"Living Adolescence in Family" parenting program: Adaptation and implementation in social and school contexts

The participants from social services in the intervention group were 438 parents, with a majority of mothers (86.8%), an average age of 42.32 years, a level of risk that is divided into normal-low risk (58.4%) and medium-high risk (41.6%) (see Table 1 for further details). Participants in the control group were 259 parents who expected to take part in the program for several editions, and whose socio-demographic composition did not differ from that of the intervention group. Participants to the program were referred by the local social services or day centers for being at pre-risk or risk conditions. The geographic areas where the program was developed were Castilla la Mancha, Lleida, Tenerife and Gran Canaria.

The participants from the school centers were 1283, from which 1040 were mothers (81.1%) and 243 participants were fathers (18.9%) (see Table 1 for further details). They came from each one of the Canary Islands, although the majority of the participants belonged to Gran Canaria (40.8%) and Tenerife (42.2%). Participants in the intervention group were 880 parents. Participants in the control group were 403 parents who could not be a part of the program for several reasons, although their children were at risk of early school dropout. As for their socio-demographic profile they did not significantly differ from the intervention group.

Table 2 shows the profiles of the 83 facilitators that implemented the program in the local social services. The majority of the facilitators were women (84.7%) with an average age of 28.26 years; a 59.1% of facilitators have a psychosocial academic background (psychology, social worker, social educator, and sociology) and they all possess an average of 4.36 years of previous experience in family work (see Table 2 for further details).

The teachers’ profile who implemented the program as facilitators in the school centers is shown in Table 3 . There were 453 facilitators, since some groups made use of two facilitators. The majority of the facilitators were women (74.7% for facilitator 1 and 87.2% for facilitator 2), with an average age of 44 years (facilitator 1) and 47 years (facilitator 2). In relation to the teacher's role in the center, the commonest role was the facilitator 1 (50%) and the teacher as a tutor in facilitator 2 (44.8%) (see Table 3 for further details). On the whole, all facilitators attended the initial training workshop.

This section describe the instruments that have been used in the local social services and the school centers, though in both cases other instruments were also used.

• -

  Family's socio-demographic aspects : Facilitators reported on the socio-demographic profile of the participants: sex, age, residence place, number of children, couple status, family structure, individual's and partner's level of education, employment status, and economic status. In the case of the families coming from school centers, these aspects were reported by the participants themselves.

• -

  Facilitators’ identifying profile : Facilitators reported on their profile: age, sex, academic degree, workplace, and years of experience in family intervention. Exclusively for the facilitators (teachers) from the school centers, data about employment status at the center, years of teaching experience, and years of experience as a teacher in the center itself were also reported.

• -

  Questionnaire on parental supervision in adolescence (Stattin & Kerr, 2000 ). It is composed of 16 items with a Likert response scale (1 = nothing to 5 = always). As for the Spanish adaptation, the statistical program Mplus was used, from which four factors were obtained, and whose factorial loadings oscillated between .431 and .912. Considering the goodness of fit indexes were satisfactory: CMIN = 3.45; RMSEA = .07; CFI = .980; TLI = .960 and WRMR = .610. The following factors were obtained: Control (5 items) that analyzes strategies for the strict control of pernicious behavior for adolescent's well-being (e.g.,“Your child should ask for permission to go out at night, before planning with their friends”); Educational monitoring (3 items) that analyzes parents’ knowledge about the adolescents’ school activities (e.g., “Knows the school tasks that their child brings from school to work at home); Leisure monitoring (4 items), which evaluates parents’ knowledge about the adolescents’ leisure activities (e.g.,“Knows what their child does at leisure time”); Self-disclosure (4 items) which evaluates the grade of spontaneity that is used by adolescents when telling about personal experiences from daily life to their parents and their whereabouts (e.g., “Your child tells you their secrets about what they do during free time”). The overall reliability of the instrument were α = .94, oscillating between α = .73 and α = .96.

• -

  Situational questionnaire for strategies and goals for family conflict resolution ( García, 2008 ). This instrument explores the strategies and goals used, reported on a Likert scale (1 = totally disagree to 5 = totally agree) in six daily situations of conflict between parents and adolescents. The version for parents was composed by 24 items for each kind of conflict (4 for each situation). The integration or negotiation style implies fitting parents’ and adolescents’ point of view in the conflict resolution through dialog and manifestation of needs, indicating the limit points for reaching agreements for both sides. The dominant style consists of taking just their own point of view for conflict resolution into account, using power or negative feelings in order to achieve their own personal desire. The avoidant style consists of solving the problem by using indifference or calm, leaving their own point of view at second place. The reliability of the instrument was α = .91, oscillating between α = .70 and α = .95.

In both contexts a series of intensive training workshops was carried out that were organized by the coordinators of the program and where there was a discussion about the set of contents, the methodology and the implementation process of the program. After that, the groups were formed and the program started. At the same time, monitoring sessions with the facilitators took place, so as to gather information about the groups, solve doubts, etc. Eventually, a final session to close the program with the delivery of certificates and a farewell party with the participant families was carried out. Afterwards, a day with the facilitators was celebrated in order for them to get some feedback about the results of the program.

For the first aim, repeated ANOVAs were performed to examine differences in the analyzed dimensions between the pre-test and post-tests measurements; and in respect to the control group, using the F statistic for the interpretation of the meaning of contrast. The effect size was explored through the R 2 statistic, whose clinical relevance is considered to be negligible when R 2 < .01, low when R 2 > .01 and R 2 < .09, medium when R 2 > .09 and R 2 < .25, and high when R 2 > .25 (Cohen, 1988 ). The same kind of analysis was carried out to calculate the modulation effect of the pre-post changes according to the families’ socio-demographic characteristics and the facilitators’ profile, both in the social services and in the school centers. In those cases in which these factors had more than two levels, post hoc analysis was done, assuming equal variances through Scheffé. The statistical analysis has been made with the SPSS statistics software, version 21.

In the local social service context, significant differences were obtained in every analyzed variable with low and medium effect sizes ( Table 4 ). In relation to parental supervision, participants have reported an increase in education and leisure monitoring as well as in self-disclosure; they also claim having decreased the degree of strict control. The results were also significant when the post-test of the intervention group was compared to the initial measure of the control group that was in a waiting list to start it. In comparison with the control group, the intervention group decreased their use of strict control ( F(1,696) = 4.02, p = .045, R 2 partial = .02 ), whereas the use of educational monitoring increased ( F(1,696) = 25.18, p = .000, R 2 partial = 1.0) and the self-disclosure (F(1,696) = 14.12, p = .000, R 2 partial = .09 ) also increased, with low and medium clinical relevance. As for the styles of conflict resolution, compared to the control group, the intervention group reported a decrease for the dominant and the avoidant style, as well as an increase on the integration style with low and medium clinical relevance. The changes affecting the control group have been noticeable, for the integration style ( F(1,696) = 19.11, p = .000, R 2 partial = .09); the dominant style (F(1,696) = 5.61, p = .018, R 2 partial = .04), and the avoidant style ( F(1,696) = 3.472, p = .05, R 2 partial = .02 ), indicating more positive findings for the intervention group, with low to medium clinical relevance.

The results from the participants in the school centers indicate that significant differences have been obtained in several analyzed variables, showing positive changes ( Table 5 ). In relation to parental control, participants reported an increase in the use of leisure monitoring and self-disclosure, with low and medium clinical relevance. There was no significance difference in educational monitoring or in strict control. With respect to the styles of conflict resolution, a decrease in the dominant style and increase of the integration style and in the avoidant style, with low to medium clinical relevance were found ( Table 5 ). The control group did not differ from the intervention group in the pre-test measures and there were no significant differences between the pre-test and the post-test measures in this group.

The age of parents influenced the self-disclosure ( F(1,392) = 2.33, p = .05 ), to such an extent that the older the parents were the more open-minded their adolescents were reported to be, after participating in the program (pre-test: M = 3.21, DT = .92; post-test: M = 3.42, DT = .97). In relation to family types ( F(1,436) = 3.46, p = .045), single-parent families showed the highest rate of adolescents’ disclosure after the program (pre-test: M = 3.24, DT = .86; post-test: M = 3.50, DT = .90); with respect to two-parent families (pre-test: M = 3.20, DT = .96; post-test: M = 3.30, DT = 1). The academic level seemed to influence on control (F(1,430) = 3.67, p = .05 ), since the parents without primary studies were the least inclined toward using control after the program ( pre-test: M = 4.03, DT = .95; post-test: M = 3.77, DT = .98 ), in comparison to parents with studies (pre-test: M = 3.87, DT = .91; post-test: M = 3.79, DT = .91).

The academic level influenced on the integration style ( F(1,287) = 3.76, p = .05), since the participants whose partners possessed academic titles seemed to be the most inclined for this style (pre-test: M = 3.46, DT = .65; post-test: M = 3.69, DT = .54), in relation to those parents whose partners did not possess any academic certificate (pre-test: M = 3.65, DT = .68; post-test: M = 3.72, DT = .60). As for the dominant style, the employment situation made a significant impact on the results ( F(1,429) = 4.05, p = .045), since the participants who were employed showed the lowest rate, regarding the use of this style (pre-test: M = 2.75, DT = .58; post-test: M = 2.54, DT = .51), with respect to those who did not have any job (pre-test: M = 2.75, DT = .71; post-test: M = 2.70, DT = .73). Finally, the family type showed interesting findings for the avoidant style ( F(1,436) = 3.85, p = .05), since single-parent families decreased the use of this style (pre-test: M = 2.79, DT = .61; post-test: M = 2.66, DT = .56), whereas changes were not significantly found in two-parent families (pre-test: M = 2.66, DT = .56; post-test: M = 2.65, DT = .56).

The changes in self-disclosure was moderated by the sex of the facilitators in the group ( F(2,305) = 3.90, p = .021); in this way the major increases were for those groups where facilitators were men and women (pre-test: M = 2.67, DT = .99; post-test: M = 3.28, DT = 1.01), in comparison to those groups whose participants were exclusively women (pre-test: M = 3.31, DT = 1; post-test: M = 3.47, DT = 1.07).

As for the integration style, the age of the facilitator was a relevant aspect ( F(1,436) = 4.57, p = .033), as it was observed that families attending groups with younger facilitators were the ones who increased their use most (pre-test: M = 3.52, DT = 0.65; post-test: M = 3.79, DT = 0.66), in comparison to the older facilitators (pre-test: M = 3.59, DT = 0.65; post-test: M = 3.71, DT = 0.52). As regards the dominant style relevant differences have been found, considering the position of the facilitator ( F(2,272) = 5.87, p = .003), and so, the families whose facilitators belonged to associations that have been hired by the service were the least inclined for this style after the program (pre-test: M = 2.76, DT = 0.44; post-test: M = 2.44, DT = 0.46), in relation to those facilitators who came from other unknown services (pre-test: M = 2.58, DT = 0.70; post-test: M = 2.47, DT = 0.67). The education of the facilitator also influenced on the dominant style ( F(2,434) = 13.60, p = .000); thus, families with facilitators who were trained in psycho-social field decreased more their support for this style (pre-test: M = 2.71, DT = 0.67; post-test: M = 2.57, DT = 0.66), in relation to those who were trained in education fields (pre-test: M = 2.84, DT = 0.70); post-test: M = 2.73, DT = 0.69). Finally, the experience of the facilitator also influenced on the dominant style ( F(12,249) = 3, p = .001), since the families with well-experienced facilitators showed the lowest rate of the dominant style after the program (pre-test: M = 2.65, DT = 0.66; post-test: M = 2.58, DT = 0.69). Lastly, the education of the facilitator influenced on the use of the avoidant style ( F(2,435) = 7.10, p = .001), the families being those whose facilitators held a degree in the psycho-social field, the ones who most diminished its use after the program (pre-test: M = 2.66, DT = 0.56; post-test: M = 2.56, DT = 0.51), in comparison to those facilitators who had a degree in education (pre-test: M = 2.78, DT = 0.59; post-test: M = 2.73, DT = 0.57).

As for the parental control dimensions, we do not find any significant difference in any socio-demographic variable. However, in the strategies for problem-solving, we find that in the avoider style the age become a relevant factor ( F(1,347) = 17.57, p = .01), in such a way that the mothers aged 46 years old or over claimed to use this style most of the time (pre-test: M = 2.77, DT = .83; post-test: M = 3.1, DT = .91) in respect to younger mothers aged 27–39 years old pre-test: M = 2.7, DT = .83; post-test: M = 2.95, DT = .83).

The years of teaching experience showed to be relevant for the variable of self-disclosure ( F(2,293) = 3.39; p = .03 ), the families being those whose facilitators had 3–10 years of teaching experience the ones who reported less changes in disclosure after participation in the program (pre-test: M = 4.6, DT = 1.09; post-test: M = 4.8, DT = .85), in relation to those facilitators with 11–16 years of teaching experience whose families reported major changes (pre-test: M = 4.8, DT = .95; post-test: M = 5.10, DT = .60).