Sue Frantz

Correlation example: Number of social media platforms used and depression/anxiety

Blog Post created by Sue Frantz on Dec 26, 2016

After introducing correlations, I share a number of examples with students. Here are couple new ones that going into my pool.

 

Young adults (age 19-32) who spend their time on a lot (7 to 11) of different social media platforms are more likely (than those who spend their time on 0 to 2 different social media platforms) to report symptoms of depression. That same relationship exists between the number of different social media platforms and anxiety (Primack, et.al., 2017).

 

Ask students if these represent a positive, negative, or zero correlation (clicker question, show of hands, shout out). (Both correlations are positive.) If students say negative, they may be caught in the trap of thinking that this is a “bad thing,” so it’s negative. Remind students that that is not what positive and negative mean in this context. It can help students in identifying correlations to first note what the two variables are (“number of social media platforms used” and “depression”; “number of social media platforms used” and “anxiety”) and then sort out whether those variables are moving in the same direction (positive) or opposite directions (negative).

 

Next, to help students see that correlations don’t mean causation, ask students to consider the causes, why anxiety and depression may be related to the number of social media platforms used. In think-pair-share, give students a couple minutes to jot down their thoughts, then give students a couple minutes to share their ideas with one or two classroom neighbors, then ask for volunteers to share their thoughts.

 

Students may surmise that jumping from one social media platform to another may cause depression or anxiety. The authors note three possibilities here: “participation in many different social media platforms may lead to multitasking between platforms, which is known to be related to poor cognitive and mental health outcomes;” since each platform has its own rules and customs “as the number of platforms used increases, individuals may experience difficulty navigating these multiple different worlds successfully, leading to potentially negative mood and emotions;” and the more social media platforms you are on you run an “increased risk of damaging gaffes.”

 

Students may also surmise that those who are depressed or anxious may choose to jump from one social media platform to another. The authors suggest “[t]his may be because these individuals tend to search multiple avenues for a setting that feels most comfortable and in which they feel most accepted.” 

 

A third possibility is that something else, a third variable, could be affecting both social media use and depression/anxiety, perhaps loneliness. Those who are more isolated may be more likely to seek out community in social media and being more isolated may contribute to depression/anxiety.

 

Remind students that the value in correlations comes from revealing a relationship between two variables. Students identified a number of possible reasons as to why there is a relationship between those variables. The next step is to do more research on which of those possibilities are right -- and it may be one, some, or all of them. 

 

Side note. Even though their research is clearly correlational, the article's authors are comfortable suggesting that it’s the social media use that’s causing or at least contributing to depression/anxiety. They write “it may not be too soon to suggest that individuals with depressive and/or anxiety symptoms, and who use a high number of different social media platforms may wish to decrease the number of platforms used.” Ask students if this is a fair statement for the authors to make. 

 

Primack, B. A., Shensa, A., Escobar-Viera, C. G., Barrett, E. L., Sidani, J. E., Colditz, J. B., & James, A. E. (2017). Use of multiple social media platforms and symptoms of depression and anxiety: A nationally-representative study among U.S. young adultsComputers in Human Behavior, 69, 1-9. doi:10.1016/j.chb.2016.11.013

Outcomes