Smoking Selfies: Using Instagram to Explore Young Women’s Smoking Behaviors

Abstract

Our research provides social scientists with areas of inquiry in tobacco-related health disparities in young adult women and opportunities for intervention, as Instagram may be a powerful tool for the public health surveillance of smoking behavior and social norms among young women. Social media has fundamentally changed how to engage with health-related information. Researchers increasingly turn to social media platforms for public health surveillance. Instagram currently is one of the fastest growing social networks with over 53% of young adults (aged 18-29) using the platform and young adult women comprise a significant user base. We conducted a content analysis of a sample of smoking imagery drawn from Instagram’s public Application Programming Interface (API). From August 2014 to July 2015, 18 popular tobacco- and e-cigarette-related text tags were used to collect 2.3 million image posts. Trained undergraduate coders (aged 21-29) coded 8,000 images (r = .91) by type of artifact, branding, number of persons, gender, age, ethnicity, and the presence of smoke. Approximately 71.5% of images were tobacco-relevant and informed our research. Images of cigarettes were the most popular (49%), followed by e-cigarettes (32.1%). “Selfies while smoking” was the dominant form of portrait expression, with 61.4% of images containing only one person, and of those, 65.7% contained images of women. The most common selfie was women engaged in “smoke play” (62.4%) that the viewer could interpret as “cool.” These “cool” images may counteract public health efforts to denormalize smoking, and young women are bearing the brunt of this under-the-radar tobacco advertising. Social media further normalizes tobacco use because positive images and brand messaging are easily seen and shared, and also operates as unpaid advertising on image-based platforms like Instagram. These findings portend a dangerous trend for young women in the absence of effective public health intervention strategies.

Keywords

tobacco smoking vaping e-cigarettes young adult women Instagram selfie

Introduction

Cigarette use continues to be the number one preventable disease in the world and in the United States (Bauer, Briss, Goodman, & Bowman, 2014; Centers for Disease Control and Prevention, 2012). Young adults are particularly vulnerable to tobacco use due to targeted tobacco marketing campaigns (Cortese, Lewis, & Ling, 2009; Ling & Glantz, 2002), with approximately 20% of young adults smoking cigarettes, with 18 to 25 having the highest prevalence at 41% (Richardson, Williams, Rath, Villanti, & Vallone, 2014). Although the smoking trends for young adult men show higher prevalence when compared to young adult women (Nelson et al., 2008), it is not as if young adult women have little risk of tobacco use (Li, Holahan, & Holahan, 2015) which can lead to tobacco addiction and multiple tobacco use (Lee, Hebert, Nonnemaker, & Kim, 2014; Ling & Glantz, 2002). Anti-tobacco marketing to youth such as those from the Truth Campaign, as well as anti-tobacco advertising and interventions in bars and clubs targeting young adults, helps in reducing or preventing smoking (Agaku, King, & Dube, 2014; Ling, Holmes, Jordan, Lisha, & Bibbins-Domingo, 2017). As Laestadius, Wahl, and Cho (2016) demonstrate in their research using Instagram hashtags (e.g., #ecig and #vape), the tobacco industry is already circumventing advertising bans on Instagram with corporate users comprising half of their sample of e-cigarette or vape posts, and none of the posts they found were critical of vaping tobacco use. Are we witnessing the tobacco industry using the shifting landscape of social media to advertise by influencing the creation and perpetuation of the “smoking selfie?” Who is mostly sharing these selfies, and what are in these images that are shared? What social groups are most at risk by seeing a social reconstruction of smoking re-normalization by means of a rapidly growing and minimally monitored technology that captivates a demographic most vulnerable to smoking initiation?

In this article, we provide early data on tobacco use in selfies through Instagram, and provide examples of how the selfies Instagram users share normalize tobacco use, thereby putting at risk the smoking rate declines seen over the past decades. We conclude that time is of the essence for public health professionals to begin using media technologies in their anti-tobacco counter-marketing strategies to stave off an uptick in use because, as we argue, the tobacco industry is already perfecting their marketing in social media.

Instagram represents one of the fastest growing social networks; at the time the Instagram sample was extracted, the platform had more than 400 million daily users, 90% of whom are under age 35 (Instagram, 2016; Smith, 2018); in 2014, Pew reported that 53% of young adults aged 18 to 29 used Instagram, compared to just 37% who used Twitter (Duggan, Ellison, Lampe, Lenhart, & Madden, 2015). A huge draw to the platform is that Instagram is visually based, and users can manipulate their photographs through a variety of filters. Portraits of oneself (i.e., “selfies”) are highly common on Instagram (Hu, Manikonda, & Kambhampati, 2014).

Selfies may seem insignificant or fleeting to the casual observer, but it is an integral part of the self-identity construction, normative gender performance expectations, and the subculture of young adult women. Although nearly anyone can take—and probably has taken—a self-photo, or “selfie” as it is commonly called, the dearth of research on selfies suggests that these are gendered performances done by young women (Gorey, 2014; Tifentale & Manovich, 2014). In one such study that analyzed thousands of selfies from users in five cities across the world (Bangkok, Berlin, Moscow, New York, and Sao Paulo), women were more likely to take selfies than men (range 55.2% women in Bangkok to 82% in Moscow, with New York 61.6% women), and young women in particular (23.7 estimated median age, with women having the younger average age across all cities). Normative cultural expectations of young women center on physical appearances (Strahan, Wilson, Cressman, & Buote, 2006) and help explain why nearly half of all selfies fall into the “appearance” category (Deeb-Swihart, Polack, Gilbert, & Essa, 2017).

Sociologists and social psychologists often note the importance of this type of presentation of self (Goffman, 1959), as the clothes we wear and the roles we perform help establish and perpetuate self-identity. Selfies are an even faster way in which individuals can manage these performances because they can receive almost instantaneous validation (or correction) in their gendered performances. Although selfie images are consumed quickly and individually, their impacts on the consumer are cumulative like advertising (Kilbourne, 2016) and other print media (Cortese & Ling, 2011), with a recent study of young women finding that even the most fleeting perusal of selfies had a negative impact on their self-image (Eckler, Kalyango, & Paasch, 2017). Selfies can attract followers and increase online popularity (Marwick, 2015), making users “mircocelebrities” (see Senft, 2013) through their consistent online self-presentations. In short, selfies are impactful to both the selfie-taker and the observer consuming the imagery.

We caution the reader not to interpret selfies as exclusively a young woman’s domain. The selfie is a nuanced cultural phenomenon used by all genders, documenting and exhibiting experiences from the banal to the advert of all users—from the common-folk to the celebrity (Abidin, 2016; Marwick, 2015). However, it is important to note that our research findings comport with earlier research on stealth tobacco advertising in bars and clubs that use popular people as “influencers” to use and thereby encourage product use of others. Although the tobacco industry historically built advertising campaigns on the use of female gender-normative performances in tobacco print advertising in the 1970s and 1980s (Anderson, Glantz, & Ling, 2005), in the 1990s and 2000s, the tobacco industry took their advertising to a stealthier method. The tobacco industry spent years psychographic profiling young men and women to understand their lifestyles, desires, aspirations, and gender expressions to market their tobacco products beyond traditional print advertising (Cortese et al., 2009; Cortese & Ling, 2011; Katz & Lavack, 2002; Ling & Glantz, 2002; Sepe, Ling, & Glantz, 2002). This article reifies another way the tobacco industry uses the gendered performances of young women—unknowingly or not—as tools for and targets of their stealth tobacco advertising methods, and that by doing so, it makes young adult women who use social media particularly vulnerable to these “crafty” and “cool” tobacco marketing efforts.

The recent rapid evolution of the media landscape and the widespread use of social media fundamentally altered the ways in which audiences receive health-related information (Cappella, Kim, & Albarracín, 2015; Himelboim, Smith, & Shneiderman, 2013; Napoli, 2011). Not only are people simply exposed to such information, but they also search for, create, share, and respond to content across multiple media platforms (Cappella et al., 2015). Researchers increasingly turn to the rich datasets offered by social media (e.g., Twitter) to conduct public health surveillance, tracking such topics as disease outbreaks (Culotta, 2010; Lampos & Cristianini, 2010), depression (De Choudhury, Gamon, Counts, & Horvitz, 2013), insomnia (Jamison-Powell, Linehan, Daley, Garbett, & Lawson, 2012), and overall county health (Schwartz et al., 2013). This evolving communications ecosystem has implications for substance use trends, including tobacco use; indeed, the tobacco control community has begun to recognize the importance of social media for influencing health behaviors (Allem et al., 2016; Hefler, Freeman, & Chapman, 2013).

Evidence has shown that substance use frequently is portrayed and discussed on social media, often in positive terms, and that young people easily may access and engage with these posts (Egan & Moreno, 2011; Morgan, Snelson, & Elison-Bowers, 2010; Myslín, Zhu, Chapman, & Conway, 2013; Seidenberg, Rodgers, Rees, & Connolly, 2012). This finding is meaningful when considering tobacco use because, according to social cognitive theory (Bandura, 2001; Stefanone, Lackaff, & Rosen, 2010), people—especially young people—are strongly influenced by peers or others in social media and reality shows who “look” like they could be peers (Stefanone et al., 2010). Actual or perceived behaviors of peers and/or other valued social referents (such as celebrities) have been established as key risk factors for tobacco and other substance use among adolescents and young adults (Bandura, 1971; Stefanone et al., 2010; Steinberg & Monahan, 2007; Turner, Mermelstein, & Flay, 2004), and the advertising methods of microcelebrities (Abidin, 2016; Marwick, 2015) may be an additional risk factor in the rapidly growing social media advertising landscape. Repeated exposure to images modeling tobacco use behavior by members of one’s social network may increase the perceived prevalence of tobacco use among one’s peers, serving to normalize the behavior and potentially resulting in imitation (Cialdini & Trost, 1998; Rivis & Sheeran, 2003).

Social media platforms also play a significant role in tobacco marketing; such marketing falls largely under the radar of the tobacco control community and outside the restrictions stipulated by the 1998 Master Settlement Agreement (Huang, Kornfield, & Emery, 2016; Huang, Kornfield, Szczypka, & Emery, 2014; Kostygina, Huang, & Emery, 2016; Kostygina, Tran, Shi, Kim, & Emery, 2016; National Association of Attorneys General, 1998). Advertising for tobacco products on social media offers high brand visibility (Richardson, Ganz, & Vallone, 2013), reaches a primarily young audience, and effectively circumvents the restrictive media marketing policies for tobacco. Users promote tobacco products organically on social media by incorporating them into narratives about their lives, while brands leverage consumer-generated content to extend their marketing reach when users post images in which those brands are visible (Carah & Shaul, 2016). Furthermore, celebrity presence on Instagram is quite high, and the extant studies examining celebrity substance use–related Instagram posts to date suggest that these communications have potential to reshape social norms about substance use behavior (Addiction-Treatment.com, 2014; Richardson et al., 2013; Richardson & Vallone, 2014). Tobacco control seized upon this opportunity in the Truth Initiative’s (2015) “unpaid spokesperson” television advertisement which featured images of young celebrities smoking and tagged them as unwitting marketers for the tobacco industry (Ad Age, 2014). The “unpaid spokesperson” ad first aired during the MTV Music Awards program and was part of Truth®’s “Finish It” campaign (Truth Initiative, 2015).

Given Instagram’s popularity among young adults and celebrities—high-profile and microcelebrities alike—(Khamis, Ang, & Welling, 2017; Laestadius et al., 2016; Marwick, 2015) and the platform’s potential to influence social norms related to tobacco, characterizing the extent and nature of consumer-generated image marketing of tobacco products on Instagram becomes an important research question. In this study, we undertook a content analysis of smoking imagery on Instagram and analyzed how such tobacco-related imagery is shared among users.

Methods

Data Collection

The data in this study were accessible to the public at the time of collection (April through July 2015) through the Internet, and our sample of 8,000 from 2,000,000+ images was collected through Instagram’s public Application Programming Interface (API) through the use of “tags.” Instagram users communicate through hashtags or “tags.” These tags are signposts for users to connect images around a theme and are discoverable via the Instagram platform for other users to view. For example, a user who has an affinity for clouds of smoke or vapor produced by tobacco, electronic cigarette, or vaping products might post an image of this action and tag the post as #smokeclouds. Other users can then search for the tag to view the images associated with that particular tag. It is important to note that, at the time, the API of Instagram had permitted this type of research methodology (see Highfield & Leaver, 2015), but no more. Today, Instagram is under ownership of Facebook (Marwick, 2015), and the shift in the Instagram API (Instagram, 2018a, 2018b) prevents replication of this and similar studies. Furthermore, the Terms of Service (TOS; Instagram, 2018c) prohibit researchers from using images or any other content that goes into a posted image found through hashtag searches without permission from the user, which is why we cannot include any images in this article.

We used websta.me, a web-based Instagram analytic platform connected to the Instagram API, to identify popular tobacco, electronic cigarette, and vaping tags. This search created a preliminary list of 683 tags that included brands, common terms, and slang. We narrowed this list to 18 tags to be included in our study based on research team expert consensus. Our final list of tags included the following: #cig, #cigarette, #cigarettes, #cigarillo, #cigarillos, #cigs, #ecig, #ecigs, #hookah, #nicotine, #rellos, #rillo, #rillos, #smokeclouds, #smokeselfie, #smokingselfie, #smokingtricks, and #vape. Data collection began on August 1, 2014 and finished July 31, 2015, and yielded a corpus of 2,334,271 images, of which we yielded a random sampling of 8,000 images.

Content Coding

We developed our coding classification system, instruments, and codebook based on a random sample of 1,000 images from the 8,000 sample noted above. The co-authors first coded this set to identify its relevance to tobacco or vaping. If an image contained an act or artifact of smoking or vaping, then the image could be coded for content. We developed a coding instrument and codebook based upon this set of relevant images. In cases where images were ambiguous or unclear in meaning, coders were instructed to use hashtags or text to contextualize intended meanings of the images.

The instrument and codebook included categories for objective measures: (1) artifact visible (cigarette, e-cigarette or cigalike or other vaping device, cigarillo or blunt, hookah, vaping liquid, tobacco package, cannot tell, and none), (2) person number (one, two, three or more, and none), (3) artifact likely has a brand name or logo (yes or no), smoking or vaping normally (yes or no), (4) smoke play or vape play (yes or no), (5) presence of marijuana (yes or no), and (6) image is an anti-smoking message (yes or no). Coders were trained with two training sessions and four test samples of 20 on how to determine the more subjective categories: (1) gender (all female, all male, both, and cannot tell), (2) age of primary person (probably/definitely high school age or younger, probably/definitely college age but under 30, probably/definitely older than 30, and cannot tell), and (3) ethnicity (mostly/all White, mostly/all Black, mostly/all another race or ethnicity, and cannot tell). Due to the nature of the in-app abilities for the selfie-taker to use photographic filters and vignetting that could affect the perceived skin tone, age, and gender, the subjective categories required “probability hedging” to allow coders the flexibility to self-assess with strong likelihood the strength of the claims without feeling they needed to exhibit absolute certitude.

We invited six (three males and three females) Dean’s List students aged 21 to 29 from the Anthropology and Sociology Program at Governors State University, and four students (three males, one female) agreed to the coder training for the Summer 2015 semester. To minimize differences in coding interpretations, we sought students of the similar ages of those who most frequently post selfies on social media. Coders were blinded to our hypotheses to avoid acquiescent or social desirability biases. Coders were trained in two half-day sessions with a codebook and instructions on how to determine with certitude the ages, genders, and race ethnicities (the most subjective categories of the instrument), as well as the more objective categories of content. Under the training supervision of Cortese, teams coded four test sets of 100 images each to ensure consistent intercoder reliability (ICR) for each set. Cortese met with the coders to discuss agreement and discrepancies among coded categories, and subsequently definitions were further refined by these meetings. Further refinement of the codebook continued until ICR measures from the test sets met normative standards in content analyses (Cortese et al., 2009; Lombard, Snyder-Duck, & Bracken, 2002; Riff, Lacy, & Fico, 2014).

Coders were separated into two teams and, in 500 image sets (to consistently test for ICR), each team content analyzed a random sample of 4,000 images (8,000 in total) based upon the instrument and codebook, and tested for ICR at the end of each set. Coders within each team achieved high agreement (Krippendorff’s α = .91), and we consistently measured ICR throughout the study to ensure that it exceeded a percentage agreement greater than 85%, a normative expectation for studies of this type (Cortese et al., 2009). On average, each coder team exceeded 85% for ICR on each set, with variances averaging between 80% and 90% for the subjective categories of race, age, and gender, and 85% to 100% for the objective categories of artifact visibility, branding, marijuana use, and smoke/vape play. Approximately 2,279 images were coded “not relevant” to tobacco or vaping because the tags used by the poster did not comport with its corresponding image. This left 5,721 images coded for content.

This article does not include a semiotic analysis of the imagery’s meanings, nor examples of images, due to Instagram’s TOS that limits the reproduction of images without permission from the original users. We encourage readers to view images by the tags we provide in the “Methods” section either online or through the Instagram app.

Results

Content coding of the 5,721 Instagram photographs relevant to tobacco smoking or vaping behavior revealed that combustible tobacco cigarettes were the most popular images displayed, with the most prevalent type of image posted being one of a young White woman smoking a cigarette in a conventional fashion.

Coding of demographics among the images is presented in Table 1. By far the highest proportion of smoking-related images was coded for females at 67.4% compared to 29.7% showing males, and 2.7% in which both genders were visible. A majority of the coded photographs were found to be selfies showing a single individual at 61.4%, while 4.1% were group selfies containing two or more people and 34.5% did not include any people in the image frame. The most prevalent age group was college-aged at 87.5%, followed by high school age or younger at 8.9%, and the least prevalent was the “older than 30” age group at 3.6%. More than half of the images were judged to depict people of White ethnic background at 85.7%, with 4.6% Black and 9.6% “all other” ethnicities.

Table 1.

Demographics Among Smoking-Related Photographs on Instagram.

Characteristic	No. of photos	%
Age
College	2,566	87.5
High school	261	8.9
Older than 30	106	3.6
Gender
Female	2,383	67.4
Male	1,056	29.7
Both visible	97	2.7
Ethnicity
White	2,699	85.7
Black	162	4.6
All others	338	9.7
Number
Individual selfie	3,514	61.4
Group selfie	235	4.1

Content coding also indicated high presence of smoking artifacts and branding in the images. As shown in Table 2, of the 5,721 photographs coded, 2,299 (40.2%) contained artifacts likely displaying a brand name or logo. Artifacts featured in the photographs were most likely to be cigarettes (47.2%), followed by e-cigarettes, other vaping devices, or vaping liquid at 32.1%, cigarillos at 3.6%, and hookah (2.4%).

Table 2.

Tobacco Artifacts and Brands Present in Instagram Images.

Characteristic	No. of photos	%
Branding
Artifact likely has a brand name or logo	2,299	40.2
Tobacco artifact
Cigarette	2,803	47.2
E-cigarette or other vaping device	948	16
Vaping liquid	886	14.9
Tobacco package	357	6
Hookah	140	2.4
None	590	10.3

Table 3 presents the results of coding for presence of smoke and/or smoke play in the images. Smoke or vape clouds were visible in 60% of the photographs, and 16% depicted smoke play or vape play. A very small proportion of the coded sample of photographs was judged to be presenting an anti-tobacco message (1.9%).

Table 3.

Smoke Imagery in Instagram Photographs.

Characteristic	No. of photos	%
Smoke play
Smoking or vaping	3,430	60
Smoke play or vape play	894	16
Marijuana	584	10.2
Anti-smoking message	108	1.9

Discussion

In this exploratory study, the results indicate that smoking selfies are prevalent and widely shared across the Instagram platform, which is particularly popular among youth and young adult women. The prevailing sentiment in smoking-related Instagram posts appears to be pro-use and the most common form of smoking image is one of a young White woman smoking a combustible cigarette. Public health professionals should note this gendered social media experience, given that smoking rates for young adult males are higher than for females (Jamal, 2016), but that the smoking presence on Instagram skewed toward young adult women. Greenwood, Perrin, and Duggan (2016) have reported that teenage girls share more on visually oriented platforms like Instagram than boys do, and the same holds true for female adults (Lenhart, 2015). Therefore, even though posts are more common among young women, it does not necessarily reflect higher prevalence of tobacco use. However, when selfies demonstrate smoking behavior, they are easily seen and potentially shared among young adult social networks. Young White women in particular, who are establishing their social media personas, may be inadvertently encouraging and reinforcing pro-smoking norms, with potential to influence the uptake or maintenance of smoking behavior (Rivis & Sheeran, 2003). The prevalence and glamorization of such images may serve to counteract recent public health gains toward de-normalizing tobacco use, particularly in groups who may not be considered health disparate. Tobacco companies are likely exploiting this gendered selfie norms through under-the-radar advertising that is to the tobacco industry’s benefit only.

These findings imply that Instagram can be a powerful tool for the surveillance of tobacco use. Tobacco brands also appear to be commonly shared in user-generated Instagram posts. This finding has policy implications because brand sharing in the “organically sharing” features of popular image-based social media platforms essentially represents covert, unpaid advertising for the tobacco industry that it would have little interest in curbing. The present study is consistent with other research indicating that the tobacco companies remain a step ahead of tobacco control, subverting advertising restrictions by exploiting Web 2.0 for marketing purposes (Elkin, Thomson, & Wilson, 2010; Freeman, 2012), and offers support for including Instagram and other social media in future policy considerations. Furthermore, very few images were found to promote anti-tobacco sentiments and, given Instagram’s potential (and social media in general) to reach and engage with young smokers, anti-tobacco advocates and public health professionals may be inadvertently overlooking an important opportunity to spread counter-marketing messages.

This study has certain limitations and suggests directions for future research. First, our methodology relied upon Instagram text tags to capture relevant content during data collection. There is risk that some relevant tags may have been omitted, the inclusion of which might have altered the findings. Second, and related to the above, the coding of data was primarily of image-only, and textual content such as hashtags, comments, or emojis were used as supplementary material to clarify ambiguous or unclear imagery. Therefore, the original context of many users’ posts may have been lost (see Highfield & Leaver, 2016). Third, as described in the “Methods” section, we could neither replicate this study under the current API of Instagram nor include examples of photos because of copyright restrictions placed on user content by Instagram’s TOS. Gaining future access to Instagram data presents a challenge which tobacco control must surmount to remain relevant in today’s evolving communications environment. Public policy professionals must inform policymakers on the need to re-open social media to examination, given the ubiquity of the platform to shape and transform the Zeitgeist on a whole range of social norms, ideas, and belief systems, particularly with regard to tobacco and vape use among young adults.

This study offers preliminary evidence that pro-smoking imagery is widely viewed and shared across increasingly popular image-based media platforms such as Instagram. Smoking behavior may become re-normalized as the tobacco industry gains covert and unregulated advertising through this medium. Future research should confirm these results and further explore Instagram’s potential as a surveillance tool to inform behavioral and regulatory science.

Footnotes

Acknowledgements

We would like to thank our coders, who were undergraduate students from Governors State University: Heather Glockler, Joshua Gravitt, Giancarlo Narcisi, and Matthew Prusak. We also thank our anonymous reviewers who provided helpful suggestions to make this a stronger paper.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Daniel K. Cortese

Author Biographies

Daniel K. Corte (PhD, University of Texas at Austin) is Professor of Sociology at Governors State University in University Park, Illinois. His research interests include identity construction processes, gender and sexuality, and semiotics.

Glen Szczypka (MA, DePaul University) is a Principal Research Scientist in Public Health Research and Deputy Director of the Health Media Collaboratory at NORC at the University of Chicago. His research interests include health communications, public health campaign evaluation, and social media analysis.

Sherry Emery (PhD, University of North Carolina–Chapel Hill) is a Senior Fellow in Public Health Research and Director of the Health Media Collaboratory at NORC at the University of Chicago. Her research interests include health communications, social media analysis, and public health policy.

Shuai Wang (PhD Candidate, University of Illinois at Chicago) is a Research Assistant at the University of Illinois at Chicago. His research interests include data mining, natural language processing, machine learning, and sentiment analysis.

Elizabeth Hair (PhD, MA, Texas A&M University) is a Vice President for Truth Initiative’s Schroeder Institute. Her research interests include issues related to health communication and tobacco. Currently, she oversees a longitudinal evaluation of the Truth® campaign’s effectiveness in communicating anti-smoking messages to young adults.

Donna Vallone (PhD, MPH, Columbia University) is a Chief Research Officer at Truth Initiative’s Schroeder Institute. Her research interests focus on examining the influence of media messages to reduce tobacco use, particularly among lower socioeconomic status (SES) and racial/ethnic minority groups.

References

Abidin

(2016). Aren’t these just young, rich women doing vain things online? Influencer selfies as subversive frivolity. Social Media + Society, 2. doi:2056305116641342

Ad Age. (2014, August 25). “Truth” targets Rihanna, R-Patz in new ads from “Finish It” campaign. Ad Age. Retrieved from http://creativity-online.com/work/truth-unpaid-tobacco-spokesperson/36888

Addiction-Treatment.com. (2014). Instagram intoxicated. Addiction-Treatment.com. Retrieved from http://www.addiction-treatment.com/in-depth/instagram-intoxicated/

Agaku

I. T.

King

B. A.

Dube

S. R.

(2014). Current cigarette smoking among adults—United States, 2005–2012. Morbidity and Mortality Weekly Report, 63, 29–34.

Allem

J.-P.

Escobedo

Chu

K.-H.

Soto

D. W.

Cruz

T. B.

Unger

J. B.

(2016). Campaigns and counter campaigns: Reactions on Twitter to e-cigarette education. Tobacco Control, 26, 226–229.

Anderson

S. J.

Glantz

S. A.

Ling

P. M.

(2005). Emotions for sale: Cigarette advertising and women’s psychosocial needs. Tobacco Control, 14, 127–135. doi:10.1136/tc.2004.009076

Bandura

(1971). Social learning theory. New York, NY: General Learning Press.

Bandura

(2001). Social cognitive theory of mass communication. Media Psychology, 3, 265–299.

Bauer

U. E.

Briss

P. A.

Goodman

R. A.

Bowman

B. A.

(2014). Prevention of chronic disease in the 21st century: Elimination of the leading preventable causes of premature death and disability in the USA. The Lancet, 384, 45–52.

10.

Cappella

J. N.

Kim

H. S.

Albarracín

(2015). Selection and transmission processes for information in the emerging media environment: Psychological motives and message characteristics. Media Psychology, 18, 396–424.

11.

Carah

Shaul

(2016). Brands and Instagram: Point, tap, swipe, glance. Mobile Media & Communication, 4, 69–84. doi:10.1177/2050157915598180

12.

Centers for Disease Control and Prevention. (2012). Current cigarette smoking among adults—United States, 2011. Morbidity and Mortality Weekly Report, 61, 889–894.

13.

Cialdini

R. B.

Trost

M. R.

(1998). Social influence: Social norms, conformity and compliance. In Gilbert

D. T.

Fiske

S. T.

Lindzey

(Eds.), The handbook of social psychology (4th ed., Vols. 1 and 2, pp. 151–192). New York, NY: McGraw-Hill.

14.

Cortese

D. K.

Lewis

M. J.

Ling

(2009). Tobacco industry lifestyle magazines targeted to young adults. Journal of Adolescent Health, 45, 268–280.

15.

Cortese

D. K.

Ling

P. M.

(2011). Enticing the new lad: Masculinity as a product of consumption in tobacco industry-developed lifestyle magazines. Men and Masculinities, 14, 4–30.

16.

Culotta

(2010, July 25–28). Towards detecting influenza epidemics by analyzing Twitter messages. Paper Presented at the 1st Workshop on Social Media Analytics (SOMA), Washington, DC.

17.

De Choudhury

Gamon

Counts

Horvitz

. (2013). Predicting depression via social media. ICWSM, 13, 1–10.

18.

Deeb-Swihart

Polack

Gilbert

Essa

(2017). Selfie-presentation in everyday life: A large-scale characterization of selfie contexts on Instagram. Retrieved from https://www.cc.gatech.edu/~irfan/p/2017-Deeb-Swihart-SELLCSCI.pdf

19.

Duggan

Ellison

N. B.

Lampe

Lenhart

Madden

(2015). Social media update 2014. Washington, DC: Pew Research Internet Project. Retrieved from http://www.pewinternet.org/2015/01/09/social-media-update-2014/

20.

Eckler

Kalyango

Jr. Paasch

(2017). Facebook use and negative body image among U.S. college women. Women & Health, 249, 267.

21.

Egan

K. G.

Moreno

M. A.

(2011). Alcohol references on undergraduate males’ Facebook profiles. American Journal of Men’s Health, 5, 413–420. doi:10.1177/1557988310394341

22.

Elkin

Thomson

Wilson

(2010). Connecting world youth with tobacco brands: YouTube and the internet policy vacuum on Web 2.0. Tobacco Control, 19, 361–366. doi:10.1136/tc.2010.035949

23.

Freeman

(2012). New media and tobacco control. Tobacco Control, 21, 139–144. doi:10.1136/tobaccocontrol-2011-050193

24.

Goffman

(1959). The presentation of self in everyday life. New York, NY: Anchor Books.

25.

Gorey

(2014, February 24). Major study of selfies finds major demographic differences between countries. Siliconrepublic. Retrieved from https://www.siliconrepublic.com/life/major-study-of-selfies-finds-major-demographic-differences-between-countries

26.

Greenwood

Perrin

Duggan

(2016). Social media update 2016. Washington, DC: Pew Research Internet Project. Retrieved from http://www.pewinternet.org/2016/11/11/social-media-update-2016

27.

Hefler

Freeman

Chapman

(2013). Tobacco control advocacy in the age of social media: Using Facebook, Twitter and change. Tobacco Control, 22, 210–214. doi:10.1136/tobaccocontrol-2012-050721

28.

Highfield

Leaver

(2015). A methodology for mapping Instagram hashtags. First Monday, 20. Retrieved from http://firstmonday.org/article/view/5563/4195

29.

Highfield

Leaver

(2016). Instagrammatics and digital methods: Studying visual social media, from selfies and GIFs to memes and emoji. Communication Research and Practice, 2, 47–62. doi:10.1080/22041451.2016.1155332

30.

Himelboim

Smith

Shneiderman

(2013). Tweeting apart: Applying network analysis to detect selective exposure clusters in Twitter. Communication Methods and Measures, 7, 195–223. doi:10.1080/19312458.2013.813922

31.

Manikonda

Kambhampati

(2014, June 1-4). What we Instagram: A first analysis of Instagram photo content and user types. Paper Presented at the International AAAI Conference on Web and Social Media, Ann Arbor, MI. Retrieved from https://asu.pure.elsevier.com/en/publications/what-we-instagram-a-first-analysis-of-instagram-photo-content-and

32.

Huang

Kornfield

Emery

S. L.

(2016). 100 million views of electronic cigarette YouTube videos and counting: Quantification, content evaluation, and engagement levels of videos. Journal of Medical Internet Research, 18(3), e67. doi:10.2196/jmir.4265

33.

Huang

Kornfield

Szczypka

Emery

S. L.

(2014). A cross-sectional examination of marketing of electronic cigarettes on Twitter. Tobacco Control, 233, iii26–iii30. doi:10.1136/tobaccocontrol-2014-051551

34.

Instagram. (2016). Instagram press page. Retrieved from https://instagram-press.com/

35.

Instagram. (2018a). Platform changelog. Retrieved from https://www.instagram.com/developer/changelog/

36.

Instagram. (2018b). Platform policy. Retrieved from https://www.instagram.com/about/legal/terms/api/

37.

Instagram. (2018c). Terms of Service. Retrieved from https://help.instagram.com/478745558852511

38.

Jamal

(2016). Current cigarette smoking among adults—United States, 2005-2015. MMWR: Morbidity and Mortality Weekly Report, 65, 1205–1211.

39.

Jamison-Powell

Linehan

Daley

Garbett

Lawson

(2012, May 5-10). “I can’t get no sleep”: Discussing #insomnia on Twitter. Paper Presented at the SIGCHI Conference on Human Factors in Computing Systems, Austin, TX. Retrieved from https://dl.acm.org/citation.cfm?id=2208612&bnc=1&preflayout=tabs

40.

Katz

S. K.

Lavack

A. M.

(2002). Tobacco related bar promotions: Insights from tobacco industry documents. Tobacco Control, 11, I92–I101.

41.

Khamis

Ang

Welling

(2017). Self-branding, “micro-celebrity” and the rise of social media influencers. Celebrity Studies, 8, 191–208. doi:10.1080/19392397.2016.1218292

42.

Kilbourne

(2016). Beauty and the beast of advertising. In Scott

Cayleff

Donadey

Lara

(Eds.), Women in culture: An intersectional anthology for gender and women’s studies (2nd ed., pp. 183–186). Malden, MA: Wiley.

43.

Kostygina

Huang

Emery

(2016). TrendBlendz: How Splitarillos use marijuana flavours to promote cigarillo use. Tobacco Control, 26, 235–236.

44.

Kostygina

Tran

Shi

Kim

Emery

(2016). “Sweeter than a Swisher”: Amount and themes of little cigar and cigarillo content on Twitter. Tobacco Control, 25, i75–i82.

45.

Laestadius

L. I.

Wahl

M. M.

Cho

Y. I.

(2016). #Vapelife: An exploratory study of electronic cigarette use and promotion on Instagram. Substance Use & Misuse, 51, 1669–1673.

46.

Lampos

Cristianini

(2010, June 14-16). Tracking the flu pandemic by monitoring the social web. Paper Presented at the 2nd IAPR Workshop on Cognitive Information Processing (CIP), Elba, Italy. Retrieved from https://ieeexplore.ieee.org/document/5604088/

47.

Lee

Y. O.

Hebert

C. J.

Nonnemaker

J. M.

Kim

A. E.

(2014). Multiple tobacco product use among adults in the United States: Cigarettes, cigars, electronic cigarettes, hookah, smokeless tobacco, and snus. Preventive Medicine, 62, 14–19. doi:10.1016/j.ypmed.2014.01.014

48.

Lenhart

(2015). Teens, social media & technology overview 2015. Washington, DC: Pew Research Internet Project. Retrieved from http://www.pewinternet.org/2015/04/09/teens-social-media-technology-2015

49.

Holahan

C. K.

Holahan

C. J.

(2015). Sociodemographic and psychological characteristics of very light smoking among women in emerging adulthood, national survey of drug use and health, 2011. Preventing Chronic Disease, 12, Article 140547. doi:10.5888/pcd12.140547

50.

Ling

P. M.

Glantz

S. A.

(2002). Why and how the tobacco industry sells cigarettes to young adults: Evidence from industry documents. American Journal of Public Health, 92, 908–916.

51.

Ling

P. M.

Holmes

L. M.

Jordan

J. W.

Lisha

N. E.

Bibbins-Domingo

(2017). Bars, nightclubs, and cancer prevention: New approaches to reduce young adult cigarette smoking. American Journal of Preventive Medicine, 53, S78–S85. doi:10.1016/j.amepre.2017.03.026

52.

Lombard

Snyder-Duch

Bracken

C. C.

(2002). Content analysis in mass communication: Assessment and reporting of intercoder reliability. Human Communication Research, 28, 587–604.

53.

Marwick

A. E.

(2015). Instafame: Luxury selfies in the attention economy. Public Culture, 27, 137–160. doi:10.1215/08992363-2798379

54.

Morgan

E. M.

Snelson

Elison-Bowers

P. R.

(2010). Image and video disclosure of substance use on social media websites. Computers in Human Behavior, 26, 1405–1411.

55.

Myslín

Zhu

S.-H.

Chapman

Conway

(2013). Using Twitter to examine smoking behavior and perceptions of emerging tobacco products. Journal of Medical Internet Research, 15(8), e174. doi:10.2196/jmir.2534

56.

Napoli

P. M.

(2011). Audience evolution: New technologies and the transformation of media audiences. New York, NY: Columbia University Press.

57.

National Association of Attorneys General. (1998). Master Settlement Agreement. Retrieved from http://www.naag.org/naag/about_naag/naag-center-for-tobacco-and-public-health/master-settlement-agreement/master-settlement-agreement-msa.php

58.

Nelson

D. E.

Mowery

Asman

Pederson

L. L.

O’Malley

P. M.

Malarcher

Pechacek

T. F.

(2008). Long-term trends in adolescent and young adult smoking in the United States: Metapatterns and implications. American Journal of Public Health, 98, 905–915.

59.

Richardson

Ganz

Vallone

(2013). The cigar ambassador: How Snoop Dogg uses Instagram to promote tobacco use. Tobacco Control, 23, 79–80.

60.

Richardson

Vallone

D. M.

(2014). YouTube: A promotional vehicle for little cigars and cigarillos? Tobacco Control, 23, 21–26. doi:10.1136/tobaccocontrol-2012-050562

61.

Richardson

Williams

Rath

Villanti

A. C.

Vallone

(2014). The next generation of users: Prevalence and longitudinal patterns of tobacco use among US young adults. American Journal of Public Health, 104, 1429–1436.

62.

Riff

Lacy

Fico

(2014). Analyzing media messages: Using quantitative content analysis in research. New York, NY: Routledge.

63.

Rivis

Sheeran

(2003). Descriptive norms as an additional predictor in the theory of planned behaviour: A meta-analysis. Current Psychology, 22, 218–233. doi:10.1007/s12144-003-1018-2

64.

Schwartz

H. A.

Eichstaedt

J. C.

Kern

M. L.

Dziurzynski

Agrawal

Park

G. J.

Ungar

(2013, June 28). Characterizing geographic variation in well-being using tweets. Paper Presented at the 7th International AAAI Conference on Weblogs and Social Media (ICWSM), Ann Arbor, MI. Retrieved from http://wwbp.org/papers/icwsm2013_cnty-wb.pdf

65.

Seidenberg

A. B.

Rodgers

E. J.

Rees

V. W.

Connolly

G. N.

(2012). Youth access, creation, and content of smokeless tobacco (“dip”) videos in social media. The Journal of Adolescent Health, 50, 334–338. doi:10.1016/j.jadohealth.2011.09.003

66.

Senft

T. M.

(2013). Microcelebrity and the branded self. In Hartley

Burgess

Bruns

(Eds.), A companion to new media dynamics (pp. 346–354). Malden, MA: Wiley.

67.

Sepe

Ling

P. M.

Glantz

S. A.

(2002). Smooth moves: Bar and nightclub tobacco promotions that target young adults. American Journal of Public Health, 92, 414–419. doi:10.2105/ajph.92.3.414

68.

Smith

(2018). 41 incredible Instagram statistics. Brandwatch. Retrieved from https://www.brandwatch.com/blog/37-instagram-stats-2016/

69.

Stefanone

Lackaff

Rosen

(2010). The relationship between traditional mass media and “social media”: Reality television as a model for social network site behavior. Journal of Broadcasting & Electronic Media, 54, 508–525. doi:10.1080/08838151.2010.498851

70.

Steinberg

Monahan

K. C.

(2007). Age differences in resistance to peer influence. Developmental Psychology, 43, 1531–1543. doi:10.1037/0012-1649.43.6.1531

71.

Strahan

Wilson

Cressman

Buote

(2006). Comparing to perfection: How cultural norms for appearance affect social comparisons and self-image. Body Image, 3, 211–227. doi:10.1016/j.bodyim.2006

72.

Tifentale

Manovich

(2014). Selfiecity: Exploring photography and self-fashioning in social media. Retrieved from http://manovich.net/content/04-projects/086-selfiecity-exploring/selfiecity_chapter.pdf

73.

Truth Initiative. (2015, September 29). Fall tour continues to take the “finish it” message across the United States. Truth Initiative. Retrieved from http://truthinitiative.org/news/fall-tour-continues-take-%E2%80%9Cfinish-it%E2%80%9D-message-across-united-states

74.

Turner

Mermelstein

Flay

(2004). Individual and contextual influences on adolescent smoking. Annals of the New York Academy of Sciences, 1021, 175–197. doi:10.1196/annals.1308.023