What good mentorship does for young people: what the research says
Will a mentor make a useful difference to my child, and how would I judge a programme before choosing it? Research on youth mentoring gives parents grounds for measured optimism. Reviews find small average benefits and suggest looking closely at the relationship and how a programme is run. The evidence supports modest expectations; promises of personal transformation go further than the research allows.
How much difference does mentoring make?
Raposa and colleagues' 2019 meta-analysis in the Journal of Youth and Adolescence brought together 70 evaluations involving 25,286 young people, whose average age was 12. It examined one-to-one mentoring between generations in English-language studies from 1975 to 2017. The 70 evaluations were not all randomised trials, and their participants did not represent Indian teenagers as a whole.
The pooled effect was Hedges' g = 0.21, roughly one-fifth of a standard deviation. This puts the average difference on a shared statistical scale. The authors describe it as small by the usual thresholds. They also note that it compares more favourably with typical effects of universal youth prevention programmes.
A value of 0.21 does not mean a 21 per cent improvement, or that 21 per cent of participants benefited. It is an average, and individual children's experiences may differ. An earlier meta-analysis by DuBois and colleagues, published in 2002, reviewed 55 evaluations and also found modest or small average benefits. The reviews may include some of the same studies, so avoid adding their totals as separate bodies of evidence.
Having a purpose seems to help
Christensen and colleagues' 2020 follow-up meta-analysis examined 48 studies. It compared targeted programmes, which matched activities to the difficulties young people faced, with non-specific programmes focused on the mentoring relationship. Average effects were g = 0.25 for targeted programmes and g = 0.11 for non-specific programmes; the overall pooled effect was g = 0.19.
This makes a programme's purpose worth examining. Ask what difficulty its activities address and how it will judge progress. The figures are averages, rather than evidence that every targeted programme works twice as well. Ask how a service described as personalised compares with the models studied.
The programmes differed in their participants, size, duration, delivery and evaluation. Missing data meant the researchers could not account for every possible difference. Targeted evaluations also measured outcomes more closely linked to their goals. These limits leave open whether targeting itself produced the larger effects.
The quality of the relationship
Formal programmes are only part of the evidence. Van Dam and colleagues' 2018 review examined natural mentoring: an existing relationship with a significant person other than a parent or step-parent. Its 30 studies looked at these existing relationships, rather than mentors assigned through a programme.
Across 24 studies with 63,327 participants, having a natural mentor was modestly associated with better youth outcomes, with a correlation of r = 0.106. The association between relationship quality and outcomes was r = 0.208. Quality included closeness and trust, social support and support for the young person's autonomy.
These correlations show links rather than proof of cause. They use a different statistic from Hedges' g, so keep the findings on natural mentoring and formal programmes separate. For a parent, the useful questions concern the relationship itself: does the young person feel supported, is there trust, and is there room for their own choices?
What supports a programme
The DuBois review identified several promising practices associated with better results: ongoing mentor training, structured activities, expectations for contact, parental support or involvement, and monitoring of implementation. These are concrete matters to discuss with a provider. The review did not independently prove each practice through a randomised trial.
Ask how mentors continue to learn, what sessions involve, what contact is expected and how the programme checks that its plans are being followed. Ask the provider to explain these arrangements clearly, using the review's findings to guide the conversation.
Raposa's analysis also found larger effects in programmes with a greater proportion of mentors from helping professions. This was a link across programmes, rather than a randomised test of qualifications. Judge a mentor's ability beyond their profession or the prestige of their job.
Implementation deserves attention for another reason. The 2002 review warned that poorly implemented programmes could have adverse effects for young people with personal vulnerabilities. The warning concerns poor delivery, rather than mentoring in general.
What is being measured?
A 2026 systematic review by Tiraieyari and Krauss examined 27 school-based mentoring studies published between 2010 and January 2025. It described and compared the findings rather than calculating a new pooled effect size. Read it alongside the earlier meta-analyses; its later date does not supply a new numerical estimate.
The review grouped assessments into school functioning, psychological well-being, social and interpersonal functioning, and substance use and behavioural conduct. School functioning dominated the measures. These categories tell us what researchers assessed, rather than showing that every programme improved every area.
When a provider cites research showing that mentoring works, ask which outcome was measured. Match any promise to the area studied, while remembering that school measures may leave out other benefits of a mentoring relationship.
What can Indian families take from this?
All but two studies in the 2020 analysis were conducted in North America, and only English-language studies were included. The authors warn against assuming that the findings apply to programmes elsewhere. Most studies in the 2026 school-based review were conducted in the USA, with differences in methods, sample sizes, delivery and school context.
For families choosing a programme in India, these findings offer questions to ask a provider. The effect sizes are not established estimates for Indian learners, and evidence from schools does not directly show how well a private online programme works. The reviews discussed here provide no independent evaluation of Learnacy. Ask the service for its own answers.
Questions to take into the decision
Start with what the young person wants help with, then ask how the mentoring activities relate to that need. Ask what support mentors receive, how often they are expected to be in contact and how the programme checks its delivery. Discuss whether the relationship offers trust, support and room for the young person's own judgement. Keep expectations proportionate to the modest average benefits in the research.
Parents exploring an option can read about Learnacy Mentorship, which One Young India describes as serving learners in Grades 8 to 12 and undergraduates, primarily online. Its page describes onboarding that maps a learner's interests and gaps before pairing them with mentors to design a personalised journey.