Skip to content

#Instainsights

Instagram was founded in 2010 and now has more than 50 billion shared photos by 1 billion monthly active users. It is the sixth most popular social network worldwide and especially teenagers love the social network site (72% of teens use Instagram). Earlier this year, Instagram announced to hide likes in some countries to create a less pressured environment according to Instagram’s CEO Adam Mosseri.

But what do we actually know about Instagram users, addiction to Instagram and relationships between Instagram use and mental health? I put together some interesting studies for you!

“We will make decisions that hurt the business if they help people’s well-being and health”

– Adam Mosseri, Head of Instagram

Study 1: A questionnaire to measure Instagram addiction

Researchers from Turkey and the UK created a new scale to measure Instagram Addiction. For that, they modified the Internet Addiction Test (simply by changing the word Internet with Instagram) and validated their findings. Their factor analyses showed that Instagram addiction comprises of two factors: social effect and compulsion.

Social effect refers to the negative effects that an individual may have in their real life due to their Instagram addiction. An item for measuring social effect is “How often do you snap, yell, or act annoyed if someone bothers you while you are on Instagram?”. Compulsion consists of the need for using Instagram as well as the tendency that users forget about the time while using Instagram and avoiding troubles in their real life through using Instagram. One of the compulsion items is “How often do you find yourself saying “just a few more minutes” when on Instagram?”.

After creating their Instagram addiction scale they wanted to find out the relationships between personality traits and Instagram addiction. For that, 752 university students with a mean age of 20 years had to fill out questionnaires regarding their big five personality traits and the new Instagram addiction scale. Moreover, they were asked about their daily internet use and their level of self-liking.

Using a path analysis, the study findings show that agreeableness, conscientiousness, and self-liking are negatively associated with Instagram addiction. To put it in other words, people that have problems to fit in, organize themselves and don’t like themselves too much are more vulnerable to Instagram addiction. For me it was not too surprisingly that they found a positive correlation between daily Internet use and Instagram addiction. What is especially interesting for me is that the authors discuss that addicted Instagram users have different personality traits than addicts of other social network sites such as Facebook or Twitter. The researchers suggest that personality traits determine which platform the individuals prefer in the first place.

Path Analysis of Kircaburun & Griffiths (2018)

Study 2: Picture Perfect

This study was especially worrying in my point of view. Dutch scientists conducted an online experiment with 144 adolescent girls. To some of the girls, they presented ten original Instagram photos and to other ones ten manipulated ones.

“We edited the faces and bodies visible in the photos, by removing eye bags, wrinkles, and impurities, and by reshaping legs to be thinner and waist to be slimmer”

– citation from study Kleemans et al. (2018)

Next, they asked the girls about their body satisfaction and their tendency to social comparison. Their findings show that manipulated Instagram pictures were rated more attractive (bad) and that the girls thought that even the manipulated pictures provide a representative view of reality (really bad). In addition, seeing the edited pictures directly led to lower body satisfaction, an effect that is even stronger for girls with higher social comparison tendency (also very bad). This study really shows how dangerous the practice of editing pictures on social media is and what effects it has on psychological factors of adolescent girls.

Manipulated photos (right) and original photos (left), stimuli material by Kleemans et al. (2018)

Study 3: Show me your Instagram account and I tell you who you are

This study recruited 113 participants with a median age of 30 years to fill out a big five personality questionnaire and give access to their Instagram accounts. The authors crawled more than 22.000 pictures and looked for picture features such as brightness and saturation that can be associated with the personality characteristics of the participants.

It has to be said that most correlations were only marginally significant and that some personality traits are easier to predict than other ones. Still, this study is very interesting in my point of view and shows the potential of visual data for psychometrics. The personality trait openness to experience (this one friend who always know the best restaurant int town) goes along with pictures with a lot of green, more vivid colours than other personalities and few faces (maybe because of the #instafood content? ;-)).

Summary of findings by Ferwerda et al. (2015)

Study 4: Instagram analysis for depression screening?

US researchers got access to Instagram data from 166 individuals (which led to 43.950 pictures) and applied machine learning tools to identify markers of depression. Similarly to the study before, they used metadata components such as color, brightness or saturation of the pictures. With their trained model they were able to diagnose depression better than general practitioners and their models held even when only analyzing posts made before the individual was first diagnosed.

They found out that depressed individuals’ photos tend to be bluer, darker and grayer. Depressed individuals use fewer filters and if they use filters, they prefer “Inkwell”, which converts colour to black-and-white. Next, posts of depressed individuals tend to receive less likes but more comments (let’s see how this change with the like ban).

Instagram filter usage difference for healthy and depressed individuals
reported by Reece and Danforth (2017)

I think this is an interesting piece of research as it shows us the possibilities of computational models in an increased digitalized society. It shows that psychological changes are transmitted in social media use and therefore present an interesting avenue for further research on mental health screening.

That was a ride … Four completely different studies having Instagram as a common theme. I hope you enjoyed the insights and I’m looking forward what the “likes ban” will bring.

Sources:
Study 1: Kircaburun, K., & Griffiths, M. D. (2018). Instagram addiction and the Big Five of personality: The mediating role of self-liking. Journal of behavioral addictions7(1), 158-170.
Study 2: Kleemans, M., Daalmans, S., Carbaat, I., & Anschütz, D. (2018). Picture perfect: The direct effect of manipulated Instagram photos on body image in adolescent girls. Media Psychology21(1), 93-110.
Study 3: Ferwerda, B., Schedl, M., & Tkalcic, M. (2015, September). Predicting personality traits with instagram pictures. In Proceedings of the 3rd Workshop on Emotions and Personality in Personalized Systems 2015 (pp. 7-10). ACM.
Study 4: Reece, A. G., & Danforth, C. M. (2017). Instagram photos reveal predictive markers of depression. EPJ Data Science6(1), 15.