An Exploratory Study of Adolescents' Social Media Sharing of Marijuana-Related Content

Abstract

Adolescents often post content related to risk behaviors online, and social media posting related to substance use has been found to be associated with intentions to use substances. However, there is a dearth of research about what may motivate an adolescent to share marijuana-related content. This study aimed to examine the types of content related to marijuana that adolescents are posting on social media and what may influence such sharing. We conducted an online survey in Washington state (N = 350)—a state in which recreational marijuana use is legal for persons at age 21 years and older—and examined adolescents' social media sharing of marijuana-related content. Thirty-one percent of adolescents in our study reported sharing marijuana-related content on social media, with 24 percent of adolescent participants sharing marijuana-related memes. Peer marijuana use and perceived parents' approval of marijuana were positively associated with the likelihood of posting marijuana-related content on social media. Increased perceived parental monitoring was negatively associated with posting marijuana-related content on social media; however, increased parental media monitoring was not. Our research highlights the factors that may influence adolescents' decisions to share marijuana-related content on social media. Internal factors, such as personal beliefs, were not associated with sharing such content in the final model, whereas external factors, such as perceived peer use and parent approval, were positively associated. Our results have implications for communicators who may want to address adolescents' norm perceptions, and parents who may want to informally monitor their children instead of monitoring their children's social media.

Introduction

Young people may be exposed to a plethora of marijuana-related content on social media,^1–3 and such exposure may impact marijuana use.⁴ This is disconcerting as the American Academy of Pediatrics⁵ opposed the use of marijuana among youth because of potential negative health effects and effects on brain development. Little work, however, has examined adolescents' creation and distribution of marijuana-related content on social media. Recent studies demonstrated youth and young adults' active engagement in displaying risky behaviors,^6–9 including marijuana use,^6,7 on social media, and these investigations highlighted a shared concern with the normalization of risky behaviors among young people. As research has often focused on young adults, adolescents' posting behaviors are less studied. In addition, the motives behind such sharing are still unclear. As young people use social media for a variety of reasons, including to present themselves to others,^8,9 it is important to assess the risk-related messages youth display on social media and what may be associated with such sharing.

The theory of planned behavior posits that attitudes, subjective norms, and perceived behavioral control influence intentions to engage in a behavior and, in turn, intentions influence behavior.¹⁰ This theory has been applied to predicting online content creation and sharing¹¹ and marijuana intentions and use among U.S. adolescents.¹² Therefore, this study focused on the potential impact of personal outcome beliefs, which are proximal to the adolescent, and perceived norms related to peers and parents, which are more distal to the adolescent.

Parental media monitoring can also influence how adolescents spend their time online and their behaviors.^13,14 Families connected on social media have a stronger connection in real life predictive of prosocial behavior and negatively associated with delinquency.¹⁵ Parental knowledge, such as knowing their children's whereabouts, is related to adolescents' intentions to use marijuana and marijuana use.¹² In addition, adolescents who perceive their parents are not monitoring them are more likely to participate in risky health behaviors, including using marijuana.¹⁶ Peers are also influential in adolescents' marijuana use.⁴ Therefore, we posed the following:

RQ: What are adolescents posting on social media related to marijuana, and where are they sharing such content?

Positive outcome beliefs about marijuana use (H1a), perceived peer marijuana use (H2), perceived friends' approval about marijuana use (H3), and perceived parents' approval about marijuana use (H4) will be positively associated with sharing social media posts about marijuana.

Negative outcome beliefs about marijuana use (H1b), adolescents' perceived parental monitoring (H5), and adolescents' perceived parental media monitoring (H6) will be negatively associated with sharing social media posts about marijuana.

Methods

Participants and procedures

We recruited 350 adolescents (13–17 years old) residing in Washington state from an online panel in 2018. Qualtrics provided compensation in the format standard to their panel studies. We obtained parents' informed consent and adolescents' assent before participation. The authors' institutional review board approved the study.

Participants were between 13 and 17 years old (M = 15.08, SD = 1.44). About half (51.7 percent, n = 181) identified as male gender, and 1.4 percent (n = 5) identified their gender as “other.” As a proxy for family income, we asked if participants received free or reduced-price breakfast or lunch at school, and 44 percent (n = 154) reported they did. About 73 percent of participants identified as White (n = 254). Participants' social media use is given in Table 1.

Table 1.

Descriptive Statistics for Participants' Social Media Use (n = 350)

Variable and questions items	Frequency	Percentage
Social media use
Never	15	4.3
Monthly	10	2.9
Weekly	38	10.9
Once a day	56	16
Multiple times a day	231	66
Most used social media (based on daily use)
YouTube	289	82.5
Instagram	208	69.5
Facebook	207	69.1
Most used social media for marijuana-related posting
Facebook	78	22.3
Snapchat	68	19.4
Instagram	61	17.4
Ever posted marijuana related content on social media^a
Yes	108	30.9
No	242	69.1
Ever posted a picture of someone smoking marijuana on a social media site
Yes	45	12.9
No	305	87.1
Ever posted a marijuana related meme on a social media site?
Yes	84	24
No	266	76
Ever posted a video of someone smoking marijuana on a social media site
Yes	39	11.1
No	311	88.9

This variable was recoded as “never = 0, and monthly to daily posting = 1.” When recoded as “never to monthly = 0, and weekly to daily = 1,” then 18.3 percent (n = 64) participants posted marijuana-related content.

Measures

Sharing of social media posts about marijuana was assessed by asking, “have you ever posted anything related to marijuana on the following social media sites” on a 7-point scale (1 = never, 7 = multiple times a day). Eight social media sites were listed. We recoded this construct as a dichotomous variable (0 = never, 1 = have posted marijuana-related messages on any of the listed social media, frequency ranging from monthly to multiple times a day).

Positive outcome beliefs (α = 0.78) and negative outcome beliefs (α = 0.91) were each measured by three items and nine items respectively, adapted from a previous study¹⁷ on a 5-point scale that asked how likely outcomes were if participants used marijuana nearly every month for the next 12 months (e.g., “be like the coolest kids,” “become forgetful”).

Peer marijuana use was measured by three questions adapted from a previous study¹⁸ on a 4-point scale (1 = none, 2 = almost none, 3 = less than half, 4 = more than half) that asked how many of their best friends, longest time friends, and friends they have associated with have smoked marijuana (α = 0.95).

Perceived friends' approval was measured by one question adapted from a previous study¹² on a 5-point scale (1 = strongly disapprove, 5 = strongly approve) that asked whether their close friends would approve if they used marijuana nearly every month for the next 12 months. Perceived parents' approval was measured by one question adapted from a previous study¹² but used the term parents/caregivers in place of friends.

Parental monitoring¹⁹ (e.g., “my parents know where I am in my spare time,” α = 0.95) and parental media monitoring²⁰ [e.g., “my parent(s) have checked websites I visit,” α = 0.79] were both measured by five questions adapted from previous research on 5-point scales (1 = strongly disagree, 5 = strongly agree).

Control variables included age (years), gender (male, female, and other), and receiving a free or reduced-price meal at school (yes, no), as a proxy for family income. For the regression analysis, identifying as male was coded as 1, and all other participants were coded as 0, as research indicates men may be more likely to use drugs.²¹

Analysis

We conducted analyses using SPSS Statistics 20. Descriptive statistics were conducted to show participants' general social media use and creation of marijuana-related social media posts. We conducted logistic regression to explore how variables influenced adolescents' marijuana-related social media posting. We entered control variables and outcome beliefs into the first logistic regression model. Peer and parent-related variables were entered into the second model. Two-way interactions between peer-related variables and parent-related variables were tested in an alternative model but were not significant, thus were omitted from the report. The sample size met the recommendation of a simulation study on the sample size requirement of logistical regression.²²

Results

As given in Table 1, 66 percent of adolescents reported using social media multiple times a day. Descriptive results and correlations among key variables are reported in Tables 2 and 3, respectively. Based on the correlations, multicollinearity was not an issue for the logistic regression analysis. All predictors, except for age and parental media monitoring, were significantly correlated with marijuana-related social media posting.

Table 2.

Descriptive Statistics

Variable	M	SD	Range
Age	15.08	1.44	13–17
Positive outcomes beliefs	2.70	1.02	1–5
Negative outcomes beliefs	3.69	.91	1–5
Perceived friends' approval	2.06	1.20	1–5
Perceived parents' approval	1.53	1.04	1–5
Peer use	2.26	1.00	1–4
Parental monitoring	4.41	.67	1–5
Parental media monitoring	3.69	1.05	1–5

M, mean; SD, standard deviation.

Table 3.

Pearson and Point-Biserial Correlations Among Key Variables

	1	2	3	4	5	6	7	8	9	10
1. Ever posted marijuana message	1
2. Gender	−0.14^**	1
3. Age	0.04	0.08	1
4. Receiving free meal at school	−0.14^**	0.14^**	0.12^*	1
5. Positive outcomes beliefs	0.29^***	−0.09	0.03	−0.14^*	1
6. Negative outcomes beliefs	−0.22^***	0.03	−0.19^***	0.18^***	−0.31^***	1
7. Perceived friends' approval	0.36^***	0.06	0.11^*	0.25^***	0.45^***	−0.53^***	1
8. Peer use	0.36^***	−0.03	0.25^***	−0.19^***	0.42^***	−0.40^***	0.56^***	1
9. Perceived friends' approval	0.38^***	0.08	0.02	0.22^***	0.31^***	−0.41^***	0.63^***	0.34^***
10. Parental monitoring	−0.26^***	0.06	−0.04	0.03	−0.20^***	0.26^***	−0.23^***	−0.17^**	−0.24^***	1
11. Parental media monitoring	−0.031	0.06	−0.1	−0.1	−0.01	0.06	0.01	0.04	0.03	0.13^*

Point-Biserial correlation was calculated to measure the relationship between dichotomous variable and continuous variables.

p < 0.05, ^**p < 0.01, ^***p < 0.001.

The research question sought to examine patterns of adolescents' marijuana-related social media posting. About 31 percent of adolescents in this study reported posting marijuana-related messages on social media sites, and the top three most used social media sites for marijuana posts were Facebook, Snapchat, and Instagram (Table 1). The most common formats of marijuana-related posts were memes (24 percent), pictures (12.9 percent), and videos (11.1 percent).

The results of the logistic regression are reported in Table 4. The final model that contains all predictors demonstrated a good model fit given the significant chi-square test (p < 0.001), a nonsignificant Hosmer and Lemeshow test result, Pseudo R² (0.35), and fairly high classification accuracy (78.7 percent).

Table 4.

Summary of Logistic Regression Models Predicting Social Media Posting

Variables	Model 1						Model 2
Variables	B	SE	Wald'sχ²	p	OR	95% CI	B	SE	Wald's χ²	p	OR	95% CI
Gender^a	0.51	0.26	3.87	0.05	1.66	1.00–2.75	0.51	0.28	3.24	0.07	1.66	0.96–2.89
Age	0.05	0.09	0.35	0.56	1.06	0.88–1.27	−0.01	0.10	0.01	0.91	0.99	0.80–1.21
Receiving free breakfast/lunch^b	0.47	0.26	3.34	0.07	1.61	0.97–2.67	0.31	0.29	1.12	0.29	1.36	0.77–2.41
Positive outcomes beliefs	0.61	0.14	20.15	0.000	1.84	1.41–2.41	0.26	0.16	2.68	0.10	1.30	0.95–1.77
Negative outcomes beliefs	−0.29	0.15	3.97	0.05	0.75	0.56–1.00	0.26	0.19	1.87	0.17	1.29	0.89–1.87
Perceived friends' approval^c							0.13	0.16	0.65	0.42	1.14	0.83–1.57
Peer marijuana use							0.62	0.17	13.17	0.000	1.85	1.33–2.59
Perceived parents' approval							0.48	0.17	8.51	0.004	1.62	1.17–2.24
Parental monitoring							−0.58	0.21	7.93	0.005	0.56	0.37–0.84
Parental media monitoring							−0.14	0.13	1.08	0.30	0.87	0.67–1.13
Model χ²	χ² = 47.47, df = 5, p < 0.001						χ² = 96.64, df = 10, p < 0.001
Hosmer and Lemeshow test	χ² = 12.42, df = 8, p = 0.13						χ² = 8.17, df = 8, p = 0.42
Classification accuracy	74.1%						78.7%
Pseudo R² (Nagelkerke's R²)	0.18						0.35

1 = male, 0 = female.

1 = yes; 0 = no.

Higher scores represent higher approval from other.

CI, confidence interval; OR, odds ratio.

Overall, as given in Table 4, the results of the logistic regression suggested that peer marijuana use (H2) and perceived parent's approval (H4) were positively associated with adolescents' posting. Parental monitoring (H5) was negatively associated with posting, after controlling for other variables in the final model. Positive and negative outcome beliefs about marijuana use (H1a and H1b), perceived friends' approval of marijuana use (H3), and parental media monitoring (H6) were not associated with marijuana-related posting on social media after controlling for other variables in the final model.

Discussion

Close to one-third of Washington adolescents surveyed had shared marijuana-related content—primarily memes, pictures, and videos—on social media. Thirteen percent of the youth reported they had posted content depicting people smoking marijuana. Such content creation and sharing suggest adolescents participate in online conversations about and promotion of marijuana use, although they are not legally able to purchase or use the product (recreational use is legal in Washington state for those at age 21 years and older). The prevalence of self-reported posting was comparable with previous research of college students' posts containing substance use.²³

We found that personal beliefs regarding marijuana use were less associated with adolescents' social media marijuana posting behaviors compared with external influences (including perceptions of parental approval of using marijuana, perceptions of peer use, and parental monitoring). This finding implies that adolescents may be posting content incongruent with personal beliefs in a desire to conform to their peers. This is particularly disconcerting given that young people tend to overestimate their peers' use and acceptance of substance abuse.²⁴

Our results suggest perceptions of peer use were positive predictors of youths' marijuana-related content sharing. Of interest, we also found that perceived peer approval of use of marijuana was not associated with posting marijuana-related content. These results have implications for potential health communication efforts. Social norms campaigns, which have been used successfully to address substance use perceptions among youth,²⁵ may help adolescents make more informed decisions. Such campaigns should focus on correcting descriptive norms, which are norms that describe the quantity or frequency of engaging in a behavior,²⁶ as perceptions of peer use were associated with posting.

Perceptions of parental approval of using marijuana were positively associated with youths' marijuana-related content sharing, whereas parental monitoring was negatively associated with adolescents' social media posting. Previous research has suggested that, regardless of their own use, parents do not want their children to use substances.²⁷ This study suggests it is important for parents to communicate their disapproval of youth's marijuana use to their children. This study also suggests that parental monitoring may be associated with youth's marijuana-related posts. Previous research has found decreased parental knowledge and monitoring were associated with risk behaviors in youth.^12,16 Our work highlights that this may also extend to content displayed on social media. However, we found that parental media monitoring, measured by parents' review of their teens' media use, was not significantly associated with their adolescents' marijuana-related social media sharing. Our results suggest that parents may need to create a family environment that balances parental monitoring and teen autonomy.

There are limitations to consider with this research. The sample was drawn from an online Qualtrics panel, so participants may differ from the general population. Although using a dichotomous variable to assess social media sharing made the most sense analytically and for our objectives, we acknowledge the limitation of measuring sharing frequency on a monthly to daily time scale. In addition, it may be challenging for participants to recall their content sharing over such a time period. Diary studies may be useful in the future. This study did not measure adolescents' use of marijuana, because of sampling limitations and ethical considerations, which could contribute to their sharing behaviors. Future studies should explore this potential influence. Furthermore, future research could benefit from looking at the content shared in peer groups, as we found peers to be a predictor of marijuana-related social media sharing among youth. Finally, parental monitoring and peer use were both measured based on participants' perceptions, and therefore might not be accurate. Future research may want to use dyadic samples that account for others' behaviors.

Conclusion

This research provides insights into adolescents' social media sharing of marijuana-related content. Approximately one-third of adolescents had shared marijuana-related content on social media. Increased perceived parental approval of marijuana use, increased perceived peer marijuana use, and decreased parental monitoring were associated with adolescents' marijuana-related displays on social media, although personal beliefs toward marijuana were not. Such findings support existing research that suggests as young people enter a developmental period characterized by increased autonomy, their perceptions of their peers and parents may become more salient than their personal beliefs.²⁸ The findings suggest that external factors, as opposed to internal beliefs about marijuana, may motivate adolescents' marijuana-related social media sharing.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This investigation was supported in part by funds provided by the State of Washington Initiative Measure No. 502 (PIs: J.F.W. and S.J.T.H.).

References

Cavazos-Rehg

, Krauss

, Fisher

, et al. Twitter chatter about marijuana. Journal of Adolescent Health, 2015; 56:139–145.

Krauss

, Sowles

, Sehi

, et al. Marijuana advertising exposure among current marijuana users in the U.S. Drug and Alcohol Dependence, 2017; 174:192–200.

Krauss

, Sowles

, Stelzer-Monahan

, et al. “It takes longer, but when it hits you it hits you!”: videos about marijuana edibles on YouTube. Substance Use & Misuse, 2017; 52:709–716.

Roditis

, Delucchi

, Chang

, et al. Perceptions of social norms and exposure to pro-marijuana messages are associated with adolescent marijuana use. Preventive Medicine, 2016; 93:171–176.

Ammerman

, Ryan

, Adelman

, et al. The impact of marijuana policies on youth: clinical, research, and legal update. Pediatrics, 2015; 135:e769–e785.

Groth

, Longo

, Martin

. Social media and college student risk behaviors: a mini-review. Addictive Behaviors, 2017; 65:87–91.

Moreno

, Kerr

, Lowry

. A longitudinal investigation of associations between marijuana displays on Facebook and self-reported behaviors among college students. Journal of Adolescent Health, 2018; 63:313–319.

Pew Research Center. (2018). Teens, social media, & Technology. https://www.pewresearch.org/internet/2018/05/31/teens-social-media-technology-2018/ (accessed May 3, 2020).

Spies Shapiro

, Margolin

. Growing up wired: social networking sites and adolescent psychosocial development. Clinical Child and Family Psychology Review, 2014; 17:1–18.

10.

Ajzen

. The theory of planned behavior. Organizational Behavior and Human Decision Processes, 1991; 50:179–211.

11.

Kim

, Lee

J-A

, Sung

, et al. Predicting selfie-posting behavior on social networking sites: an extension of theory of planned behavior. Computers in Human Behavior, 2016; 62:116–123.

12.

Lac

, Alvaro

, Crano

, et al. Pathways from parental knowledge and warmth to adolescent marijuana use: an extension to the theory of planned behavior. Prevention Science, 2009; 10:22–32.

13.

Padilla-Walker

, Coyne

. “Turn that thing off!” parent and adolescent predictors of proactive media monitoring. Journal of Adolescence, 2011; 34:705–715.

14.

Padilla-Walker

, Coyne

, Fraser

, et al. Parents and adolescents growing up in the digital age: latent growth curve analysis of proactive media monitoring. Journal of Adolescence, 2012; 35:1153–1165.

15.

Coyne

, Padilla-Walker

, Day

, et al. A Friend request from dear old dad: associations between parent–child social networking and adolescent outcomes. Cyberpsychology, Behavior, and Social Networking, 2014; 17:8–13.

16.

DiClemente

, Wingood

, Crosby

, et al. Parental monitoring: association with adolescents' risk behaviors. Pediatrics, 2001; 107:1363–1368.

17.

Yzer

, Cappella

, Fishbein

, et al. The effectiveness of gateway communications in anti-marijuana campaigns. Journal of Health Communication, 2003; 8:129–143.

18.

Lee

, Akers

, Borg

. Social learning and structural factors in adolescent substance use. Western Criminology Review. 2004; 5:17.

19.

Willoughby

. BrdsNBz: sexually experienced teens more likely to use sexual health text message service. Health Education & Behavior, 2015; 42:752–758.

20.

Pew Research Center. (2016). Parents, teens and digital monitoring. https://www.pewresearch.org/internet/2016/01/07/parents-teens-and-digital-monitoring/ (accessed May 3, 2020).

21.

Degenhardt

, Chiu

W-T

, Sampson

, et al. Toward a global view of alcohol, tobacco, cannabis, and cocaine use: Findings from the WHO World Mental Health Surveys. Hay P, ed. PLoS Medicine, 2008; 5:e141.

22.

Peduzzi

, Concato

, Kemper

, et al. A simulation study of the number of events per variable in logistic regression analysis. Journal of Clinical Epidemiology, 1996; 49:1373–1379.

23.

Morgan

, Snelson

, Elison-Bowers

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

24.

Sanders

, Stogner

, Seibert

, et al. Misperceptions of peer pill-popping: the prevalence, correlates, and effects of inaccurate assumptions about peer pharmaceutical misuse. Substance Use & Misuse, 2014; 49:813–823.

25.

Berkowitz

. (2005) An overview of the social norms approach. In: Lederman

, Stewart

eds. Changing the culture of college drinking: a socially situated health communication campaign. Cresskill, NJ: Hampton Press, pp. 193–214.

26.

Cialdini

, Kallgren

, Reno

. (1991) A focus theory of normative conduct: a theoretical refinement and reevaluation of the role of norms in human behavior. In: Zanna

ed. Advances in experimental social psychology. Vol. 24. Cambridge, MA: Academic Press, pp. 201–234.

27.

Kosterman

, Bailey

, Guttmannova

, et al. marijuana legalization and parents' attitudes, use, and parenting in Washington State. Journal of Adolescent Health, 2016; 59:450–456.

28.

Meeus

. Adolescent psychosocial development: a review of longitudinal models and research. Developmental Psychology, 2016; 52:1969–1993.