Monitoring the sentiment of cannabis‐related tweets in the lead up to New Zealand's cannabis referendum

Abstract

Introduction and Aims. In October 2020, New Zealanders will vote on whether cannabis should be legalised for recreational use. With this in mind, the aim of the present study is to gauge the views and opinions of the New Zealand population on cannabis via tweets. To achieve this, we conducted a sentiment analysis of all historic cannabis-related tweets and referendum-specific tweets written in New Zealand. Design and Methods. We used a Twitter-sponsored commercial platform to access all historic cannabis-related tweets written in New Zealand and used search terms to remove non-cannabisrelated terms. Next, we used the platform's machine learning function to code the sentiment of tweets (i.e. positive/pro-cannabis, negative/anti-cannabis or neutral). Results. Between July 2009 and August 2020, 304 760 cannabis-related tweets were written in New Zealand. Overall, the tweets were predominantly positive (62.0%) and there was a higher proportion of positive tweets written in 2020 (65.3%) compared to negative or neutral tweets. Similarly, for referendum-specific tweets, the 2020 data reveal a generally positive view of cannabis (53.5%). Discussion and Conclusions. Both cannabis-related, and referendum-specific tweets, suggest that Twitter users in New Zealand have a generally positive view of cannabis. Given the nature of Twitter, the current method will allow us to study whether views toward cannabis change as the referendum nears and capture any late swings in pro-or anti-cannabis sentiment (abcd-lab.shinyapps.io/cannabis_sentiment/).

#TurntTrending: a systematic review of substance use portrayals on social media platforms

“…A total of 15 905 182 substance‐related social media posts (15 804 353 text posts, 68 023 image posts, and 32 470 video posts) were assessed across all 73 studies. Twitter data was analysed in 34 studies [20–53]; 23 studies examined YouTube [54–76]; 10 assessed Instagram content [77–85]; 4 studies used Pinterest images [85–88]; TikTok videos were analysed in 2 studies [89, 90]; and Weibo content was assessed in a single study [91]. E‐cigarettes were the most common category analysed ( n = 24 studies), followed by tobacco ( n = 20 studies), cannabis ( n = 18 studies), opiates ( n = 6 studies), alcohol ( n = 4 studies), psychostimulants ( n = 1 study), stimulants/amphetamines ( n = 1 study), inhalants ( n = 1 study), novel psychoactive substances (NPS) ( n = 1 study) and polysubstance use ( n = 1 study).…”

“…Sample size of included studies varied within social media platforms, with the coding method influencing the quantity of substance‐related content posts coded (Supporting information Table S2). Favoured coding methods included manual coding [20–26, 29, 30, 32–34, 37–39, 42, 43, 48, 50–60, 62–91], machine learning [27, 28, 31, 35, 41, 45, 47, 49, 61] or a combination of manual and machine learning [36, 40, 44, 46]. Manual coding was the most common method, and typically involved one or more human coders categorizing data thematically and sentimentally using a codebook derived from data subsets.…”

“…Studies included in this review identified that user‐generated content was most likely to be responsible for the dissemination of substance‐related content on social media [21–23, 25, 26, 28–31, 33–37, 39–52, 54–56, 58–82, 84, 85, 87–90]. Personal accounts typically expressed positive sentiments toward substance use and shared content relating to Informative or Entertaining themes.…”

“…ies, 12 assessed e-cigarette-related content[24, 26-28, 33, 34, 36, 37, 39, 43, 48, 53], nine assessed cannabis-related content[22,23,38,41,42,46,47,50,51], five assessed opiate-related content[20,21,25,40,56], five assessed tobacco-related content[32,35,44,45,52], one assessed alcohol-related content[30], and a single study assessed polysubstance use-related content[29].SentimentTwenty-Three studies[21, 22, 25, 29, 30, 32, 34-37, 40-48, 50, 52, 53, 56] reported sentiment analyses of substance-related content on Twitter. Overall, substance-use sentiment on Twitter was positive, with 76.0% of all analysed content depicting substance use positively.…”

The Use and Validation of Drug Screening Technologies in Night-Time Entertainment Districts

“…Van Geest et al, 2017). For instance, the increased adoption of smartphones, social media, identification scanning, and global positioning system devices have shown to be instrumental in improving methods of collecting data, identifying and monitoring peoples substance use practices and the harms they experience, and trialling novel interventions (e.g., Byrnes et al, 2016;Cárdenas & Stormshak, 2019;Devilly, Greber et al, 2019;Kuntsche & Labhart, 2014;Merril et al, 2020;Piron & Devilly, 2022;Riordan et al, 2021;Steers et al, 2021;Wright et al, 2018). Hand-held Breath Approximated Blood Alcohol Concentration (BrAC) devices are another popular apparatus used by field-researchers to improve the overall reliability and validity of their findings (c.f.…”