Content uploaded by Aqdas Malik
Author content
All content in this area was uploaded by Aqdas Malik on Mar 04, 2019
Content may be subject to copyright.
326
The past decade has seen rapid evolvement
of ENDS (electronic nicotine delivery sys-
tems) commonly known as electronic ciga-
rettes, e-cigarettes, or vaporizers. Although the use
of tobacco products such as combustible cigarettes,
cigars, snus, and pipes has decreased signicantly
in this same time period, there has been an uptick
in the use of e-cigarettes particularly among ado-
lescents (12-17 years old) and young adults (18-21
years old).1-4 A 2015 report by the United States
(US) Centers for Disease Control and Prevention
(CDC) and the US Food and Drug Administra-
tion (FDA) on tobacco use estimates 2.39 million
high school students and 620,000 middle school
students have used e-cigarettes at least once during
the past 30 days.4 Since 2006, the estimated usage
of e-cigarettes has soared from 3% to 20% among
6-12 graders.5
e surge in use of e-cigarettes (also referred as
vaping) has created a major public health concern
as it can create a new generation dependent on
nicotine.6 Like traditional cigarettes, e-cigarettes
also contain toxins and nicotine, a powerful ad-
dictive chemical, harmful to adolescent brain de-
velopment with negative consequences for their
attention and learning abilities.7,8 Although e-cig-
arettes are less toxic than traditional combustible
cigarettes, they could facilitate transition to other
smoking products and illicit substances. One study
carried out among 808 students of 3 public schools
in Connecticut indicates that e-cigarettes use was
associated with combustible cigarette use.9 Simi-
Aqdas Malik, Department of Information Sciences & Technology, George Mason University, Fairfax, VA. Yisheng Li, Department of Information
Sciences & Technology, George Mason University, Fairfax, VA. Habib Karbasian Department of Information Sciences & Technology, George Mason
University, Fairfax, VA. Juho Hamari, Gamication Group, Faculty of Humanities, University of Turku, Finland. Aditya Johri, Department of
Information Sciences & Technology, George Mason University, Fairfax, VA.
Correspondence Dr Malik; malik.aqdas@gmail.com
Live, Love, Juul: User and Content
Analysis of Twitter Posts about Juul
Aqdas Malik, DSc
Yisheng Li, BEng
Habib Karbasian, MSc
Juho Hamari, DEcon
Aditya Johri, PhD
Objectives: In this study, we identied patterns of communication around Juul use and users on
Twitter. Methods: Public tweets were collected from April 27, 2018 until June 27, 2018. We cate-
gorized 1008 randomly selected tweets on 4 dimensions: user type, sentiment, genre, and theme.
Results: Most tweets were through personal accounts followed by ones of the tobacco industry.
Participation by anti-tobacco campaigners, educational, and governmental entities was limited.
Posts were mostly about rst-hand use, use intentions, and personal opinions. Tweets advocating
Juul were most common; meanwhile a handful of tweets discouraged Juul use. Young women,
young men, and the tobacco industry expressed positive sentiments about Juul. Conclusions:
Twitter data are a rich source of public communication to complement surveillance of emerging
tobacco products. Youth actively and positively communicate about Juul on Twitter. Educational
content and strategies must be examined for curtailing dissemination of positive sentiments and
advocacy that normalize and promote Juul use among youth and non-smokers. We observed lim-
ited evidence supporting a claim for Juul to be a smoking cessation adjunct.
Key words: JUUL; electronic cigarettes; smoking; Twitter, social media; content analysis
Am J Health Behav.™ 2019;43(2):326-336
DOI: https://doi.org/10.5993/AJHB.43.2.9
Malik et al
Am J Health Behav.™ 2019;43(2):325-336 327 DOI: https://doi.org/10.5993/AJHB.43.2.9
larly, other studies found substantial evidence that
e-cigarette use increases the risk of transitioning
to the habitual use of combustible cigarettes.3,10-12
Moreover, studies also specify that increased expo-
sure to e-cigarettes also could facilitate transition
from e-cigarettes to consuming illicit substances
such as marijuana.13,14
Whereas smoking combustible cigarettes among
youth has declined steadily during the last decade,
use of e-cigarettes has increased.3,15 Despite numer-
ous eorts to regulate the sales and promotion of
e-cigarettes, many adolescents and young adults
perceive vaping as a healthier and safer substitute
to combustible cigarettes.16-18 Moreover, these de-
vices have extended traction due to their perceived
socially desirable characteristics that helps them in
avoiding the “smoker” tag. For instance, perception
of being cool, supporting one’s social image, and
peer validation are some salient social attributes
linked to the popularity of e-cigarettes.12,14,16,19
Studies also suggest that marketed wellness (ie,
less harmful and less addictive), easy accessibility,
price, and an assortment of avors further make e-
cigarettes more appealing to adolescents and young
adults.2,3,12,18,20
Juul, a product developed by Pax Labs and
launched in 2015, has provided a strong boost to
the vaping industry. Juul is a closed-system vaping
device that has become the largest e-cigarette brand
in the US capturing approximately 55% of the e-
cigarette market.21 e company actively markets
Juul as a smoking cessation device and harm reduc-
tion alternative through messages, such as: “Juul
is an alternative to all the distasteful elements of
smoking” and “Our mission is to eliminate ciga-
rettes by oering existing adult smokers a true al-
ternative.”22 Despite these targeted promotions to
adult smokers, measures such as age verication
for online ordering (over 21 years), and awareness
and prevention campaigns by the parent company,
Juul’s popularity has increase rapidly among ado-
lescents and young adults. In addition to Juul’s
novelty and design aesthetics, its ease of use, por-
tability, marginal smell, and ease of concealment
from parents and teachers has been attributed for
this surge.23,24 For instance, its size (L= 9.48 cm,
W= 1.51 cm) makes it slightly larger than a com-
bustible cigarette and easy to hide. e device looks
like a USB ash drive and can be recharged through
a USB port. It is virtually smokeless and the smell
of its vapor can be mistaken for fruits or candies.
e disposable nicotine pods (5% by weight), are
almost the size of a thumb and available in dier-
ent avors including mango, mint, cucumber, and
fruit medley. A single Juul pod contains about the
same amount of nicotine as supplied by a pack
of 20 combustible cigarettes.25 e high dose of
nicotine in a single pod of Juul can have negative
consequences for adolescents’ attention and learn-
ing abilities.7,8 Moreover, sleek design of the device
itself and the disposable pods being available in a
variety of fruity favors entice minors toward its use.
Flavored nicotine in Juul pods is not only attrac-
tive, but also can mislead users about the harmful-
ness of the product.26
Traditional media, campus newspapers, school
and college websites/social media accounts, and
more recently, scholarly work have initiated stress-
ing the penetration and consequences of underage
Juul use.24,27,28 An active discussion on a number
of social media platforms also highlights the popu-
larity of Juul among school and college students.
For instance, Twitter is used by students to discuss
dierent places for using Juul at school including
class rooms, bathrooms, and libraries.28 Anecdotal
evidence further suggests that it is easy for these
groups to access Juul through peers as well as pur-
chase the device and accessories through a bread-
ing underground marketplace on social platforms
such as Reddit.29 e popularity and widespread
adoption of Juul has been labeled as alarming and a
catastrophic public health crisis as it holds a strong
potential of creating a new generation of smokers
addicted to nicotine.30 To curb Juul use among stu-
dents, a principal of an Annapolis, Maryland high
school ordered removal of bathroom doors. Like-
wise, a New Jersey School System installed detec-
tors to alert school administration about Juul use
(as well as other forms of e-cigarettes) in school.
Meanwhile, a school district in eastern Pennsylva-
nia has banned USB ash drives to eect control
of Juul use by minors.30 e urgency around use of
Juul by teens has reached such proportions that the
parent company is purported to be scrutinizing the
use of Bluetooth technology to combat use among
teenagers by regularly verifying age as well as auto-
matically shutting down the devices by geo-fencing
around schools.31
“Live, Love, Juul”: User and Content Analysis of Twitter Posts about Juul
328
Twitter is one of the leading microblogging so-
cial media platforms where users can share personal
opinions or information appended by a photo, vid-
eo, or a URL. In addition to other popular social
media platforms such as Facebook and Instagram,
Twitter has been used for tobacco-related conver-
sations, not only by the public, but also by the
tobacco industry (eg, @RAI_News, @the_tma),
public health agencies (eg, @CDCTobaccoFree,
@FDATobacco), smoking advocacy groups (@
CASAAmedia, @AVABoard) and anti-tobacco
campaigners (@TobaccoPrev, @truthinitiative).32
As a rich data source, Twitter also has been used
by researchers for investigating use and percep-
tions of tobacco-related products in the general
public and among vulnerable population such as
adolescents, young adults, African Americans, and
Hispanics.16,23,33,34 Furthermore, Twitter has been
considered an important data source for addressing
emerging issues and products in public health sur-
veillance or epidemiological research.16 Social me-
dia platforms provide novel data streams that can
support public health surveillance, survey develop-
ment, educational campaigns, and policymaking.16
Tobacco-related research has started to leverage
Web-based data streams (including social media
platforms) to bridge the gap in tobacco-related re-
search – use and trends in e-cigarettes and tobacco-
control campaigns, for example.19,35-37
Presently, the popularity of a given product or
service among youth is increasingly tied to social
media use as new products and services are actively
followed, discussed, and reviewed on platforms
such as Facebook, Instagram, and Twitter. Due to
the high acceptance and acclaim of e-cigarettes,
adolescents and young adults, enthusiastically re-
search them on various social media platforms
where they familiarize themselves with new a-
vors, devices, and dierent hacks and tricks.17 As
Juul adoption has increased, data from social media
platforms can be a potential source to gather the so-
cietal as well as the environmental context in which
the device is being used, perceived, and debated
by various entities. Despite a strong prevalence of
Juul among youth, highlights in broadcast media,
and alarming concerns among school authorities,
parents, and the public health community, for-
mal research documenting relevant conversations
and communication to improve understanding
of the shared perceptions and attitudes about Juul
has been scarce. Furthermore, the current surveil-
lance systems for tobacco use that primarily rely
on traditional methods fail to capture the rapid
developments of emerging products such as Juul.17
In addition, many young adults and adolescent e-
cigarette users do not identify themselves as being
smokers.38 Considering these challenges in track-
ing and monitoring the use of emerging smoking
substitutes and accessories marketed by the tobacco
industry, the focus of the present study is to exam-
ine and understand the snapshots of dialogues on
Twitter by dierent entities surrounding Juul. e
key objective of our analysis is to identify key user
groups and conversational patterns about Juul by
conducting a content analysis of relevant tweets.
Findings from this study can inform the assessment
and design of control and prevention campaigns
for use of tobacco alternatives and the development
of future regulatory policies. Furthermore, nd-
ings from the current study pose several signicant
implications for research investigations associated
with emergent tobacco industry-inspired com-
modities, and the public health community inter-
ests in tobacco use surveillance, particularly among
minors and young adults.
METHODS
Data Collection Strategy
We used Twitter API for data collection by que-
rying the keywords juul, juuls, juuled, and juuling,
and hashtags #juul, #juuls, #juuled, and #juuling.
Using this approach, during 2 months (April 27,
2018 until June 27, 2018), we were able to retrieve
69,374 original tweets (excluding retweets and
retweets with comments).
Analytical Sample and Coding Approach
Of the total sample (N = 69,374), 2% of the
randomized tweets (N = 1387) were selected for
analysis. e coding process consisted of 2 stages.
First, drawing on prior work related to e- ciga-
rettes and smoking communication on social me-
dia platforms, a tentative coding category list was
established.33,34,39 e rst author then read 200
randomly selected tweets from the collected data;
categories identied from the literature were re-
vised to ensure that they appropriately represented
the Juul dataset. e nal codebook consisted of 4
primary dimensions: user type, sentiment, genre,
Malik et al
Am J Health Behav.™ 2019;43(2):325-336 329 DOI: https://doi.org/10.5993/AJHB.43.2.9
and theme. Concurrent with this process, exclu-
sion criteria for the analytical sample were estab-
lished (eg, non-English tweets, inaccessible tweets,
meaningless/irrelevant tweets, and tweets originat-
ing from social bots).
During the second stage, 1387 tweets were hand-
coded by one of the authors. In the coding process,
379 tweets were excluded from the analytical sam-
ple: tweets not in English (N = 52), deleted tweets
or suspended accounts (N = 177), tweets contain-
ing tags only (N = 27), tweets not directly related
to Juul or unable to comprehend (N = 30), and
spammed tweets, presumably from social bots (N =
93). e nal analytical sample consisted of 1008
tweets. To validate coding reliability, a randomly
selected sub-sample (N = 200) from the analytical
sample was coded by another author. Based on the
coding process of 2 reviewers, a substantial agree-
ment was reached across all the assessed dimensions
as calculated using Cohen’s kappa (user type: k =
0.89, sentiment: k = 0.95, genre k = 0.90, theme
k = 0.87).
Assessed Topics
User type. User type represents the tweet sender
and is determined based on the information gath-
ered through the description and picture pro-
vided in the respective user prole. Within user
type, personal accounts were sub-categorized as
young women, young men, adult women, adult
men, and unclassied individuals. In the prole
image, if a person appeared to be under 21 years
old, he/she was categorized as young, otherwise an
adult. If the prole description did not provide rel-
evant information, or the prole picture was dif-
cult to comprehend (eg, avatar), or there was no
prole picture, or there were multiple persons in
the prole picture, then it was categorized under
unclassied individual user type. As mass media,
educational institutes, anti-tobacco campaigners,
healthcare units, and government entities are vi-
tal in smoking prevention and control eorts,40
pertinent categories also were included to assess
their participation in Juul related conversations on
Twitter (Table 1).
Sentiment. is dimension refers to the stance in
a tweet whether positive, negative, or neutral (Table
2). Sentiment observed in the tweets can be helpful
in evaluating whether the conversation positively
or negatively supports Juul in general, Juul users,
Juul use, or anti-tobacco eorts. ese observations
can help us understand the prevailing social norms
about Juul.41 To determine the sentiment in a post,
we subjectively assessed whether the tweet was sup-
portive, against, or neutral towards Juul. Any tweet
dominantly expressing positive emotions, feelings,
or reactions (eg, great, excited, relaxing, enjoying,
admiration, etc) signied support for Juul in gen-
eral, and its use was categorized as having a positive
sentiment. Conversely, tweets emphasizing nega-
tive emotions, feelings, or reactions (eg, disgust,
guilt, unpleasant, ashamed, etc) were classied as
posts with negative sentiment. Finally, the tweets
that did not portray either a positive or negative
sentiment were categorized as having neutral senti-
ment. While carrying out the sentiment analysis,
we specically focused on Juul; we disregarded sen-
timent towards any other concept or topic beyond
Juul.
Genre. Tweet genre characterizes the tweet for-
mat,33 whether it’s about Juul’s rst-hand use or
intention to use, personal opinion, news story, or
related to marketing. Overall, there were 6 catego-
ries for genre classication as presented along with
the corresponding denition and a representative
tweet in Table 3.
eme. e dimension of theme refers to the
topic of the actual content presented in the tweet.
Fourteen categories were established that represent
non-exclusive topical domains such as advocacy,
youth health and safety, promotion, avors, and
smoking cessation (Table 4).
In addition to individual assessment of the afore-
mentioned dimensions, an inter-category relation-
ship among all 4 coding dimensions (use type,
sentiment, genre, and theme) was carried out (See
Appendix A; Appendices can be retrieved by con-
tacting lead author, malikaqdas@gmail.com). e
inter-category analysis enabled assessment of the
relationship of each study dimension with the oth-
er dimensions to provide additional insights about
Juul communication on Twitter.
RESULTS
As Table 1 shows, over 72% of the coded tweets
originated from personal accounts. Almost one-
fourth of the coded tweets (22.4%) were from
young women followed by young men (15.8%).
“Live, Love, Juul”: User and Content Analysis of Twitter Posts about Juul
330
Adult men and adult women represent 13.6% and
8.5% of the coded tweets respectively. Tweets from
the tobacco industry accounted for 13% of the
coded data.
Overall, a majority of the sampled coded tweets
presented a positive sentiment (57.4%). Less than
one-third of the tweets were deemed negative
(30.1%); the remaining tweets (12.5%) conveyed
a neutral sentiment (Table 2).
With respect to genre, rst-hand use or intent to
use was the most common genre at 41.9%. About
16.4% of the coded tweets were categorized as a
personal opinion; news and information genres ac-
counted for around 14% tweets in each category.
Tweets related to marketing made up 12.3% of the
coded tweets (Table 3).
In the themes category (Table 4), the largest pro-
portion of coded tweets were related to advocacy
Table 1
Distribution of Tweets by User Type
Category Denition User count
(%)
Personal account
Young women
Twitter account used by an individual for personal use
226 (22.4%)
Young men 159 (15.8%)
Adult men 137 (13.6%)
Adult women 86 (8.5%)
Unclassied 120 (11.9%)
Tobacco Industry Manufacturer, retailer, vendor or any other company involved in
sales and promotion of tobacco products or industry 131 (13.0%)
Business A non-tobacco related for-prot company 14 (1.4%)
News
A media company broadcasting TV programs (eg, BBC, Al
Jazeera), or involved in physical printing (eg, New York Times,
Forbes), or web-based publishing (eg, WebMD, Mashable)
37 (3.7%)
Government A governmental agency or organization 9 (0.9%)
Private healthcare A private hospital, care unit, agency or organization working for
public health and safety 29 (2.9%)
Educational A school, college, university or other entity involved in educating 8 (0.8%)
Non-prot/foundation
(Anti-tobacco)
A non-business entity advocating against the use of tobacco
products or smoking 26 (2.6%)
Non-prot/foundation
(Other)
A non-business entity advocating a particular social cause not
related to tobacco products or smoking 14 (1.4%)
Miscellaneous Any other account/user that cannot be classied in the above
categories 12 (1.2%)
Table 2
Distribution of Tweets by Sentiment
Category Denition Example Sentiment count (%)
Positive Tweet that supports or favors Juul,
its usage, or its user(s)
Juuling with @jonesandjills. We are always in
a party mode, so is our #juul. 579 (57.4%)
Negative Tweet that is against or condemns
Juul, its usage, or its user(s)
First shot of a #juul and tbh this is absolutely
disgusting. Never gonna try this again….ever… 303 (30.1%)
Neutral
Tweet that cannot be categorized
either under positive or negative
category
Debating juul with my buddies at our weekend
meetup in downtown #Fairfax #Virginia. 126 (12.5%)
Malik et al
Am J Health Behav.™ 2019;43(2):325-336 331 DOI: https://doi.org/10.5993/AJHB.43.2.9
(28.2%) followed by tweets with a youth health
and safety theme (16.2%). Other tweets were
about prevention (8.1%), tips and queries (8%),
promotion (7.9%), and avors (6.9%).
Inter-category analysis among all the 4 coding di-
mensions provided further insights (Appendix A).
Whereas young women, young men, and the to-
bacco industry predominantly provide positive sen-
timents towards Juul, adult women and adult men
largely oer a negative sentiment. First-hand use or
intent to use tweets and marketing-related tweets
were primarily positive; tweets coded as news and
information genres were mostly negative in senti-
ment. Tweets about advocacy, avors, promotions,
and tips and queries were largely positive; negative
sentiment tweets related to discouraging use, pre-
vention, youth health and safety, and general health
and safety themes. Tweets coded as rst-hand use or
intent to use mostly advocated Juul use, discussed
tips, tricks, and avors; most of the news and infor-
mation tweets discussed youth health and safety as
well as prevention. Finally, young women followed
by young men were the most vocal user groups ad-
vocating for Juul use; tweets originating from adult
women, adult men, and private healthcare providers
primarily voiced concerns related to youth health
and safety. With respect to the tobacco industry,
most of the tweets were promotional in nature.
As tweets of individuals (rather than organiza-
tions) comprised most of the data of the present
study, they allowed us to investigate the relation-
ships between the demographic variables of age and
sex on the sentiment of the tweets. A cross-tabula-
tion and chi-square test was performed to examine
the relation of tweeter age (younger than 21 years
vs 21 or older), tweeter sex (female or male), and
tweet sentiment (negative, neutral, or positive).
e relation between age and sentiment was statis-
tically signicant [χ² (2, N = 608) = 104,016, p <
.000]; however, the eect of sex was not statistically
signicant [χ² (2, N = 608) = 2.845, p = .241]. Ex-
amination of the cross-tabulation tables (Appendix
A) shows that the eect of the age of the tweeter is
pronounced – ie, the discrepancy between frequen-
cies and expected frequencies. However, although
the eect of tweeter sex was not statistically signi-
cant, it can be observed that in neutral and positive
tweet sentiments, the discrepancies of frequencies
and expected frequencies are inverse for men versus
women, suggesting that there may be a slight ten-
dency for men to display positive sentiment.
erefore, we further performed a tiered cross-
tabulation and chi-square test to investigate
whether there was a signicant dierence in tweet
sentiments when age and sex were combined in the
analysis. e results show that for tweets made by
Table 3
Distribution of Tweets by Genre
Category Denition Example Genre count (%)
Personal opinion Personal thought or opinion about
Juul in general or use by others
I have always thought of #Juul as a
portable charge for mobile phones (emojis) 165 (16.7%)
First-hand use
or intent to use
Reporting personal use, intent, or
interest in julling
Having a great time with my second juul
in a month 422 (41.9%)
Marketing
Marketing or advertising Juul,
vaping, or other smoking products
for commercial purposes
Run fast. Fresh delivery of #juul and
#juulpods including #mango and #mint just
landed.
124 (12.3%)
News
A news story about the latest trend,
current event, or happening to
inform and educate the online user
@News_of_the_day story: #juuling among
school students across different states of
#USA is on the rise with a potential of a
new generation of #nicotine #addicts
144 (14.3%)
Information
General information about Juul or
link to an online source (eg, forum,
blog, website, video)
A new documentary created by @AlfaS-
choolsto educate parents and educators
about #Juul and other #vaping products:
https://t.co/abcdef12345
145 (14.4%)
Miscellaneous Any other tweet that cannot be
classied in the above categories Grandpapa, hit the juul please, yooohooo 8 (0.8%)
“Live, Love, Juul”: User and Content Analysis of Twitter Posts about Juul
332
young people, there was no signicant dierence in
tweet sentiment by sex [χ² (2, N =608) = 2.598, p
= .273], indicating that tweets made by both young
men and women exhibit similar sentiment. How-
ever, for tweets made by adults, there was a statis-
tically signicant dierence in tweet sentiment by
sex [χ² (2, N =608) = 11.517, p = .003]; whereas
tweets made by adult men were less likely than
expected to display negative sentiment, and more
likely to display positive sentiment, tweets made by
adult women were more likely to display negative
sentiment and less likely to display positive senti-
ment than expected. Tweets that expressed neutral
sentiment were similarly likely by both sexes.
Table 4
Distribution of Tweets by emea
Category Denition Example Theme
count (%)
Promotional
Sale, coupon, voucher, discount,
offer, review or other activities
promoting Juul in general
10% off on #juul and #juulpods plus 15% off on all
#vapes and #juices 89 (7.9%)
Flavors Tweet discussing or promoting
different avors
I am falling in love with the #mango #juul #pod. This is
absolutely refreshing 78 (6.9%)
Features
Tweet highlighting or promoting
design related aspects of Juul (eg,
size, usb charging etc.)
Just bought a super charging kit for my #juul as a
backup ;) this thing charges quick 43 (3.8%)
Advocacy
Tweet glorifying or signifying
Juul use by attributing it as not
harmful, attractive, cool, or trendy
Fully charged and a new pod. @myfriend lets meet at
the #Lisbongrounds to rock and roll with #juul.
319
(28.2%)
Tips and
queries
Tweet mentioning tips, trick,
hacks, and support related to Juul
Is it possible to get the cheap rells for #juul as the
original ones are too expensive? And how to rell? 91 (8.0%)
Smoking
cessation
Using Juul to quit smoking or
other tobacco products
#Juuling has been great so far. It’s been 3 weeks since I
smoked a cigarette, and the credit goes to #juul 42 (3.7%)
Fun Joke, humor, meme, or funny
tweet
A new rhyme in our club: Lets juul and be cool at the
school. (emojis) 37 (3.3%)
Prevention
Preventive measures, legislations,
policy, regulations, educational
programs
Please mark your calendar for our latest webinar on
#juuling among #teenagers and #FDA. August 16, 12-2
pm Eastern hosted by Abraham Wellford and Jenny
Krispock. Register: https://t.co/Njy64645sde2
92 (8.1%)
General health
and safety
Health consequences, concerns,
dangers, threats, or risks in
general
It is quite risky to engage in #juuling as we are still
not sure about the long term side effects of #vaping
products. Stay safe
43 (3.8%)
Youth health
and safety
Health consequences, concerns,
dangers, threats, or risks among
underage/youth in schools/col-
leges
Just met a panel of principals from @Mills_county_
schools and all of them synonymously agreed that #juul
and #vaporizer is likely to cause health epidemic among
our new generation.
183
(16.2%)
Illicit substance
use
Mention of an illicit substance use
with Juul
My sister who is still 16 recently started using #CBD
with her #juul. Is this dangerous? 17 (1.5%)
Discouraging
use
Tweet discouraging or rejecting
the use and adoption of Juul
#Juul is one of the most hated, unattractive, and
dangerous piece of technology I have ever seen. Stay
away from #juuling
46 (4.1%)
Research Research / scientic information /
publication / statistics
The latest results from our survey reveal that #juuling
among youth has increased gradually since its launch in
2015. https://t.co/ee9434dfgh6
13 (1.1%)
Miscellaneous Tweet that cannot be classied in
any of the above categories
I was sitting at the beach and heard guys playing volley
and having juul. 40 (3.5%)
Note.
a: Non-exclusive category
Malik et al
Am J Health Behav.™ 2019;43(2):325-336 333 DOI: https://doi.org/10.5993/AJHB.43.2.9
DISCUSSION
Our study extends current understanding about
Juul by examining communication regarding use,
attitudes, and acceptability. Adopting the manual
coding approach, thorough insights from our re-
sults complement other work that deployed au-
tomated methods to understand the ongoing
communication about Juul on Twitter.27,28 By le-
veraging data from one of the popular social media
platforms, ie, Twitter, insightful observations about
various user groups, public sentiments, and conver-
sational themes are identied and expanded. Our
ndings will further inform e-cigarette surveillance,
regulations, and future research. Juul’s proliferation
among youth in the US has spawned a lively debate
among citizens, the media, and the public health
community.
Most tweets in our dataset communicated rst-
hand use or intent to use as well as marketing by
conveying a positive sentiment towards Juul. ese
tweets predominantly originate from youth, and
address aliation/advocacy or avors (pods), and
inquire about Juul’s features and relevant aspects.
Similar observations citing communication by
young students about Juul pods, various forms of
inquires about Juul (tips and tricks), as well as its
discrete features that may facilitate its use at school/
college also have been by others.28 It is also likely
that the positive perceptions, social image building,
and validation often depicted in Juul-related tweets
might inuence non-smokers as well as youngsters
(who are more prone to peer inuence) to experi-
ment the device.
We also observed an active participation by the
tobacco industry through its positive portrayal of
Juul. Most of the entities in this category are re-
tailers involved in marketing Juul, avored pods
(and related accessories), as well as bestowing it as a
smoking cessation device. Despite e- cigarettes be-
ing promoted and asserted as a smoking cessation
device by the tobacco industry,6 only a few indi-
viduals tweeted about Juul as a smoking alternative,
thereby exhibiting only week evidence in support
of this claim. ese ndings also echo work that
found a few Twitter users communicating about
using Juul to quit smoking.28 Further investigation
of this argument frequently used by the tobacco
industry is required, as tweets from the tobacco
industry have strong potential for reaching a large
audience, including vulnerable populations such as
adolescents, as these groups are the most active us-
ers of social media platforms, including Twitter.42
Studies also show that e-cigarette promotions can
give rise to interest among youth and non-smokers
in experimenting with the product.43,44 Given that
Juul, and the tobacco industry overall, is utilizing
online platforms (as well as oine channels) for
promoting their products with few restraints,45,46
marketing regulations might be judiciously consid-
ered to mitigate the interest and reach among youth
and non-smokers.
e overall positive sentiment and patronage
with respect to Juul-related tweets suggests that a
dominant proportion of the content on Twitter in-
tentionally normalizes and promotes its use. Our
ndings conrm the notion that social media, es-
pecially Twitter, has emerged as a platform for in-
dividuals to express their attitudes and perceptions
about various trends and topics. Social media are
communal conversational spaces that provide af-
nity groups, such as people interested in Juul, an
opportunity to share their experiences. ese nd-
ings align with prior literature suggesting most of
the tweets related to use of dierent tobacco prod-
ucts (eg, e- cigarettes, hookah) normalize or make
smoking appear common by portraying positive
experiences with smoking.19,33,34 Similar to the lter
bubbles that form around political issues, by inter-
acting with those who have positive experiences or
attitudes related to Juul, other individuals can gain
self-armation.
On the other hand, the negative sentiment (less
than one-third of the coded tweets) is expressed
mostly by adults who are likely to be parents, teach-
ers, and other concerned individuals. ese tweets
communicate information and news stories con-
cerning the threats and relevant implications for
youth health, preventive measures, and discourag-
ing use. Consistent with prior ndings highlight-
ing the addictive characteristics associated with
Juul use,27 we also observed a few disclosures that
depict frequent cravings and diculties in quitting
Juul among users. Given that Juul is designed as
a closed-system device, that is, being non-modi-
able,27 a small number of tweets also mention usage
of illicit substances such as marijuana and cannabi-
diol in Juul pods.
We also observed an active role of media com-
“Live, Love, Juul”: User and Content Analysis of Twitter Posts about Juul
334
panies (predominantly online print media) in pre-
senting stories concerning youth/general health
and safety as well as various preventive measures.
is nding shadows prior literature highlighting
the signicance and reach of media outlets on Twit-
ter for dierent campaigns.47,48 In contrast, anti-
tobacco campaigners, educational institutions, and
governmental agencies have low participation in
the observed data. Also consistent with prior stud-
ies,27,28,33 despite the concerns expressed by public
health authorities and educational institutions in
print media regarding Juul use among youth, analy-
sis of ongoing discourse on Twitter fails to reect
those apprehensions. e overall positive inclina-
tion of Twitter posts related to Juul advocates a
stronger need for educational content and strategies
that not only can reduce the misconceptions about
Juul, but also support curtailment of the prevalent
positive social norms and acceptance of the prod-
uct. Tobacco control programs engaged in risk
communication may devise mechanisms to circum-
vent positive sentiments by posting messages and
information amplifying negative sentiments. Fur-
thermore, these messages can be targeted to the in-
dividuals involved in these positive sentiments (eg,
adolescents and young adults). Social media plat-
forms especially Twitter, Facebook, and Instagram
oer a strong potential of reaching certain vulner-
able and dicult to access communities (eg, youth,
African Americans, and Latinos) and can deliver
health-related information and communication
about the risks associated with tobacco use.49 One
study has shown that interventions through dier-
ent social media platforms have potential for getting
the general public to consider smoking cessation.50
Finally, current surveillance systems on tobacco use
should rigorously build upon the existing investiga-
tions that employ novel data streams from various
social media platforms for tracking and monitoring
behavior of users of emerging tobacco products in
the fast-paced tobacco market.19,28,35,51-53
Limitations and Future Research
First, we used a limited number of hashtags
and keywords (juul, juuls, juuled, juuling) for 2
months. Even though these are the most frequently
used terms, observing a more exhaustive list of Juul-
related terms over an extended period of time could
have provided a further nuanced analysis of users,
sentiments, and themes. Although within 2 months
we were able to retrieve a sizable analytical sample,
other work on Juul communication over Twitter
collected data for a slightly longer period.27,28 e
larger timeframe can be attributed to the automated
classication adopted by these studies that usually
require a much larger analytical sample.
Second, for the content analysis, we did not con-
duct a systematic analysis and categorization of
image, video, or URL appended to the tweet. Fur-
thermore, the current study did not account for the
reach and popularity indices such as the number of
retweets, number of favorites, and number of fol-
lowers, etc. A detailed analysis of these indices could
have provided additional insights. Finally, as the
current study specically focused on Twitter data,
the results might not be generalizable to other social
media platforms due to diverse forms of commu-
nication and demographics. A quick search of Juul
on dierent social media outlets reveals an active
debate on YouTube, Instagram, and Reddit. Future
research might delve into these platforms to supple-
ment our ndings.
Another useful avenue for future research might
be a comparison of Juul’s uptake among youth with
that of traditional cigarette smoking in the past
when smoking came to be associated with being
“cool.” We nd evidence from our data that Juul’s
image as being a “cool” activity stems from the fact
that it is seen as not smoking, because smoking cig-
arettes is now perceived as an uncool activity that
causes harm; Juul, and vaping more broadly, has
been appropriated as activity in opposition to smok-
ing. Yet, there is a large component of peer pressure
and peer endorsement attached to it, similar to the
uptake of smoking in past. Social media have just
become a bigger arena in which these dynamics are
playing out. Finally, the extent of Juul usage among
cigarette smokers as a harm reduction (or smoking
cessation) alternative predominantly among adult
smokers needs to be investigated further.
Conclusion
Social media can be a potential source for seeking
public health information. e data streams origi-
nating from these channels can help in exploring
and understanding user attitudes, adoption pat-
terns, and public sentiments towards emerging and
critical health-related phenomena. Data from plat-
Malik et al
Am J Health Behav.™ 2019;43(2):325-336 335 DOI: https://doi.org/10.5993/AJHB.43.2.9
forms such as Twitter can help us understand real-
time public opinions and sentiments by listening
to what people are commenting about and in their
own words. Social media data streams can support
tobacco surveillance and guide social media cam-
paigns as well as inform policies and strategies for
tobacco control by numerous entities. Despite a few
limitations of the current study, we demonstrated
using data from a popular social media platform, ie,
Twitter to understand the pulse of public opinions
about a prevailing phenomenon (Juul) particularly
among youth. e current investigation informs
and supports policymakers, the public health com-
munity, anti-tobacco campaigners, and educators in
their eorts to understand users’ attitudes towards
Juul and counteract its growing reach, popularity,
and acceptance.
Human Subjects Statement
e current research exclusively focused on freely
and publically available social media data. In accor-
dance with George Mason’s University IRB policy
such public information does not meet the regula-
tory requirements for Human Subjects Research.
Conict of Interest Disclosure Statement
ere is no conict of interest for any of the
authors.
Acknowledgements
e National Science Foundation (NSF -
1707837) provided support for this research.
References
1. Choi K, Grana R, Bernat D. Electronic nicotine deliv-
ery systems and acceptability of adult cigarette smoking
among Florida youth: renormalization of smoking? J Ado-
lesc Health. 2017;60(5):592-598.
2. Harrell MB, Weaver SR, Loukas A, et al. Flavored e-cig-
arette use: characterizing youth, young adult, and adult
users. Prev Med Rep. 2017;5:33-40.
3. Modesto-Lowe V, Alvarado C. E-cigs... are they
cool? talking to teens about e-cigarettes. Clin Pediatr.
2017;56(10):947-952.
4. Singh T, Arrazola RA, Corey CG, et al. Tobacco use among
middle and high school students – United States. MMWR
Morb Mortal Wkly Rep. 2016;65(14):361-367.
5. Hammig B, Daniel-Dobbs P, Blunt-Vinti H. Electronic
cigarette initiation among minority youth in the United
States. Am J Drug Alcohol Abuse. 2017;43(3):306-310.
6. Grana RA, Ling PM. “Smoking revolution”: a content
analysis of electronic cigarette retail websites. Am J Prev
Med. 2014;46(4):395-403.
7. England LJ, Bunnell RE, Pechacek TF, et al. Nicotine and
the developing human: a neglected element in the elec-
tronic cigarette debate. Am J Prev Med. 2015;49(2):286-
293.
8. US Department of Health and Human Services (USD-
HHS). E-Cigarette Use among Youth and Young Adults. A
Report of the Surgeon General. Atlanta, GA: USDHHS;
2016. Available at: https://e-cigarettes.surgeongeneral.
gov/documents/2016_SGR_Full_Report_non-508.pdf.
Accessed October 29, 2018.
9. Bold KW, Kong G, Camenga DR, et al. Trajectories of
e-cigarette and conventional cigarette use among youth.
Pediatrics. 2017;141(1):e20171832.
10. Cardenas VM, Evans VL, Balamurugan A, et al. Use of
electronic nicotine delivery systems and recent initia-
tion of smoking among US youth. Int J Public Health.
2016;61(2):237-241.
11. National Academies of Sciences, Engineering, and Medi-
cine. Public Health Consequences of E-cigarettes. National
Academies Press 2018. Available at: https://www.nap.
edu/catalog/24952/public-health-consequences-of-e-ciga-
rettes. Accessed October 29, 2018.
12. Schneider S, Diehl K. Vaping as a catalyst for smoking? An
initial model on the initiation of electronic cigarette use
and the transition to tobacco smoking among adolescents.
Nicotine Tob Res. 2016;18(5):647-653.
13. Giroud C, de Cesare M, Berthet A, et al. E-cigarettes: a
review of new trends in cannabis use. Int J Environ Res
Public Health. 2015;12(8):9988-10008.
14. Unger JB, Soto DW, Leventhal A. E-cigarette use and
subsequent cigarette and marijuana use among Hispanic
young adults. Drug Alcohol Depend. 2016;163:261-264.
15. McMillen RC, Gottlieb MA, Shaefer RMW, et al. Trends
in electronic cigarette use among US adults: use is increas-
ing in both smokers and nonsmokers. Nicotine Tob Res.
2015;17(10):1195-1202.
16. Ayers JW, Leas EC, Allem JP, et al. Why do people use
electronic nicotine delivery systems (electronic cigarettes)?
A content analysis of Twitter, 2012-2015. PLoS One.
2017;12(3):e170702.
17. Hilton S, Weishaar H, Sweeting H, et al. E-cigarettes, a
safer alternative for teenagers? A UK focus group study of
teenagers’ views. BMJ Open. 2016;6(11):e013271.
18. Cheney MK, Gowin M, Wann TF. Electronic cigarette
use in straight-to-work young adults. Am J Health Behav.
2016;40(2):268-279.
19. Romito LM, Hurwich RA, Eckert GJ. A snapshot of the
depiction of electronic cigarettes in YouTube videos. Am J
Health Behav. 2015;39(6):823-831.
20. de Andrade M, Angus K, Hastings G. Teenage percep-
tions of electronic cigarettes in Scottish tobacco-education
school interventions: co-production and innovative en-
gagement through a pop-up radio project. Perspect Public
Health. 2016;136(5):288-293.
21. Weisman S. What’s the hype? Juul electronic cigarettes
popularity with youth & young adults. Public Health
Law Center. 2018. Available at: http://www.publi-
chealthlawcenter.org/sites/default/.../Juul-Webinar-Slides-
Apr262018.pdf. Accessed October 15, 2018.
22. JUUL (@JUULvapor) | Twitter. Available at: https://twit-
“Live, Love, Juul”: User and Content Analysis of Twitter Posts about Juul
336
ter.com/JUULvapor. Accessed October 15, 2018.
23. Huang J, Duan Z, Kwok J, et al. Vaping versus JUULing:
how the extraordinary growth and marketing of JUUL
transformed the US retail e-cigarette market. Tob Control.
2018 May 31. pii: tobaccocontrol-2018-054382. doi:
10.1136/tobaccocontrol-2018-054382. [Epub ahead of
print]
24. Willett JG, Bennett M, Hair EC, et al. Recognition, use
and perceptions of Juul among youth and young adults.
Tob Control. 2019;28(1):115-116.
25. JUULpods and E-Liquid FAQs - JUUL Support. JUUL-
pod Basics. Available at: https://support.juul.com/home/
learn/faqs/juulpod-basics. Accessed October 26, 2018.
26. Truth Initiative. Juul e-cigarette craze highlights the dan-
gers of avored tobacco. 2018. Available at: https://truthi-
nitiative.org/news/juul-e-cigarette-craze-highlights-why-
avored-tobacco-products-are-so-dangerous. Accessed
October 15, 2018.
27. Kavuluru R, Han S, Hahn EJ. On the popularity of the
USB ash drive-shaped electronic cigarette Juul. Tob Con-
trol. 2019;28(1):110-112.
28. Allem J-P, Dharmapuri L, Unger JB, Cruz TB. Character-
izing JUUL-related posts on Twitter. Drug Alcohol Depend.
2018;190:1-5.
29. Ross B. Juul Vaporizers: nicotine alternative hits campus.
e Daily Evergreen. 2018. Available at: https://dailyever-
green.com/24025/mint/juul-vaporizers-nicotine-alterna-
tive-hits-campus/. Accessed October 15, 2018.
30. Bui L. Juuling: If you don’t know what it is, ask your kids.
Washington Post. 2018. Available at: https://www.wash-
ingtonpost.com/local/public-safety/juuling-if-you-dont-
know-what-it-is-ask-your-kids/2018/05/09/37e2f026-
4d65-11e8-84a0-458a1aa9ac0a_story.html?utm_
term=.0fc2a53b4ed5. Accessed October 25, 2018.
31. Locklear M. Juul wants to use Bluetooth e-cigs to lock
out teen smokers. Engadget. Available at: https://www.
engadget.com/2018/08/02/juul-bluetooth-e-cigs-lock-
out-teen-smokers/. Published 2018. Accessed October 15,
2018.
32. Allem J-P, Escobedo P, Chu K-H, et al. Campaigns and
counter campaigns: reactions on Twitter to e-cigarette
education. Tob Control. 2017;26(2):226-229.
33. Cole-Lewis H, Pugatch J, Sanders A, et al. Social listening:
a content analysis of e-cigarette discussions on Twitter. J
Med Internet Res. 2015;17(10):e243.
34. Krauss MJ, Sowles SJ, Moreno M, et al. Hookah-relat-
ed twitter chatter: a content analysis. Prev Chronic Dis.
2015;12:e121.
35. Chu K-H, Allem J-P, Cruz TB, Unger JB. Vaping on Insta-
gram: cloud chasing, hand checks and product placement.
Tob Control. 2017;26(5):575-578.
36. Allem J-P, Chu K-H, Cruz TB, Unger JB. Waterpipe pro-
motion and use on Instagram:# hookah. Nicotine Tob Res.
2017;19(10):1248-1252.
37. Ayers JW, Althouse BM, Allem J-P, et al. Revisiting the rise
of electronic nicotine delivery systems using search query
surveillance. Am J Prev Med. 2016;50(6):e173-e181.
38. Agaku I, Odani S, Vardavas C, Ne L. Self-Identied to-
bacco use and harm perceptions among US youth. Pediat-
rics. 2018;141(4):e20173523.
39. van der Tempel J, Noormohamed A, Schwartz R, et al.
Vape, quit, tweet? Electronic cigarettes and smoking cessa-
tion. Int J Public Health. 2016;61(2):249-256.
40. Lantz PM, Jacobson PD, Warner KE, et al. Investing in
youth tobacco control: a review of smoking prevention
and control strategies. Tob Control. 2000;9(1):47-63.
41. Lienemann BA, Unger JB, Cruz TB, Chu KH. Methods
for coding tobacco-related twitter data: a systematic re-
view. J Med Internet Res. 2017;19(3):e91.
42. Smith A, Anderson M. Social media use in 2018. Pew In-
ternet & American Life Project. 2018. Available at: http://
www.pewinternet.org/2018/03/01/social-media-use-
in-2018/. Accessed October 15, 2018.
43. Pepper JK, Emery SL, Ribisl KM, et al. Eects of adver-
tisements on smokers’ interest in trying e-cigarettes: the
roles of product comparison and visual cues. Tob Control.
2014;23(3):31-36.
44. Fischer F, Kraemer A. Secondhand smoke exposure at
home among middle and high school students in the
United States – does the type of tobacco product matter?
BMC Public Health. 2017;17(1):98.
45. Vincent D, Potts J, Durbin J, et al. Adolescent use
of electronic nicotine delivery systems. Nurse Pract.
2018;43(3):17-21.
46. Tobacco Prevention and Control Branch. TPCB: E-Cig-
arettes. Available at: https://www.tobaccopreventionand-
control.ncdhhs.gov/ecigs/. Accessed October 15, 2018.
47. Johri A, Karbasian H, Malik A, et al. How diverse users and
activities trigger connective action via social media: lessons
from the twitter hashtag campaign # Ilooklikeanengineer.
In Proceedings of the 51st Hawaii International Confer-
ence on System Sciences. 2018. Available at: http://hdl.
handle.net/10125/50161. Accessed October 29, 2018.
48. Malik A, Johri A, Handa R, et al. # EngineersWeek:
broadening our understanding of community engagement
through analysis of Twitter use during the national engi-
neers week. Proceedings of 125th ASEE Annual Confer-
ence, Salt Lake City, UT;. 2018. Available at: https://par.
nsf.gov/biblio/10066212. Accessed October 29, 2018.
49. Pechmann C, Zhao G, Goldberg ME, Reibling ET. What
to convey in antismoking advertisements for adolescents:
the use of protection motivation theory to identify eec-
tive message themes. J Mark. 2003;67(2):1-18.
50. Naslund JA, Kim SJ, Aschbrenner KA, et al. Systematic re-
view of social media interventions for smoking cessation.
Addict Behav. 2017;73:81-93.
51. Sears CG, Walker KL, Hart JL, et al. Clean, cheap, con-
venient: promotion of electronic cigarettes on YouTube.
Tob Prev Cessat. 2017 Apr;3. pii: 10. doi: 10.18332/
tpc/69393. Epub 2017 Apr 7.
52. Guidry J, Jin Y, Haddad L, et al. How health risks are pin-
pointed (or not) on social media: the portrayal of waterpipe
smoking on Pinterest. Health Commun. 2016;31(6):659-
667.
53. Ben Taleb Z, Laestadius LI, Asfar T, et al. # Hookahlife:
the rise of waterpipe promotion on Instagram. Health
Educ Behav. 2018 Jun 1:1090198118779131. doi:
10.1177/1090198118779131. [Epub ahead of print]
