Surveys measuring loneliness across age groups have produced a counterintuitive finding with reasonable consistency: reported loneliness is high in adolescence and young adulthood, frequently higher than in older age groups.
This contradicts the standard assumption that loneliness is primarily a condition of the elderly, and it has been replicated in several large surveys across different countries.
Loneliness is not solitude
The distinction is fundamental and frequently missed.
Loneliness is the subjective experience of a gap between the relationships you have and the relationships you want. Social isolation is the objective state of having few contacts.
They correlate weakly. A person can be surrounded by people constantly and be lonely; a person can spend most of her time alone and not be.
This explains why adolescent loneliness is high despite adolescents having more social contact than almost any other group. The quantity is not the variable.
What drives it at that age
Several features of adolescence make the gap between desired and actual relationships particularly likely to open.
Expectations rise. Adolescence involves a shift from friendship based on shared activity to friendship based on intimacy and disclosure. The standard for what counts as a real friendship rises sharply, and relationships that were adequate at eleven are not at sixteen.
Groups reorganise. Friendship groups in adolescence are unstable. Studies tracking friendship networks find high turnover, and each reorganisation produces a period of mismatch.
Comparison is visible. Social media makes others' social lives observable in a way they were not previously. A person who is alone on a Friday evening can now see who is not, which converts a private experience into a comparative one.
Disclosure feels riskier. The thing that resolves loneliness — telling someone — is the thing loneliness makes hardest, because admitting it feels like confirming a social failure.
The gender pattern
Findings here are mixed and depend on measurement.
Girls generally report closer friendships and more disclosure within them. They also report higher rates of relational stress and are more affected by friendship disruption in several studies.
The proposed explanation is that closer relationships raise the cost of their loss. A friendship built on high disclosure is more valuable and its ending is more consequential.
Some studies find girls reporting higher loneliness, some find no difference, and the direction varies with the instrument used — which is a reasonable indication that the effect is not large or robust.
Why it matters
Loneliness is associated with depression, anxiety and poor sleep, and longitudinal studies suggest the relationship runs in both directions.
There is a substantial literature on physical health associations in adults, including cardiovascular outcomes and mortality. The adolescent evidence is thinner but the same associations appear.
The mechanism proposed in the research involves a cognitive shift: chronic loneliness is associated with increased vigilance to social threat, which produces interpretation of ambiguous social signals as rejection, which produces withdrawal, which increases loneliness.
That self-reinforcing loop is the reason loneliness persists rather than resolving naturally.
What works
A meta-analysis of loneliness interventions found that the most effective category was not increasing social contact, providing social support, or improving social skills.
It was addressing maladaptive social cognition — the interpretation patterns described above.
This is an important and non-obvious finding, and it explains why well-intentioned interventions frequently fail. Putting a lonely person in a room with other people does not address the interpretation that everyone in the room finds her tedious.
The interventions that work look more like cognitive therapy than like social programming: examining the evidence for beliefs about how one is perceived, testing predictions, and noticing the asymmetry between how one judges one's own social failures and others'.
What a lonely adolescent can be told that is true
That it is extremely common at this age and that almost everyone reporting it believes they are unusual in it. Surveys consistently find high rates alongside a belief among respondents that others do not experience it.
That the visible social lives of others are curated, and that the comparison is against a presentation rather than a reality.
That friendship groups reorganise repeatedly during these years, that this is structural rather than personal, and that people who lose a group at fifteen very frequently have a better one at eighteen.
That the prediction "if I say this, they will think less of me" is testable and is usually wrong. The help-seeking research consistently finds people underestimate others' willingness to respond well.
And that structured shared activity produces friendship more reliably than attempts to make friends directly, because it supplies the repeated low-pressure contact that relationships actually require. This is unglamorous advice and it is the closest thing to a reliable method that exists.