**4. Results**

The results show that the mentored students that took part in the project reduced the impact of stressful life experiences measured through the Youth Environment Stressors Index (Table 2). During the course of the programme (pre-post), they significantly reduced their levels of environmental stress (*Z* = 5.931; *p* = 0.000; *dCohen* = 1.07). This difference is even greater in the girls (*Z* = 4.679; *p* = 0.000; *dCohen* = 1.238) than in the boys (*Z* = 3.666; *p* = 0.000; *dCohen* = 0.906). The differences are also significant if controlled by age.

In addition to environmental stressors, another source of anxiety in children is the interaction with the native population, since it can expose them to more or less subtle prejudices and discrimination. In this regard, the mentoring relationship did not cause the study participants to reduce their attitude of suspicion and feeling of distrust toward others (*Z* = −1.090; *p* = 0.276). No significant differences were observed by either sex or age.

The perceived availability of social support strengthened, in all cases, the psychological well-being of the mentees. The perception of social support increased and this is reflected in a higher score of the Social Support Index for the whole sample (*Z* = −6.922; *p* < 0.01; *dCohen* = 1.319; Table 2). More specifically, having informal sources of social support correlated negatively with psychological symptoms of stress (Spearman's rho = −0.337; *p* = 0.000; *dCohen* = 0.716) and positively with the Resilience Index (Spearman's rho = 0.246; *p* = 0.002; *dCohen* = 0.508), although resilience scores were only significant for the girls (*Z* = −1.661; *p* = 0.097, and as seen when considering a 0.10 error).


**Table 2.** Non-parametric tests for pre-post measurements.

Source: the authors.

After six months, the group of adolescents perceived greater social support from their local network, especially the girls and those under 12 that had more diverse sources of help after the intervention (Table 2). Both of these groups found social support networks in the school and stated having at least one classmate that helped them with the homework (girls group: *Z* = −2.251; *p* = 0.024; *dCohen* = 0.524; under 12 group: *Z* = −2.171; *p* = 0.030; *dCohen* = 0.511; Table 2). This tangible support, which includes physical acts of help and educational support in schoolwork, was not the only kind of support they received. The support of self-esteem, related to recognising their personal worth, was also perceived by these mentees in the school, where they assured: "Someone at school makes me feel successful" (girls group: *Z* = −2.003; *p* = 0.045; *dCohen* = 0.463; under 12 group: *Z* = −1.807; *p* = 0.071; *dCohen* = 0.421; Table 2).

Regarding school dynamics, the results showed that the Nightingale project was associated with more positive attitudes of the mentees towards the school and their peers, specifically for the over-twelves group, who significantly improved their response to the item: "I would prefer to go another school" (*Z* = 2.834; *p* = 0.005; *dCohen* = 0.664; Table 2).

This connection has helped those adolescents who participated in mentoring to improve academically (Table 2). The pre-test–post-test shows an improvement in the Academic Self-Efficacy Index for the participants as a whole (*Z* = −7.693; *p* = 0.000; *dCohen* = 1.548). This improvement is greater in the group of boys (*Z* = −5.905; *p* = 0.000; *dCohen* = 1.778) and in the children under 12 (*Z* = −6.631; *p* = 0.000; *dCohen* = 2.308).

With regard to self-esteem, no changes were observed between pre-test and post-test for the whole sample (*Z* = 0.766; *p* = 0.444; Table 2), nor for boys and girls separately. Statistical significance was only observed in those over twelve years of age, whose self-esteem increased (*Z* = 1.968; *p* = 0.049; *dCohen* = 0.448). However, when controlled for age (under 12 years and over 12 years), mentoring was positively associated with the adolescents' assessment of their abilities and degree of personal satisfaction. This improvement in self-concept occurred in the oldest participants in the positive responses to the items used in the questionnaires: "I feel that I am a person of worth, at least on an equal level with others" (Z = −2.466; p = 0.014; dCohen = 0.569) and "I am able to do things as well as most other people" (*Z* = −2.376; *p* = 0.018; *dCohen* = 0.547). This correlates with an improvement in the Cognitive Engagement Index score for the children under 12, the only group in which there are statistically significant differences between the pre-test and post-test scores (*Z* = −2.088; *p* = 0.037; *dCohen* = 0.054; Table 2).
