Confidence Crash

Members of Legends Insiders have been asking about confidence in girls and the effects of social media. Here are a few research studies and data points to address those questions. As someone with two daughters myself, these are things I've thought a lot about too!

How are changes in confidence different for kids since social media?

Confidence in kids has taken a big hit during the era of social media. For example, we looked at publicly available data from a massive ongoing study of youth well-being and found that from the time that data collection began in 1991 all the way up until 2012, confidence among 8th graders was pretty stable: about 60% of kids each year had a positive attitude toward themselves, thought they were a person of worth, and felt they could do things as well as others. Between 2012 and 2018, that declined to 50% and by 2021, only 38% of eighth graders agreed with those three statements (the 2022 data will be released in October).

The leading theory for these declines comes from Jonathan Haidt and Jean Twenge. Their point of view is well-summarized in this NYT Opinion piece. They highlight the strong negative effect that social media can have (particularly on girls), and noted that 2012 was the tipping point for when a majority of Americans owned a smartphone.

They also note that the years leading up to 2012 included the invention of the 'like' button and the 'retweet', as well as social media platforms optimizing for user engagement by targeting emotional arousal). Haidt and Twenge also found that rates of loneliness among kids started to increase in 2012, coinciding with spikes in symptoms of depression and anxiety, which all aligns with our own findings on declining confidence from above.

Are there gender differences? 

Yes, there are. A few more data points about girls specifically: a Ypulse survey found a steep decrease in confidence among girls around age 12 (and this data is pre-COVID):

And that's a finding that has been replicated in other research, such as this study on self-esteem showing the steep drop for girls from childhood into adolescence (though boys experience sharp declines too):

Does the age at which kids start using social media or smartphones matter?

Yes, a new study that just came out in May found that the age a kid first gets a smartphone is correlated with mental well-being in adulthood. The later they got a smartphone, the better off they were as adults.

This is the graph from that study making the most headlines. It shows that as the age of first smartphone increases (x-axis), mental well-being in adulthood also increases.

The biggest takeaway being that smartphones should probably wait until high school, and that—again—girls are worse off than boys. To put that finding into context, currently, 53% of kids have a smartphone by age 11 and 95% of teens ages 13 to 17 have access to a cell phone.

All to say, confidence is lower in this generation of kids than at any time in the last few decades (as far back as consistent data goes). Confidence generally starts to decline for kids around ages 7 or 8, with a steeper drop coming during the middle school transition years (age 12). This drop is especially pronounced in girls, and social media seems to have a particularly strong negative effect on girls (though boys are also in the negative, just not to the same magnitude).

It's important to remember the caveat that these are all average scores, and while that's useful for writing policy and developing programming, things will work differently for every child.

I try not to assume my daughters are doomed to experience things exactly like all of these graphs show, but I'm definitely on the lookout and making parenting decisions knowing that there's an increased vulnerability there for them.

But that's part of the reason our work at Legends to individualize our programs for each kid is so exciting! Understanding general trends in confidence among kids informs our broader program and theory of change, but individual assessment informs the exercises kids receive.