What moves wellbeing: capabilities and opportunities

The measurement is organised in three levels. Wellbeing is the anchor indicator. Below it are the outcomes — the dimensions that influence wellbeing, divided into capabilities (what a person can do and be) and opportunities (what the environment offers). Below the outcomes are the factors — the specific, measurable indicators within each outcome.

The distinction between capabilities and opportunities is not merely conceptual. It determines action. If a factor is a capability with a low level, the path is to develop it. If it is a missing opportunity, the path is to provide access. Confusing the two wastes resources. A programme that tries to "teach resilience" when what's missing is a trusted adult is aiming at the wrong place.

The Huber Social framework does not impose a fixed view of what matters for wellbeing. Instead, it asks each population what drives their wellbeing locally, and then measures those drivers. This is why the 10 outcomes used in this report were not selected from a generic list — they emerged from the territorial listening described in the next section.

How the outcomes and factors were identified

The selection of what to measure followed a three-step process.

First, listening. Before defining any instrument, the SIC team held conversations with the people who live the programme's reality: the young people themselves, their parents and family members, educators, pedagogical coordination, social workers, and members of the Capão Redondo community. The goal was to understand, from lived experience, what influences these children and adolescents' lives, what makes a difference day to day, what is missing, and what is working.

Second, the action filter. From everything that emerged in the listening phase, not all of it can be measured usefully. Factors like sanitation, public safety, and family income came up in the conversations as real and serious problems. But Rizomas cannot change the neighbourhood's infrastructure or the city's public safety policy. Measurement prioritises factors over which the programme has capacity for influence. This doesn't mean ignoring the context. It means recognising that the programme's role is to develop in children and adolescents the capabilities, and to provide the opportunities, that allow them to develop even within an adverse environment.

Third, grounding in the evidence base. The factors identified in the listening phase — already filtered by the programme's capacity to act — were then cross-referenced against a proprietary evidence base built and curated by Huber Social together with its measurement partners, including SIC (Social Impact). This evidence base aggregates more than a decade of peer-reviewed research on child and adolescent development, is continually updated, and is indexed by programme purpose, mission, target demographics, and contextual variables — so that each project's measurement design is anchored in research directly relevant to its population. Cross-referencing against this base, rather than ad-hoc literature searches, is what allows the framework to combine local listening with consolidated scientific evidence in a reproducible way. The result is 10 outcomes (5 capabilities and 5 opportunities) broken down into approximately 40 measurable factors, each with a corresponding question in the data collection instrument.

Objective

To generate an initial snapshot of the wellbeing of Instituto Rizomas participants, identify the factors most associated with this wellbeing, map priority needs and strengths, and assess signals of association between participation in the programme and the observed outcomes.

Sample

The target population was the 88 young people who completed the 2025 year of the Instituto Rizomas after-school programme. Of these, 78 responded to the questionnaire — a response rate of 89%. The programme serves a larger number of young people throughout the year; the 88 are those who completed the full 2025 cycle.

Participants were divided into subgroups for comparative analyses: by time in programme (newcomers vs. 2+ years), by attendance (high ≥90% vs. low ≤73%), and by participation in the Sunday socioemotional activities (participates vs. does not participate).

Instruments

General wellbeing was measured by the Satisfaction With Life Scale (SWLS), adapted in language for the context and age range of participants, while maintaining the original five-item structure.

Outcomes and factors were measured by 5-point Likert scales, converted to a 0–100 scale for ease of interpretation. Each outcome contains 3 to 4 factors, each factor measured by specific items. In total, approximately 40 factors were measured.

Analysis

Design and its implications

This cycle used a cross-sectional design, measuring all participants at a single point in time. Between-group comparisons (for example, newcomers vs. participants with 2+ years) show association, not causation. It is possible that profile differences between groups explain part of the observed results.

From the 2026 cycle onwards, measurement will include longitudinal follow-up, with repeated measures of the same participants over time. This will allow us to observe individual change and strengthen the evidence about the programme's effect.

This section presents the full results of the measurement. The Results Report highlights the central findings. Here, all data are presented with the analytical context required to interpret them — and now with the statistical rigor indicators explained in the previous section.

General wellbeing

The group's mean wellbeing is 67.8 on a 0–100 scale, with median 69.4 and standard deviation 11.8. The range falls between "slightly satisfied" and "satisfied". The general Brazilian average on the SWLS is approximately 63 (Gouveia et al., 2009, the most recent available reference with the SWLS in Brazil).

The absolute level of wellbeing serves as a starting point. Each territory and each population has a different reality, and comparing absolute numbers across different contexts is of limited use. What matters is how aligned the programme is with the needs that most influence participants' wellbeing, and, over time, how much it manages to lift it.

Drivers of wellbeing

We computed each outcome's correlation with wellbeing to identify the key drivers of wellbeing for these young people. Now with p-value, confidence interval, and FDR correction.

All 10 outcomes show statistically significant correlation with wellbeing. Three of the five strongest drivers remain opportunities. For these children and adolescents, the surrounding context pulls wellbeing more than isolated individual capabilities — but the effect of capabilities is also real and measurable.

This does not diminish the importance of capabilities. It means that, for this group, opportunities function as a precondition. Without safety, without bonds with trusted adults, without recognition, capabilities such as emotional regulation and self-management have less room to develop. Rizomas works on both fronts at once. It builds the environment and develops the capabilities that this environment allows to strengthen.

Priority needs

Priority needs are factors with significant correlation with wellbeing and still-low scores. They are where there is most room for growth and the greatest potential return on the programme's investment: dimensions that demonstrably matter for wellbeing but are still weak in this group. All priority needs are internal capabilities — skills the programme can develop.

This pattern is consistent with what evidence shows. These are skills that depend on repeated practice and an environment that allows them to develop. Emotional regulation is not learned in a lecture. It is learned by regulating emotions in a space where it is safe to make mistakes.

Strengths

Strengths are factors with positive correlation with wellbeing and high score. They represent what the programme already delivers well and needs to protect: dimensions that matter and that young people already recognise as present in their lives.

Participants report having adults who believe in them, who provide practical guidance and feedback on how to improve. They report having a voice, friends who influence positively, and routines that work. The programme delivers the opportunities that matter most for this group's wellbeing.

The Rizomas theory says that the environment the programme offers cultivates skills in young people, and that these skills sustain wellbeing. That is the story. But is it an assertion or a fact? To become fact, it needs testing.

It can be tested. The tool is called mediation analysis. What it answers is direct: how much of the effect of the environment on wellbeing passes through the internal skills that young people develop? Does the environment affect wellbeing because of the skills, or independently of them?

The tested model

The analysis considers three aggregated variables:

• Opportunities (environment): mean of the 5 opportunity outcomes (Trusted Adult, Nurturing Environment, Positive Peer Relationships, Recognition, Personal Development).
• Capabilities (internal skills): mean of the 5 capability outcomes (Self-Management, Emotional Regulation, Social Skills, Responsibility, Future Orientation).
• Wellbeing (SWLS): the anchor indicator.

Three paths are estimated:

• a: from Opportunities to Capabilities. Does the environment develop skills?
• b: from Capabilities to Wellbeing (controlling for Opportunities). Do skills sustain wellbeing even when the environment does not change?
• c': from Opportunities to Wellbeing, controlling for Capabilities. How much of the environment's effect on wellbeing happens without having to pass through skills?

The indirect effect (a × b) is the part of the environment's impact that passes through skills. If it is large, the theory is confirmed.

What the numbers show

How to read this

The environment Rizomas creates is strongly associated with young people's internal skills (a = 0.70). This is the cycle predicted by the programme's theory: nurturing environment, trusted adult, recognition → development of emotional regulation, self-management, future orientation.

Internal skills, in turn, sustain wellbeing (b = 0.38) even when controlling for the environment. It isn't the environment directly that makes a young person feel happy — it's the skills the environment allows them to build.

The indirect effect — the share of the environment's impact on wellbeing that actually passes through skills — is 0.27, with 95% confidence interval [+0.12; +0.43]. The CI does not include zero. The mediation is statistically significant.

The proportion mediated is 30%. This means: about one third of the environment's effect on wellbeing happens through the development of skills. The other two thirds of the effect (c' = 0.63) happen by another path — likely the direct effect of the sense of belonging, safety, and care, without needing to pass through a change in skill.

Important caveat: this is a mediation model tested with cross-sectional data. It shows a pattern of association consistent with the theory, not a proof of causation. In 2026, with repeated measures of the same young people, it will be possible to verify whether changes in the environment at one moment precede changes in skills at the next. That would substantially strengthen the causal interpretation.

Mixing children (≤11 years) and adolescents (12–17) in a single analysis masks important developmental differences. For a 7-year-old, "Future Orientation" carries little weight day to day. For a 15-year-old adolescent, seeing a viable future works as a central protective factor.

This analysis compares, for each outcome, the correlation with wellbeing separately in each age group. Where the difference between the two groups is large, there is an indication that the programme can (or should) act in different ways for each age range.

How to read this

Two outcomes show a statistically significant difference between children and adolescents: Trusted Adult and Future Orientation. Both have a much stronger correlation with wellbeing for adolescents than for children.

Trusted Adult matters for both groups, but for adolescents it is practically decisive (ρ = +0.73). This is consistent with the developmental stage. Younger children still have many adult figures around them (parents, teachers), and the individual weight of a reference adult is diluted. Adolescents are differentiating themselves from their parents and looking for references outside home — when they find an adult who looks at them with real attention, that matters a great deal.

Future Orientation has the same pattern. For children aged 7 to 11, "hope in dreams" and "perseverance" have a looser connection with day-to-day wellbeing. For adolescents, projecting a viable future works as an anchor. Without that anchor, the vulnerability of the territory pulls wellbeing down.

The other 8 outcomes tend to matter equally for both groups (all with positive and significant correlation in both), but without statistical difference between age ranges. This is informative: the programme's foundation works equally well for children and adolescents. What changes is the relative weight of specific dimensions.

Where the opportunities come from

If opportunities are what matter most for wellbeing, the next question is where they come from in these young people's lives.

For each opportunity factor, we asked two questions. The first ("Where do you find this?") allowed multiple applicable contexts to be marked. The second ("Where is this most true for you?") asked for a single choice.

The table below compares how often participants attribute each opportunity to Rizomas and to school, now with a confidence interval for each proportion.

All proportions have a confidence interval that excludes 50% and p < 0.001. The difference between Rizomas and school is statistically significant across all factors. Rizomas appears as the source of these opportunities at three times the frequency of school. The programme also appears frequently alongside home. It does not replace family. It fills gaps when family cannot offer certain conditions, and provides what public schools in vulnerable contexts can rarely supply.

Association with time in programme

We compared wellbeing and key factors among adolescents with different lengths of time in the programme. Now with n per subgroup, statistical test, and effect size.

Mann-Whitney U test (0–1 year vs 3 years): U=1, p < 0.001. Cliff's δ = +0.98 (large effect).

An adolescent with 3 years of Rizomas reports, on average, life satisfaction about 57% higher than a newcomer. The difference is statistically significant and the effect size is large.

Adolescence is a phase where wellbeing typically falls. The body changes, social pressures increase, the future feels uncertain. In contexts of vulnerability, that drop tends to be sharper. The pattern observed at Rizomas goes in the opposite direction. The more time in the programme, the higher the reported wellbeing. This is consistent with the hypothesis that the programme acts as a protective factor.

Factors with the largest differences between newcomers and participants with 2+ years:

The largest differences are in Future Orientation: having dreams and plans, and seeing pathways to make them happen. Marques, Gallagher, and Lopez (2017), in a meta-analysis of studies on hope in school contexts, found that hope is associated with better academic performance, higher engagement, and lower dropout. For adolescents in contexts of vulnerability, this dimension sustains long-term choices even when the surrounding environment does not cooperate.

Association with attendance in the programme

We compared young people with high attendance (≥90%) and low attendance (≤73%). Each subgroup has about 20 young people — a comparison with reasonable statistical footing within the total n.

The factors with the largest differences are tied to Responsibility and Social Skills. These are capabilities that depend on practice, routine, and continued engagement.

Part of this association may reflect profile — young people with higher attendance may come from more organised family contexts, where responsibility is already developed at home. But even controlling for that, one factor stands out: resistance to peer pressure (+31%, p = 0.002, large effect). This capability depends less on family context and more on the social environment the young person is in. Adolescents are particularly susceptible to their reference group. When a young person attends frequently, they repeatedly enter a space with commitments, responsibilities, and structured engagement. In a territory taken over by violence and drug trafficking, this is concrete protection.

Association with participation in socioemotional activities

Among the after-school programme participants, those who also take part in the Sunday socioemotional activities show higher levels of Self-Management and Emotional Regulation — the same outcomes that appear as priority needs.

The young people who take part in Sunday activities are also completing their second year of the after-school programme at the time the measurement was carried out. The observed differences likely reflect the combination of both fronts. The after-school programme offers daily exposure to a structured environment where these capabilities are practised in the everyday, with frequency and consistency. Sunday complements this with a different format: voluntary, lighter, associated with projects and play, without the weight of the school day. One hypothesis is that daily practice builds the foundation and the Sunday format helps consolidate it, in a context where the young person is more receptive. Isolating each effect would require a separate analysis, planned for the next cycle.

This analysis asks a simple thing: within the 78 young people served, are there subgroups with distinct profiles of capabilities and opportunities?

The answer has practical value. Identifying profiles helps personalise the work — young people in a more fragile situation can receive differentiated attention, more robust young people can be invited to lead activities, and the programme gains a concrete way to track movement between profiles over time.

How it was done

We used an algorithm called K-means, which groups people with similar profiles. The algorithm automatically finds the most cohesive groupings, without us having to pre-define who goes with whom. We tested several possible numbers of groups (2, 3, 4, and 5) and the quality criterion (silhouette score) indicated that 3 groups represent the structure of the data well.

The analysis uses the 10 standardised outcomes as the basis for finding the profiles. The visualisation below projects the 78 young people onto two summary dimensions (via PCA), so it is possible to see visually how they separate.

The three identified profiles

What each profile suggests

Profile 1 — Typical middle (53% of young people). Wellbeing close to the overall mean. Shows reasonable levels across all outcomes, without notable peaks or large gaps. Probably the "standard profile" of the programme — young people already benefiting from the Rizomas environment, but with room to grow across several dimensions.

Profile 2 — Flourishing (35% of young people). Wellbeing substantially above the mean (77.1 vs 67.8 overall). Stands out in positive friendships and emotional regulation. Interestingly, this profile has relatively lower scores on some "environment" elements — suggesting that these young people have managed to develop robust internal capabilities that sustain wellbeing more autonomously. In other words: the skills are already working. They may be natural candidates for leadership roles in programme activities.

Profile 3 — Struggling (13% of young people — 10 young people). Wellbeing considerably below the mean (49.4). The most fragile group. Shows important gaps in Future Orientation and Recognition — two of the outcomes most linked to wellbeing in this sample. The proportion of adolescents in this group is higher (50%), consistent with the pattern that adolescence amplifies vulnerabilities when the foundation of support is fragile. This is the group that would benefit most from differentiated attention from educators.

The social impact model organises the programme's logic into a chain that connects what Rizomas invests, does, delivers, and changes in participants' lives.

Impact

Improve the wellbeing of children and adolescents in contexts of vulnerability, so that they are in the best position to reach their potential and live a life they value.

Outcomes (capabilities and opportunities)

Rizomas reaches this impact by developing capabilities and providing access to opportunities.

Capabilities

Self-Management. Sustained attention, inhibitory control, cognitive flexibility, planning. Moffitt et al. (2011) followed 1,000 children for 32 years and found that the level of self-control in childhood predicts health, income, and criminal involvement in adult life, independently of intelligence and family social class.

Emotional Regulation. Emotional awareness, perception of body signals, regulation strategies, recovery after frustration. Aldao, Nolen-Hoeksema, and Schweizer (2010), in a meta-analysis of 114 studies, found that emotional regulation strategies are among the factors most associated with mental health in children and adolescents.

Social Skills. Empathy, respectful communication, resistance to peer pressure, non-aggressive conflict resolution. Durlak et al. (2011), in a meta-analysis of 213 social-emotional learning programmes involving 270,034 students, found gains in social skills, reduced behaviour problems, and improved academic performance.

Responsibility. Keeping commitments, care for belongings, habits and routines, honesty and integrity. The measurement data show that habits and routines have a correlation of 0.51 with wellbeing, one of the highest among all measured factors.

Future Orientation. Self-efficacy, hope (dreams and pathways), perseverance. Marques, Gallagher, and Lopez (2017), in a meta-analysis on hope in school contexts, found that hope is associated with better academic performance, higher engagement, and lower dropout.

Opportunities

Trusted Adult. Reference adult, clear and useful guidance, encouragement, fair structure and limits. DuBois et al. (2011), in a systematic review of 73 studies on mentoring programmes, concluded that the presence of a stable reference adult is one of the most consistent protective factors for young people in vulnerability. Werner and Smith (1992, 2001), in the Kauai longitudinal study that followed 698 at-risk children for 40 years, identified that the presence of at least one stable and caring adult was the factor that most distinguished the children who managed to develop despite adversity.

Nurturing Environment. Psychological safety, clear and fair rules, predictability, respectful conflict resolution. Thapa et al. (2013) reviewed the literature on school climate and found that the perception of safety and fairness in the environment is associated with higher engagement, lower dropout, and better mental health.

Positive Peer Relationships. Pro-social norms, positive friendships, group support, belonging. Chetty et al. (2022), using data from 21 million people in the United States, demonstrated that social capital, particularly connection with people from diverse contexts, is one of the factors most associated with economic mobility.

Recognition. Specific praise, valuing effort, feedback with next step, high expectations. Hattie and Timperley (2007), in a review of more than 500 meta-analyses, found that specific feedback about the process (and not about the person) is one of the interventions with the greatest effect on learning.

Personal Development. Meaningful learning, application of learning, agency, positive role models. Roorda et al. (2011) found that affective relationships with educators increase school engagement, and that this effect is stronger in students from vulnerable contexts.