Research Question #5 was “Is the effectiveness of dispensary bans dependent on them being associated with less dispensaries located near high schools?” The hypotheses associated with RQ5 propose that, H5.1) dispensary bans are negatively associated with the number of dispensaries being located near high schools; H5.2) the number of dispensaries located near schools is positively associated with students’ likelihood of using marijuana; and H5.3) the relationship between city dispensary bans and high school students’ marijuana use is mediated by the number of dispensaries operating near schools. Table 7.20 presents the results of the mediation analysis assessing the association between the count of unlicensed dispensaries located within 2,000 feet of a school on students’ lifetime marijuana use. The relationship between dispensary bans and lifetime marijuana use was negative and non-significant , as reported in Chapter 5. Whether a city had a dispensary ban was then regressed on the number of unlicensed dispensaries located within 2,000 feet of a participant’s school . The association between dispensary bans and dispensaries being located within 2,000 feet of a student’s high school was negative and statistically significant , which supports H5.1 for lifetime marijuana use, i.e., that dispensary bans would be associated with less unlicensed dispensaries being located near schools than policies that allow dispensaries. The number of unlicensed dispensaries within 2,000 feet of the school was then regressed on the likelihood of lifetime marijuana use, which revealed a statistically significant positive association between unlicensed dispensaries being located near a high school and the proportion of students there reporting lifetime marijuana use , confirming H5.2., that the number of unlicensed dispensaries located near a school is positively correlated with lifetime marijuana use. Finally,rolling grow tables the regression analysis of dispensary bans and lifetime marijuana use was repeated including the number of unlicensed dispensaries within 2,000 feet of the school in the model.

Accounting for the number of unlicensed dispensaries within a 2000-foot radius of the school slightly increased the association between dispensary bans and lifetime marijuana use among participating high school students but did not make it statistically significant , therefore refuting H5.3, that the effectiveness of dispensary bans is dependent on how well they prevent unlicensed dispensaries from being located near schools compared to city policies that allow dispensaries.These steps were repeated for the recent marijuana use outcome. As path c and a were the same for recent marijuana I began by regressing the number of unlicensed dispensaries within 2,000 feet of the school on the likelihood of recent marijuana use, which revealed a non-statistically significant positive association with the proportion of students there reporting recent marijuana use , which in contrast to the relationship observed for the number of dispensaries located near a school and lifetime marijuana use, did not support the hypothesis that the number of unlicensed dispensaries located near schools is positively correlated with recent marijuana use among the students attending that school . Finally, the regression analysis of dispensary bans and lifetime marijuana use was repeated including the number of dispensaries within 2,000 feet in the model. Accounting for the number of dispensaries within a 2000-foot radius of the school slightly increased the effect but did not make the association between dispensary bans and lifetime marijuana use among participating high school students statistically significant , and therefore did not support , my theory that there was an indirect relationship masking an association between dispensary bans and lower rates of recent marijuana use among students, i.e., that the effectiveness of dispensary bans was dependent on them allowing less dispensaries to be located within 2,000 feet of schools compared to city policies that allow dispensaries. Experimentation with marijuana typically begins in adolescence but the later in life and less frequently a young person uses marijuana, the less likely they will be to experience mental, physical, or social problems related to marijuana use .

Preventing youth use is a frequently stated goal of dispensary bans but to date there have been no known studies focusing on the effectiveness of dispensary bans in preventing adolescent marijuana use. Despite the frustration that the California tradition of local control has caused for medical marijuana consumers and law enforcement, it has also provided a valuable opportunity to evaluate whether city policies that ban marijuana outlets have a localized impact on youth marijuana use. This is important because there are few tools available at a city level to prevent underage youth marijuana use. The question that logically follows whether dispensary bans are effective in preventing youth marijuana use is how they influence adolescent marijuana use behavior. For example, does the effectiveness of dispensary bans depend on how successful they are in keeping dispensaries out of city? Or, is the effectiveness of dispensary bans dependent on them being more effective at preventing dispensaries from being located near high schools? To answer these questions, I examined trends in high school students’ marijuana use behavior before and after a restrictive dispensary policy was enacted in the City of Los Angeles, tested for cross-sectional associations between dispensary bans and student marijuana use and explored explanatory theories for why dispensary bans and more restrictive policies might have an effect. Overall, I found limited support for the efficacy of dispensary bans but strong support for enforcement-related factors. The focal relationship for this dissertation, the effect of MMD bans on adolescent marijuana use was not statistically significant among 57 cities in LA County. There was nevertheless a great deal of information learned from the difference-in-difference analysis. I found that in the City of Los Angeles, Proposition D represented a clear voter mandate for better enforcement and tighter regulations on dispensaries that when carried out was powerful enough to reverse an increasing trend in lifetime marijuana use among city students despite the continued presence of dispensaries in the city. I found that only the continous distance to the nearest dispensary within LA County had a statistically significant mediating effect on the relationship between dispensary bans and student marijuana use , although it fell short of making the relationship between dispensary bans and student marijuana use statistically significant.

This result indicates that the prevenative influence of dispensary bans is partially dependent on the degree to which they are associated with longer distances between schools and the nearest unlicensed MMD. The results from the mediation analyses also supported the importance of enforcement, particularly when it concerns closing down or preventing unlicensed outlets and the need for further study of localized effects. It is possible there was data missing for too many cities to detect an effect for city MMD bans or the rate of MMDs at a city level but that results were found for local measures and not for comparisons conducted by city could indicate that variation in adolescent marijuana use occurs within cities. In this study, students’ marijuana use was more strongly associated with the proximity of the nearest unlicensed dispensary to their school and the density of dispensaries within a several blocks from their school. These localized effects highlight the importance of enforcing city regulations that restrict dispensaries from operating near schools,growing rack whether those regulations are minimum distance requirements or policies that ban dispensaries altogether. Furthermore, that localized effects were noted only for unlicensed outlets and not for licensed dispensaries indicates that enforcing existing ordinances by closing unlicensed outlets near schools could be an excellent first step for cities looking to prevent marijuana use among their students. My first research question concerned whether city regulations that ban or restrict dispensaries influenced trends in high school students’ marijuana use . I hypothesized that over time, cities that allow dispensaries would experience greater increases in marijuana use among high school students relative to cities that allowed dispensaries throughout the study period . I was unable to test the impact of dispensary bans over time among the all the cities in LA County due data limitations, but when I tested the impact of polict that significantly tightened regulatiuons on dispensaries in the City of Los Angeles I found that enacting tighter dispensary regulations can have an impact on rates of high school students’ marijuana use, even as a smaller number of dispensaries remain in active in a city. This hints at a threshold effect, suggesting that the impacts of dispensaries in at a city level can be minimized if their number is kept below a certain point. However, this would not prevent localized effects within cities in the neighborhoods where the dispensaries are located. When comparing all of the LA County cities represented in the data, dispensary bans had little effect on outcomes in high school students’ marijuana use. Although the observed effects for city dispensary bans in the multivariate regression models were in the expected direction, the influence of dispensary bans on students’ marijuana behaviors was qualitatively small and was not statistically significant when accounting for individual characteristics and school type.

It is possible that because people under the age of 18 are not able to access marijuana directly from dispensaries, attending school in a city that allows dispensaries has little effect on their ability to obtain marijuana, which state and national-level studies have documented as already being quite easy . It is possible that city dispensary policies are too distal to what is important to high school students to have a qualitatively important impact on their attitudes and behaviors pertaining to marijuana use. Alternatively, there may be other, more influential distal factors at work, like societal attitudes toward marijuana and social norms surrounding substance use that been changing statewide and even nationally. One possible explanation for the lack of impact of city dispensary bans on high school students’ marijuana use behavior and attitudes is that storefront medical marijuana dispensaries are not the only type of medical marijuana business in operation in LA County. Interestingly, however, analyses conducted for earlier survey years that were subject to similar data limitations . did identify a statistically significant negative association between a city having a dispensary ban and lifetime marijuana use . It appears that by the next year, the 2015/2016 school year, dispensary bans were no longer associated with lower rates of even lifetime use among high school students. What could explain this change? One difference between the 2014/2015 school year and the 2015/2016 school year was that only four LA County cities allowed medical marijuana dispensaries in 2014, compared to six in 2016 and of the two new cities included in this group one did not have any active dispensaries before the end of the study period, while the other, Huntington Park, allowed four dispensaries but only in industrial zones. As this dissertation has demonstrated, the distance to the nearest dispensary was among the most powerful influences on student marijuana use behavior. By allowing dispensaries to locate only in industrial zones, the City of Huntington Park may have prevented their city ordinance allowing MMDs from having an impact on marijuana use behaviors among the young people attending their schools. Another possible explanation for the association between dispensary bans and lifetime marijuana use noted during the 2014/2015 school year disappearing for the 2015/2016 and 2016/2017 school years may be an almost exponential growth in medical marijuana delivery services in LA County. For example, even after the number of storefront MMDs has been curbed in some areas of LA County, growth in medical marijuana delivery services seems to have continued unabated, regardless of the city policy where they are located. Only 25 percent of the cannabis consumed in the state is purchased from government-approved brick-andmortar retailers, according to a report released in February 2019 by the Cannabis Growers Association. Much of the rest is sold door-to-door by hundreds of unlicensed, small and independent couriers . I was not able to test or control for this factor in my analysis. Empirical data measuring the growth in marijuana delivery services in LA County is available only via the same commercial services by which I obtained addresses for LA County storefront dispensaries, but I do not have access to any data archives from these sources for the years between 2015 and 2017 and did not track the numbers of delivery services other than when I recorded the number and location of unduplicated dispensaries listed by Weedmaps and other dispensary listing websites in September 2016.