Is Exercise Socially Contagious?

Photo of a business man doing a curl at his computerDo you want to be happier? Of course you do, who doesn’t? Perhaps you should try to surround yourself with happy friends. A study that was published back in 2008 in the British Medical Journal called Dynamic spread of happiness in a large social network: longitudinal analysis over 20 years says that happiness can go viral within your social network (like a good Keanu meme or a hilarious cat video). The researchers concluded that, “People’s happiness depends on the happiness of others with whom they are connected. This provides further justification for seeing happiness, like health, as a collective phenomenon.”

In a similar way, if your friends are overweight, that can apparently have an effect on you too. In a a study called The spread of obesity in a large social network over 32 years, it was concluded that, “Network phenomena appear to be relevant to the biologic and behavioral trait of obesity, and obesity appears to spread through social ties.”

If you find that interesting, intriguing, or maybe even terrifying, now researchers have identified that exercise can also be contagious and spread via social media.

Exercise Contagion

In a paper called Exercise contagion in a global social network published in Nature Communications, researchers say that, “we show that exercise is socially contagious and that its contagiousness varies with the relative activity of and gender relationships between friends.”

To come to this conclusion, they looked at the running habits of about a 1.1 million runners around the world who used some variety of fitness trackers for five years. The runners collectively had about 3.4 million social network ties of which the researchers analyzed 2.1 million users for whom they could target specific geographic and weather information. Over the five years, these social media runners ran a total 350 million kilometers and all of this running was automatically posted online for their friends to see.


I am going to pause for a moment and point out that the fact the data was automatically posted is important. It reduces the inherent issues that come with self-reporting. And that is not just for exercise data but really anything that you want to study thoroughly. The problems with self-reporting include:

  • Honesty management: researchers who use self-report questionnaires are relying on the honesty of their participants.
  • Introspective ability: even if a participant is trying to be honest, they may lack the introspective ability to provide an accurate response to a question.
  • Understanding: participants may all vary in their understanding or interpretation of particular questions.
  • Rating scales: many questionnaires use rating scales to allow respondents to provide more nuanced responses than just a yes or no.
  • Response bias: the  individual’s tendency to respond a certain way, regardless of the actual evidence they are assessing.
  • Control of sample: which has become more of an issue with the boom of online questionnaire distribution sites like Survey Monkey.

When the researchers looked at the automatically uploaded run data and then weather patterns of the cities the runners lived in, they used the information to examine different parts of the network. Letting the weather set up their experiment for them, acting on the assumption that nice weather is more likely to make people run, they found some very cool correlations.

Fair Weather Runners

This experiment was able to determine whether the weather in Vancouver can cause changes in the running behavior of New Yorkers.

Let’s say that it is a nice day where you live, then you and your friends are likely to go out and run. If it’s crappy outside where I live, me and my friends will find excuses to skip our jog. Since all cities have quite different weather at any given point in the year, this experiment was able to determine whether the weather in Vancouver can cause changes in the running behavior of folks in New York. If it does, this can really only be happening due to peer to peer influences of friends who live between Vancouver and New York.

That is exactly what the researchers saw. One city’s running could have a direct effect on another city’s runners if the cities were socially connected. The paper states that “… we found strong evidence of the possibility of social contagion in running behaviours …”

In a nutshell, this is what they saw: on the same day, an additional kilometer that is run by your friend influences you to run an additional 0.3 kilometers. Or speeding up your own run an extra kilometer per minute pushes your friends to run an additional 0.3 kilometers per minute faster than usual. Or if your friends run 10 minutes longer than usual, you may be inclined to run about three minutes longer. If you burn an extra 10 calories, then your friends wind up burning 3.5 more calories.

Our Closest Peer

If we look a little closer at the data, some even more interesting trends come into play. In general, runners are more influenced by their peers whose performance is slightly worse, but not that much worse than their own. Also, the runners were influenced more by peers who perform slightly better but not significantly better than they do. So runners we think are our closest “peers” (particularly those who we consider slightly inferior) are the most likely to get us to dig deeper.

This makes sense when you think about it. Most of us don’t like to compare ourselves to the pros, because we don’t want to continually feel like failures. At the same time, we may feel like jerks if we are constantly comparing ourselves to people who are just beginning or clearly don’t have what it takes.

The study also found that less active runners influence the more active runners more significantly than more active runners influence less active runners. Which I guess could suggest that we are all looking for a reason to take the day off and if our social network is made up of people who generally bail at the first sign of rain, that could bring us down as well.

This brings up something the study calls Competitive Comparisons. A debate exists about whether we make upward comparisons to those performing better than ourselves or downward comparisons to those performing worse than ourselves. Comparisons to those ahead or above us can motivate our own self-improvement, while comparisons to those behind us may create “competitive behaviour to protect one’s superiority.” The study’s findings indicate support for both arguments, but the researchers say that the effects are much larger for downward comparisons than for upward comparisons. As much of a bummer as that may be, it appears to be working in our favour.


Men strongly influence other men but not women while women moderately influence both men and women.

Another point of interest is that gender plays a major role in who gives and receives influence. Same-sex influence is strong, while influence among mixed-sex is statistically weaker. For instance, men strongly influence other men but not women while women moderately influence both men and women.

The researchers suggested that this may be due to gender differences around exercise and competition. Men report receiving and being more influenced by social support in their decision to adopt exercise behaviours, while women report being more motivated by self-regulation and individual planning. Also, there is that old macho thing where men tend to be more competitive and specifically more competitive with each other.

If Exercise is Contagious, What’s the Tip?

So, this is interesting and all but where is the quick and dirty tip? Well, we should probably start thinking about ourselves and our networks being more intertwined than we currently do. We apparently are influenced by and project influence on our own social network in very cool and unexpected ways. When personal happiness, obesity, and exercise habits are at stake, perhaps we should take pause before we hit “follow” or “like” or “subscribe” on every friend from high school that crosses our feed.

I am not trying to suck the joy out of social media or imply that we should have a “no fatties” or “must exercise religiously” rule in place on our Facebook friends, but perhaps we should be a little more mindful as we scroll through our feed of how the behaviour of our network is affecting us, our moods, or our habits.

Conversely, if we think of ourselves as being a social media “influencer” we may want to rethink how or how much of our exercise data we are sharing. If we truly want to motivate and be motivated in return, we will want to challenge the dudes to compete and influence the ladies to self-motivate. And for goodness sake people, let’s cut it out with the weight room selfies. Seriously, you’re there to workout and I am pretty sure that there was no duck-face included in the automatically-uploaded data used in this study.

I can’t believe I’m recommending this but perhaps simply sitting down and scrolling through your feed will get you motivated.

The final tip that I can glean from this study is that perhaps we can actually use our social media feed as a force for good. Perhaps we can even use it as a motivator on those days when we just can’t seem to get our mojo, or our gumption to do our workout of the day. I really can’t believe I am recommending this, but perhaps simply sitting down and scrolling through your feed will get you pointed in the right direction—that direction being toward the gym. Just make sure you set a time limit on how long you allow yourself to scroll so you don’t accidentally turn your 45-minute upper body workout into a hasty 8-minute Tabata set.

Benefit to Society

Even if we don’t do anything actionable right now about our own use of social media and our fitness trackers, this study does show how effective monitoring these networks can be in helping scientists design many types of interventions to limit social ills and even enhance social benefits.

The study authors concluded: “The granularity and precision with which fitness tracking devices record real-world health behaviours portends a sea change in our understanding of human behaviour and social influence at scale. Compared with prior studies, which relied on imprecise and frequently inaccurate self-reports, the potential for these kinds of data to extend our understanding of social behaviour in real-world settings is difficult to overstate.”