University of Virginia researchers followed 169 adolescents for a decade and found a striking difference between having a supportive best friendship at age 15 and being widely sought after by classmates. Stronger close friendships predicted better self-worth and fewer depressive and social anxiety symptoms by age 25; broader peer preference was associated with relatively greater social anxiety later in life.
The findings, discussed again in a November 9, 2026, ScienceBlog report, are not new 2026 experimental results. The original peer-reviewed study was first published online in August 2017 and appeared in the January–February 2019 issue of Child Development.
Researchers caution that these associations do not mean popularity causes anxiety, or that everyone with a best friend will experience good mental health.
What the ten-year study actually measured
The study, led by Rachel K. Narr with Joseph P. Allen, Joseph S. Tan and Emily L. Loeb, followed 169 adolescents from ages 15 to 25. It drew participants from a larger longitudinal investigation of students attending a public school serving suburban and urban communities in the southeastern United States. The friendship study’s observations ran from 2001 to 2011.
Researchers asked each adolescent’s close friend to describe their bond, including attachment, communication, trust and closeness. They separately measured peer affiliation preference: how often schoolmates nominated an adolescent among the people they most wanted to spend free time with. In other words, the researchers compared the depth of an individual friendship with desirability as company in the wider peer group.
At age 25, 146 of the original 169 participants supplied follow-up data. Statistical models used the original sample to address missing observations and considered earlier mental health, gender and family income. The researchers reported that strong, supportive friendships predicted relative gains in self-worth and declines in social anxiety and depressive symptoms over the decade, whereas wider peer preference predicted relatively higher social anxiety.
The results are reported in “Close Friendship Strength and Broader Peer Group Desirability as Differential Predictors of Adult Mental Health.”
Being liked is not the same as having status
The word popular can obscure an important distinction. Psychologist Mitch Prinstein describes likability as being valued, trusted and inclusive, whereas status popularity emphasizes visibility, influence and dominance. Those qualities are not interchangeable.
In the Virginia study, the specific peer measure was classmates’ preference for spending time with someone, not a separate rating of dominance, fame or social-media reach. It would be misleading to claim that the researchers directly measured today’s online influencer status.
In an American Psychological Association interview, Prinstein explains why the difference matters for adolescent development and adult relationships. Being socially visible may produce short-term rewards, while the abilities needed to build reciprocal, supportive ties may matter across changing life stages.
What a friendship can provide that a crowd cannot
A trusted friend may offer something that an admiring crowd cannot reliably provide: someone to confide in, work through disagreements with, and count on when circumstances change. The Virginia investigators found that the stability of a close friendship from ages 15 to 16 helped explain some of the strongest long-term associations with mental health.
This does not make having many friends a problem. Wider social acceptance can bring confidence, opportunity and companionship. The researchers’ question was which kinds of adolescent relationships predicted mental health changes after accounting for each other.
Earlier research on the pressure to look cool
A separate 2014 Child Development study led by Allen followed 184 adolescents from age 13 to 23. It examined a pattern the researchers called pseudomature behavior, including early romantic behavior, minor rule-breaking and emphasis on associating with physically attractive peers. At 13, those behaviors were linked to being perceived as popular. By early adulthood, the young people displaying more of them were more likely to report substance-related and other difficulties and were seen by peers as less skilled in relationships.
That earlier investigation was not a test of whether all popular teenagers become troubled adults. It focused on specific behaviors, and its participants came from the same larger Virginia research program as the later 169-person friendship analysis. Read the Society for Research in Child Development’s 2014 summary.
What other U.S. research tells us
A 2009 University of Rochester study examined the difference between pursuing personal growth, close relationships and community contribution and pursuing external recognition such as wealth, fame and appearance. The researchers reported that progress toward intrinsic goals was associated with better psychological well-being, whereas progress toward extrinsic goals did not deliver the same benefits. Its small sample of recent college graduates means these results should not be generalized to everyone.
Meanwhile, the Harvard Study of Adult Development, which began in 1938, has repeatedly drawn attention to relationships and health over the lifespan. Director Robert Waldinger calls the practice of maintaining personal ties “social fitness.” That broader work supports paying attention to relationship quality, although it does not replicate Virginia’s particular measurements or prove causation in that adolescent sample.
What parents, teachers and communities can learn
For families and schools, the implications are practical. An adolescent may be invited to many events yet have no one with whom they feel safe discussing a problem. Another may have little social visibility but a dependable friend.
Adults can make room for sustained friendships, model constructive conflict resolution, take social exclusion seriously and avoid treating online follower counts as a measure of well-being. School and youth programs can value cooperation, empathy and shared activities alongside conventional achievement.
These are implications informed by the research—not interventions tested or proven effective in the ten-year Virginia study.
Limitations and what remains unknown
The study was observational, so it cannot establish that close friendships themselves caused better mental health or that peer preference caused anxiety. Unmeasured influences may affect both relationships and later well-being. The sample came from one U.S. region, and the participants grew up before today’s social-media environment. Its conclusions should not be presented as direct evidence about TikTok or Instagram.
The central finding is nevertheless substantial: friendship depth and broad social desirability were distinct experiences with different long-term associations. A school’s most recognizable students are not necessarily the students with the strongest emotional support networks.
Watch: Why popularity matters
The American Psychological Association’s original interview with Mitch Prinstein explores the difference between being liked and having status. Watch the official APA video on YouTube (published August 14, 2019).
Sources and image credits
- Narr et al., original peer-reviewed friendship study, Child Development (online 2017; print 2019).
- U.S. National Library of Medicine bibliographic record.
- Society for Research in Child Development: 2014 “cool kids” study summary.
- American Psychological Association: Mitch Prinstein on popularity.
- University of Rochester: intrinsic versus extrinsic goals.
- Harvard Gazette: social fitness and adult relationships.
- Featured illustrative photo: Loren Kerns, Tigard High School Prom 2023, CC BY 2.0. Wikimedia Commons 1280-pixel rendition; no editorial modification. The people depicted are not participants in this study.
- Inline illustrative photo: PCHS-NJROTC, Port Charlotte High School, May 2010, CC BY-SA 3.0. The people depicted are not participants in this study.
