Some methodological limitations should be considered. First, preexisting differences in neurocognition, which may increase risk for substance use , cannot be ruled out in this cross-sectional study. Second, given the studies suggesting decreased motivation associated with marijuana use , the observed cognitive differences may be due to amotivational influences on test performance. Third, we used composite scores for data reduction purposes, and although common practice, they may not reproduce in other samples. Fourth, results may not generalize to other samples with different lengths of abstinence, patterns of substance use , gender or ethnic distribution, or SES0parental income. In conclusion, the general pattern of results suggested that even after a month of abstinence, adolescent marijuana users demonstrate subtle deficits in psychomotor speed, complex attention, planning and sequencing, and verbal story memory compared with non-marijuana using teens. Increased frequency of lifetime marijuana use was also associated with decreased performance in these areas. Implications include the need for psychoeducation aimed at informing adolescents and parents of the potential long-term cognitive consequences of heavy marijuana use. Longitudinal studies are critical to help rule out premorbid influences on cognitive function and to assess the developmental trajectory of neuropsychological functioning among adolescent marijuana users over time.College-going individuals in the United States may have unique attitudes toward substance use behavior and tobacco use, grow vertical including shifts in attitudes and behaviors that are associated with the constantly changing product landscape of alternative tobacco products , such as electronic-cigarettes .
Psychosocial behaviors and campus culture, including class attendance, peer socializing, campus policies, and residential environments, may also facilitate these unique attitudes toward favorability of smoking among college subgroups, while also introducing a unique risk environments for tobacco initiation, uptake, transition, and use . In addition, part of the variation explaining these health behaviors may be influenced by the specific demographic and socioeconomic characteristics of a college campus population and community. Data from social media platforms are often used to self-report and publicly communicate health-related attitudes and behaviors . Young adults [ages 18–25 ] in the United States are much more likely than older populations to actively use social media, including popular platforms Twitter, Snapchat, and Instagram . Infoveillance research, which uses online information sources to detect trends about the distribution and determinants of disease, including health knowledge and behaviors, has been used to develop insights on numerous public health issues including infectious diseases, vaccination sentiment, opioid use disorder, mental health issues, and, relevant to the exploratory aims of this study, tobacco and alternative tobacco use attitudes and behavior . However, smoking-related discussions on social media tied to specific colleges with geographic specificity has not been widely investigated. Existing studies using social media to examine tobacco-related attitudes and behaviors in college aged populations have primarily focused on evaluating the impact of social media health promotion anti-tobacco campaigns, recruiting hard-to-reach college populations using social media platforms, and examining the influence of exposure to tobacco-related social media content and marketing on current and future behavior and use .
Other research on college-aged populations has focused on assessing tobacco initiation and transition of use patterns, particularly as new alternative and emerging tobacco products become available . Accelerating research using social media to assess tobacco-related attitudes/influences among youth has also been supported by U.S. Federal initiatives, including projects funded by the National Cancer Institute and U.S. Food and Drug Administration Tobacco Centers of Regulatory Science, which for have identified and characterized e-cigarette advertisements on image-focused social media sites and tobacco user experiences with little cigars and e-cigarettes as discussed on Twitter . Changes in local, state, and national health policy related to tobacco and other products smoked or used concurrently with tobacco and electronic cigarettes can also have an impact on attitudes and behaviors of these populations. For example, recent debate in the United States regarding the legalization of marijuana/cannabis may positively influence marijuana-related attitudes for college populations, who tend to skew toward more liberal policies regarding decriminalization, legalization, and increased access . Similarly, the 2019 outbreak of e-cigarette and vaping-related lung injury associated with products containing tetrahydrocannabinol may dissuade tobacco or THC use in certain young adult populations, particularly since they were most heavily impacted by the disease . Examining the changing public attitudes and behaviors of college-aged smokers is particularly salient for the State of California, USA. As of January 2014, all campuses in the statewide University of California system became tobacco free , and the California State University system followed suit in 2017 . In addition, voters in California approved Proposition 56 in late 2016, which added a $2.00 increase to the cigarette tax effective April 2017, with an equivalent increase on other tobacco products and electronic cigarettes . Voters in 2016 also approved Proposition 64, which legalized the use of recreational cannabis in November 2016 .
During this time, the popularity of e-cigarettes in the United States was increasing . These changes in policy and preferences underscore the interconnected nature of the Triangulum of tobacco products , including potential for dual-use, transition between products, and challenges associated with conducting surveillance and implementing cessation programs . This changing policy landscape supporting tobacco control measures, as reflected in the shift of California’s public university systems to become smoke-free, is a key impetus for this study. The ability of these colleges to eliminate on-campus smoking relies in large part on understanding past and existing knowledge and attitudes held by the campus smoking populations, along with their perceptions and behaviors that may be associated with compliance or non-compliance to smoke free campus policies. In response, this study conducted exploratory research on the popular microblogging platform Twitter. Specifically, we used big data, data mining, and geospatial approaches to identify and characterize tweets originating from Twitter users specifically geolocated at California 4-year university campuses. Our primary objective was to assess types of tobacco and ATP products mentioned by users, the distribution of user sentiment toward tobacco and smoking behavior, and to assess the feasibility of detecting self-reported smoking behavior that may represent a violation of campus smoke free policies. Secondarily, we also sought to conduct a cross-campus assessment to determine how these factors vary across different university and college communities and over time.The objective of the study’s data collection approach was to obtain a highly refined subset of tweets, which were both posted from college campus’ geolocated coordinates in California and also included user discussions about smoking, in preparation for manual review to more purposefully identify tweets that specifically discussed different types of tobacco and smoking products, sentiment of users toward smoking behavior, and selfreported smoking behavior on campus. Data were collected from the Twitter public streaming Application Programming Interface using the cloud-computing service Amazon Web Services . The public streaming API was set with filters to collect all tweets that included metadata containing latitude and longitude coordinates, initially with no filter for keywords. Tweets were collected continuously from 2015 to 2019. All tweets collected included the text of the tweet and associated metadata, including the date and time of tweets. The use of the public Twitter streaming API to collect data pre-filtered only for tweets including latitude and longitude coordinates represent a subset of all tweets posted during the time frame of the study.
There exists the potential for sampling bias associated with different Twitter APIs that are not representative of all Twitter data , grow racks and data filtered only for geocoded data may omit many conversations from college aged populations about topics, such as smoking, which may be linked to college-related user groups . Though resulting in a much smaller volume of data, our approach nevertheless allows for detection of tweets in specific geospatial bounds at the high resolution of latitude and longitude coordinates in the state of California. Therefore, by using this data collection approach, we were able to isolate tweets originating from geospatial coordinates within the formal spatial boundaries of all 4-year universities in California. To enable this geolocation, a basemap of California 4-year universities from the Stanford Prevention Research Center was obtained and cross-referenced. The SPRC’s basemap included a relational geodata base which classified polygons by college name. College areas were comprised of multiple polygons for different campuses and associated properties, though aggregation was conducted at the overall college level to enable comparison across different colleges. Tweets were then filtered for 37 keywords which were broadly related to tobacco-related topics, including the names and brands of different tobacco and ATPs and descriptive terms associated with smoking and vaping as expanded upon from those used in prior studies . Specifically, the following keywords were used: bidis, cigarette, cigarettes, cigarillos, cigars, cigie, class, dip, e-cig, hookah, huqqa, joint, JUUl, kereteks, Marlboro, Newport, njoy, pipe, roll-up, shag, smoke, smoking, snuff, snus, tobacco, vape, vaped, vapejuice, vaper, vapes, vaping, vapor, waterpipe, waxpen, and weed. The purpose of this keyword filtering was to better isolate smoking-related conversations from all other Twitter discussions occurring on college campuses. After tweets were manually reviewed to positively identify smoking-related conversations originating in these college campuses, a snowball sampling design was employed which compared the frequency of all non-keyword terms in “signal” tweets with the frequency of these words in “noise” tweets . This methodology resulted in the identification and querying for ten additional keywords: 420, 818, blunt, bong, cigs, kush, marijuana, roll, smell, and stoge.After isolating a corpus filtered for tobacco-related keywords in areas geolocated for California 4-year universities, four researchers trained in social media content analysis used an inductive coding approach to identify study characteristics of interest by manually annotating all tweets , following an approach also described in prior studies . Annotators had backgrounds in public health and had experience manually annotating social media posts for tobacco behaviors in prior published research projects . Manual annotation included: identifying the type of smoking product discussed ; assessing positive, negative, or neutral sentiment related to smoking behavior ; and identification of whether the tweet included first-person use or second-hand observation of smoking behavior. Table A1 contains further details about topics that were coded as valid and invalid for positive identification as a “signal” tweet. Tweets that did not express sentiment related to smoking were excluded from analysis of signal tweets. The primary objective of this approach was to conduct exploratory research into what tobacco and smoking products were being discussed by Twitter users at California universities, assess the overall sentiment toward tobacco and smoking by these users, and explore whether it was possible to identify self-reporting of tobacco use-related behavior . Four authors coded posts independently and achieved a high inter rater reliability for overall coding categories and equally high inter rater reliability for specific subcoding for tobacco , vape , and marijuana specific tweet categories. For inconsistent results and any discrepancies related to coding, all authors convened to discuss, confer, and reach consensus on the correct classification informed by the inductive coding approach outlined in Table A1. Analyses of variation across college campuses were limited to the top twenty colleges by tweet volume, as estimates collected from samples of tweets from other colleges may have been biased due to insufficient volume of tweets collected. A p < 0.10 was considered statistically significant for correlational analyses due to sample size limitation. Point density algorithms were used to visualize and detect geospatial trends. Analysis was conducted in R version 4.0.1 and geospatial visualization of data was done in ArcGIS Desktop version 10.7. This project was part of a broader study to examine college campus smoke-free policies using qualitative focus groups and examining social media data with the qualitative analysis approved by the Institutional Review Board at California State University, Fullerton .Data collection resulted in 83,723,435 geo-identifiable tweets located in the state of California in the 5-year period from 2015 to 2019. From these tweets, 1,381,019 originated from 88 CA 4-year colleges, with the five schools contributing the most tweets including UC Los Angeles , Stanford University , UC Riverside , University of Southern California , and UC Berkeley . Thirty-eight schools contributed over 10,000 tweets each, overall representing 89% of the entire corpus of CA 4- year college geocoded tweets. Of these tweets, 7,342 contained smoking-related keywords with approximately one third occurring after 2015. In total, smoking-related topics originating from all geocoded tweets in the state made up an extremely small proportion of all topics and tweets specifically geocoded for CA 4-year universities.