass, but it is a telling example of the movement’s challenge in terms of controlling the message for reform. Indeed, a later interview I had with a state representative that was on that committee specifically mentioned “the woman in the green shorts” in his advice to the movement that it needs to present itself more professionally.

The interviews were effective in gaining information about movement efforts. Most people seemed eager to share their experiences and expressed appreciation that a researcher was taking an interest in their efforts. Overall, my role as a participant-observer seemed to bolster respondents’ willingness to share information with me. However, it must be acknowledged that my role affected the information collected from interviews, and it is possible that some respondents were more hesitant to share their true feelings because of my involvement. I tried to mitigate this issue by making the interviews confidential, but some degree of influence over the interview was unavoidable.

The shortest and least detailed interviews occurred with opponents of reform. In contrast to reform advocates, I had no prior relationships with these individuals. I was careful not to express my own opinions during the interviews, but it was probably known to the interview participants that I supported reform. This may have made them less open during the interview. Despite efforts to maintain an impartial tone and ask neutrally worded questions, I felt that respondents remained guarded. It is possible that this assessment is inaccurate, and that I was instead projecting my own expectations onto the interviews. However, that these were the shortest interviews lends some credibility to my perceptions.

I shared the major takeaways from the research with interview respondents and eventually published the research in a peer-reviewed journal. I was pleased with the research project and the article that came from it, but I could have done more to work with activists to implement some of the lessons learned from 2015 in the 2017 legislative session. At the time of the research, I had full intentions to be actively involved in 2017 for this purpose; however, I was on maternity leave at the height of session activity. It is impossible to know now whether my attempts at greater involvement would have been well-received; on one hand, movement activists were happy to have the support of the Baker Institute behind their cause; on the other, efforts to insert myself more directly into leadership decisions could have been perceived as unwelcomed interference.

Practical Lessons Learned

This study of the marijuana reform movement in Texas drew primarily from participant observation and interviews with reform advocates and other key actors during the 2015 legislative session. I learned a great deal from this experience, and these lessons will inform my future research. Here I will distill a few of the key lessons from which I think other researchers can benefit.

Start with an understanding of your research goals and how these goals fit with your occupation. The way one goes about choosing a subject for research and designing the research project will depend to a large extent on how that project fits with one’s professional occupation. A student working on a dissertation or looking to publish in an academic journal may begin the process deductively, seeking to test a specific theory and looking for a subject to do so. If one’s work is oriented more toward a nonprofit or think tank setting, the source for a research topic may come from real-world observations of phenomena that seem relevant to the organizational mission. In my case, my professional career emphasizes research that informs practice. This does not mean that theoretical development is not important, but it is not the main driver of research. My decision to study the marijuana movement in Texas came from observations that this was a topic relevant to my general mission of advancing drug policy reform. The underlying reason for a research project—whether it is theoretical development, practical application, or both—will drive how decisions are made during the research process.

Familiarize yourself with the subject matter. An extensive literature review is essential to data collection, as it provides context for interview questions and areas of focus. Before I conducted my interviews, I read about prior efforts for marijuana reform in Texas, efforts for marijuana reform in other states, social movements in other areas of criminal justice and civil rights, and the successes and failures of social movements more generally. This review was essential to my development of the research questions and subsequent interview questions.

Use gentle persistence with hard to reach groups. Researchers must make ethical considerations regarding what they ask of research participants and should respect an individual’s decision to not participate. That said, nonresponse to an interview request should not automatically be viewed as a refusal. Some people are more likely to ignore emails from people they do not know or that request their time. This was particularly the case for the state legislators I interviewed, who constantly receive interview requests. With this population, I found that it was especially important to send follow-up emails and make phone calls. In the case of the 10 legislators/chiefs of staff who did not respond to my interview requests, I did eventually accept their nonresponse as a refusal to participate, but not until I had left multiple messages and sent multiple emails. It was important to not give up immediately on this population, or I would have had zero interviews with state lawmakers.

Do not allow your own views to invade the interview. This is especially important when the subject matter is one that has clear “sides.” I was careful to phrase all interview questions in neutral terms, but due to my professional involvement in movement activities, I struggled with keeping my own views out of the interview, especially because many respondents were aware of my views. Although in the majority of cases I think my role as a participant made respondents more willing to share information with me than they would have been if they had not known me previously, I think in some cases, particularly in interviews with reform opponents, that knowledge of my support for and participation in reform efforts could have caused respondents to be more constrained in the interview.

Conclusion

This case analyzed the impact of the marijuana reform movement on policy change in Texas. The study was a case analysis based on participant observation and interviews with key actors in the reform movement. Kingdon’s multiple streams theory of the policy process was used to provide a framework for understanding the opportunities and limitations that the marijuana movement faced in the conservative Texas political environment. Although the movement was successful in getting marijuana prohibition recognized as a problem, it was unable to move the Texas legislature to adopt its desired policy reforms. This failure was due largely to enduring elements of Texas’s conservative political climate, but observations during the legislative session and interviews with participants also revealed that one of the challenges the movement faced was the inability to distinguish adequately between the various policy goals of marijuana decriminalization, medical legalization, and adult-use legalization. This finding highlights a challenge for social movements more generally, which is the limited control they have over participation and messaging.

There were limitations to this study. Although participant observation benefited the project, there were drawbacks to this approach, as it affected the interview process. Also, interview data were heavily skewed toward supporters of reform. Despite efforts to talk to opponents, I only spoke with three people who opposed reform. The study would have benefited from greater input from this side of the issue. That said, the study was primarily focused on the movement for reform, and thus it also makes sense that reform advocates made up the bulk of interview participants.

This was a single case study and therefore not generalizable to other movements. Also, as movement efforts continued in 2017 and will again in 2019, this study did not capture movement efforts in their entirety. However, social movement scholars have stressed the need for more analyses of social movements that study them at different points in development (Andrews, 2001; Benford, 1997), and this study lays the groundwork for longitudinal analysis. This research also provides analysis regarding opportunities and limitations that social movements may face, particularly as they relate to issues of criminal justice reform, and particularly in conservative states that do not allow for ballot initiatives as a means for policy change.

A case study is ideal for research questions focused on understanding how and why human behavior has a particular impact. The methods discussed here, participant observation and interviewing, were ideal for gaining a holistic understanding of how the movement for marijuana reform operated within the constraints of Texas’s political environment. The project provided important insights into the policy process, the role social movements play in that process, and the future opportunities and challenges that movements for marijuana reform and other criminal justice reforms may face.