Girls' participation in organised sport declines steeply during early adolescence. Boys' participation declines too, but later and less sharply.
The gap in participation rates that exists among adults is largely created in this window. It is not present at eight, and it is fully formed by eighteen.
The shape of the curve
Participation surveys across many countries find a common pattern: rough parity in structured activity in late primary years, divergence beginning around eleven to thirteen, and a substantial gap established by sixteen.
Estimates of the size vary considerably by country, definition of participation and survey method. Figures suggesting girls drop out at roughly twice the rate of boys during adolescence are widely cited, though the underlying measurement is often less precise than the citation implies.
What is more consistent, and more useful, is the timing. The decline coincides closely with the onset of puberty and with the transition to secondary school, which happen at similar ages and are difficult to separate in most datasets.
What girls say when asked
Qualitative and survey research on reasons for dropping out produces a consistent set of answers, and it is not the set most adults assume.
Body self-consciousness during physical activity ranks high — specifically, discomfort with kit, with being watched, and with changing facilities.
Perceived lack of competence ranks high, often disproportionately to actual ability. The pattern mirrors the self-assessment findings in academic domains.
Competing time demands rank high, and increase sharply with academic pressure at secondary level.
Menstruation is cited frequently in studies that ask about it directly, and almost never in studies that do not, which is a methodological point worth noting.
Loss of enjoyment and increasing competitiveness of provision are cited consistently. Girls who want to play at a recreational level frequently find that provision above a certain age is organised entirely around competition and selection.
Notably, "not interested in sport" is a less common answer than the drop-out rate would suggest. Most girls who stop report having enjoyed it.
Where the drop-off is smaller
The variation between contexts is more informative than the average.
Single-sex provision shows smaller declines in several studies, consistent with the self-consciousness explanation.
Activities with low kit exposure and low observation — swimming being a notable exception in the other direction, and running or team sport in appropriate kit doing better — show different patterns.
Schools with better facilities, in particular private changing space, report smaller declines. This is the least glamorous finding in the literature and one of the most consistent.
And countries with strong community club infrastructure separate from schools show smaller declines, because the transition out of school sport does not coincide with the transition out of sport entirely.
The facilities point deserves emphasis
Changing and showering arrangements come up repeatedly in qualitative research and are almost never the subject of policy attention, because they are unglamorous and require capital spending.
A fourteen-year-old who must change in a shared space with no privacy, immediately after puberty, in a school where being observed matters intensely, has a strong reason to avoid the activity that requires it. This is not a psychological barrier to be overcome with encouragement. It is a building.
The same applies to toilet access during activity, to the availability of period products, and to whether kit is available in a form that is comfortable rather than merely compliant.
The competitive-provision problem
There is a structural feature of youth sport that affects both sexes and affects girls more.
Provision for children is broad and inclusive. Provision for adolescents is increasingly organised around identifying and developing talent, which means selection, which means that a majority of participants are told, implicitly, that they are surplus.
A thirteen-year-old who enjoys playing but will not make the team frequently has nowhere to go. The recreational tier that exists at nine has often disappeared by fifteen.
Countries and clubs that maintain genuine recreational tiers for adolescents report better retention. This is the most straightforward structural fix available and it is a question of how provision is organised rather than of anyone's motivation.
Why it matters beyond the sport
The health case is well established: physical activity in adolescence is associated with bone density accrual, cardiovascular health, and mental health outcomes, and adolescence is a critical window for bone mineral accumulation specifically.
There is also a substantial body of observational research associating adolescent sport participation with later outcomes in education and employment. The causal claims here are weaker than they are usually presented — selection into sport is not random, and the studies that attempt to address this find smaller effects — but the association is consistent.
The more defensible argument does not require the career claims. Adolescence is when most people establish whether physical activity is something they do. A person who stops at fourteen and does not restart has made a decision with a fifty-year tail, and she generally made it because of a changing room.
What has actually worked
Programmes explicitly designed around the stated reasons rather than around encouragement have performed better in evaluation.
That means: kit options, private changing, non-competitive tiers, sessions that are not observable by the rest of the school, and coaching that does not begin by identifying who is best.
Campaigns that focus on enjoyment and de-emphasise competence have reported positive results, and their premise — that the barrier is self-consciousness rather than motivation — is well aligned with the underlying research.
None of this requires anyone to become more interested in sport. Most of them already were.