**4. Results**

### *4.1. Findings from Quantitative Survey Data*

All results, including descriptive statistics, are presented in Table 2.

**Table 2.** Participation in the mentoring program and well-being outcomes: descriptive statistics, repeated measures ANOVA with between-subject (group) effect, and paired *t*-tests per group.


*Note.* Mentoring group: *n* = 21, control group: *n* = 23. *M* and *SD* represent means and standard deviations, respectively. <sup>1</sup> ANOVA contrasts in the following order: 1 = group's effects; 2 = time' effects; and 3 = interaction effects (group \* time). To calculate Cohen's *d*, we used a procedure described in Morris and De Shon (2002, p. 111), who suggest estimating the effect size for single-group pre-test–post-test designs by taking the correlation between the pre- and post-test into account. Statistically significant effects are in bold and marginally significant ones are in italics.



**Table 3.** Participation in the mentoring program and educational aspirations: the McNemar test.

*Note.* We repeated these analyses excluding participants who responded "I don't know" to the question about educational aspirations. In this case, the McNemar test was also statistically significant in the mentoring group (*p* = 0.016), whereas it was non-significant in the control group (*p* = 0.453), with 18 participants in each group.

**Table 4.** Participation in the mentoring program and educational expectations: McNemar test.


*Note.* We repeated these analyses excluding participants who responded "I don't know" to the question about educational expectations. In this case, the McNemar test was also statistically significant in the mentoring group (*p* = 0.016), whereas it was non-significant in the control group (*p* = 0.453), with 18 participants in each group.

#### *4.2. Findings from Interview Data*

The qualitative results concerning the main topics of the research (psychological wellbeing and educational futures) are shown below. Through the analysis of the interviews carried out, we highlighted different types of social support that the youths perceived from their mentors (and that the mentors mentioned that they offered), which have had a certain impact in terms of well-being and on the decisions taken regarding what educational path to follow. In addition, we discuss the absence of certain types of support in the control group, which enabled us to understand the differences between the groups (mentored and non-mentored).

In order to suggest how the mentoring programmes can promote the acquisition of an effective support network, we highlight, in the final section of the results, how the programme guides the task of the mentors. Specifically, we focus on how the support provided by the mentors is focused on the needs of the young people due to training and the programme's exhaustive monitoring of each relationship.
