Dana Litt on “Pictures of Last Night Ended Up Online: Social Networking Sites and Alcohol Use”

This past Wednesday September 28, Dr. Dana Litt of the University of Washington Center for the Study of Health and Risk Behaviors discussed the role of social media on alcohol use in adolescents in a talk titled “Pictures of Last Night Ended Up Online: Social Networking Sites and Alcohol Use” (a play on the lyrics from a Katy Perry song). According to Dr. Litt, 75% of drinking in the USA is done in heavy episodic drinking (formerly known as binge drinking). The percentage is trending upwards, with women under the age of 25 being the fastest growing group. Dr. Litt discussed her research finding ways to explore the influences of social media on drinking and possible ways to lower heavy drinking in at risk populations.

Dr. Litt started by introducing us to the Theory of Reasoned/Planned Action, the Prototype Willingness Model, and the Social Reaction Pathway. Engaging in risk is volitional, social in nature, and images influence decisions. Unfortunately, the Theory of Reasoned/Planned Action falls apart on those younger than 20 years old, so they are at increased risk. adb_20_2_185_fig1a

Facebook Experimental Design & Results:

So how does the average 15 year old get her exposure to social influences? According to Dr. Litt, it’s probably from Facebook, Twitter, Instagram, Sanpchat, Reddit, etc. The average 15 year old has 800 Facebook friends. Dr. Litt designed an experiment creating 48 different fake Facebook profiles. A sample group of 13-15 year olds were randomly assigned to view either alcohol or non-alcohol using profiles. They were then asked how willing they would use alcohol in the future. It was no surprise that those exposed to alcohol positive profiles were more willing to use alcohol in the future compared to the non-alcohol profiles.

Twitter Study Design & Results:

Dr. Litt then turned to Twitter because she felt that most social media studies focused on Facebook, and less was known about Twitter. Twitter presented a unique case because fewer adults are active on Twitter compared to Facebook. Dr. Litt recruited study participants through the social media platform. Her research partner then compiled the twitter data and analyzed the tweet contents using a computer code. Dr. Litt’s research team believes they were able to collect a diverse sample of high school/college aged adolescents of varying ethnic and socioeconomic identities. When checking for bias in their sample, they found no cognitive and drinking difference with the non-Twitter population. Some interesting results they found when asking their Twitter sample users about certain posts or photos was that sometimes the appearance of the solo cup wasn’t actually a cup of alcohol. Sometimes their social media presence was an outward appearance not true to the real story of what happened. The team was also able to collect the frequencies of certain words used associated with alcohol, such as ‘wasted,’ ‘wine,’ ‘beer,’ ‘keg,’ etc.

Future Research:

The number of new social media platforms continue to rise every year, so it is difficult to keep up on all fronts of social media exposure adolescents experience. Dr. Litt would like to dig deeper into Twitter to examine what type of Twitter feed her sample users view. So far, her team has managed to go within 3 degrees of followers. Her research team is also currently developing a coding system to analyze Instagram and visual posts for content and hashtags. She believes pretty much every social media platform is valuable public data, but unfortunately, the only one they haven’t been able to look closely at is Snapchat. This is even more unfortunate because they believe Snapchat is a major factor in exposing others to alcohol use.


I enjoyed hearing about Dr. Litt’s psychology work as I take into context what I learn in my Developmental Psychology course. I find the subject very interesting — it’s definitely a subject that encompasses more than just psychology. Alcohol use has a profound influence in music, sociology, politics, — almost everything, almost everywhere. Personally, I’d say my exposure to alcohol use and party culture in general is pretty low because I go to Bryn Mawr and my friends don’t drink heavily. I find alcohol use on Facebook more prevalent among people I went to high school with. I’d say I’m not sheltered from the fact that it happens, and I’m probably not aware of to what extent some of my classmates drink.

Here’s an extra song on alcohol and some song’s I’ve put together about drug use:

A little bonus: