Introduction

All children and adolescents have basic needs that, in most cases, are met in their family environment. However, some of them experience a lack of care from their parents and suffer situations of adversity and abuse throughout their childhood, which generate a serious threat to their welfare (^{Norman et al., 2012}). To deal with this type of highly serious situations in Spain, the Child Protection System (CPS) resorts to different child and adolescent protection measures (^{Amorós & Palacios, 2004}). In this regard, our legislative framework gives priority to family-based measures, as opposed to residential care (^{Law 26/2015}, on the modification of the Child and Adolescent Protection System). This is carried out on the basis of international recommendations, which advocate in favor of the growth of children in a family environment (^{Palacios et al., 2019}), and the abundant empirical evidence that confirms the negative consequences in the short, medium and long-term resulting from institutionalization (^{Berens & Nelson, 2015}; ^{Palacios, 2003}; ^{Palacios et al., 2019}; ^{Woodhouse et al., 2018}). Despite this, there are still 16,991 children and adolescents living in protection centers in Spain (^{Ministry of Social Rights and 2030 Agenda, 2021}). However, based on the above assumptions, it should be taken into account that residential care in our country is currently intended to fulfill a very specific role, being configured as a highly specialized measure. Thus, in recent years, the age of children in residential foster care has gradually risen, being a measure that is mainly used when there is no availability of families to practice a family foster care; to a lesser extent, when it is not desired by young people who enter the system at older ages (for example, cases of parental control incapacity); and, especially, for unaccompanied migrant youths (^{Bravo & Del Valle, 2009}; ^{Bravo et al., 2022}; ^{Ministry of Social Rights and 2030 Agenda, 2021}; ^{Poole et al., 2021}).

Residential care, as a protection measure, must therefore ensure that the safety and needs of children are met, promoting their comprehensive development (^{Consejería de Inclusión Social, Juventud, Familias e Igualdad, 2022}). However, nowadays, on the one hand, a majority of children with the profile described above are characterized by a high incidence of emotional and behavioral distress, as well as other difficulties such as the prospect of emancipation without returning to the family home when reaching legal age (^{Bravo et al., 2022}; ^{Del Valle et al., 2011}). On the other hand, despite the improved qualities that have been developed and continue to be introduced in protection centers in Spain and other countries (^{Del Valle & Bravo, 2013}; ^{Whittaker et al., 2016}, ^{in press}), these do not always have the necessary human resources to sufficiently cover the affective and stimulation needs of these children (^{Palacios, 2003}; ^{UNICEF & Eurochild, 2021}). Under these circumstances, it is common for children in residential care to present more behavioral, emotional, cognitive, and social problems, compared to their community peers and to those living in family-based protection measures, such as foster care or adoption (^{Campos et al., 2019}; ^{Fernández-Molina et al., 2011}; ^{Fonseca-Pedrero et al., 2011}; ^{Jiménez-Morago et al., 2015}).

Psychological adjustment, understood as the person's ability to function adequately in their environment through their cognitive, behavioral, emotional, and social resources (^{Schoeps et al., 2019}), is one of the areas in which children in residential care show greater difficulties with respect to their community counterparts and other children in family-based measures (^{Amorós & Palacios, 2004}; ^{Fernández-Molina et al., 2011}; ^{Nowacki & Schoelmerich, 2010}). This greater psychological maladjustment occurs even when they start with similar levels of adversity (^{Cousins et al., 2010}; ^{Jiménez-Morago et al., 2015}; ^{Juffer & Van IJzendoorn, 2009}; ^{Palacios et al., 2013}). On the one hand, difficulties in this aspect among minors in residential care are related to the presence of behavioral problems, including high levels of aggression and deficits in impulse control (^{Del Valle et al., 2011}; ^{Fernández-Molina et al., 2011}; ^{Heflinger et al., 2000}). On the other hand, some studies have evidenced emotional alterations or symptoms, such as mood and anxiety disorders, sadness, anger, irritability, depression or inhibition (^{Fernández-Daza & Fernández-Parra, 2017}; ^{Heflinger et al., 2000}; ^{Simsek, et al., 2007}). In addition, other authors point out the greater propensity of these children to present difficulties related to hyperactivity and attention deficit (^{Barroso et al., 2018}; ^{Del Valle et al., 2011}). At the social level, difficulties among this population are reflected, for example, in the development of problems when relating to peers in the protection center and school context (^{Balluerka et al., 2013}; ^{Fernández-Daza & Fernández-Parra, 2017}; ^{Martín et al., 2012}), as well as in a tendency for low self-esteem and poor academic motivation, showing, in general, a notably negative view of themselves (^{González & Cortés, 2015}; ^{Greger et al., 2016}).

The scientific evidence on the psychological adjustment of children in residential care also points to the fact that some results vary according to the informant, differentiating between the vision that the children have of themselves and that of the adults caring for them. It is for this reason that some authors consider that child and adolescent behavioral assessment should have multiple perspectives, preferably reporting on different contexts (^{Achenbach, 2006}; ^{Janssens & Deboutte, 2009}; ^{Vostanis, 2006}). Thus, several studies have found low and moderate levels of agreement between children and their residential caregivers when assessing their externalizing and internalizing problems, with the latter informants identifying more problems and there being more agreement for externalizing than internalizing ones (^{Gearing et al., 2014}; ^{Sainero et al., 2015}). In the same direction, ^{Delgado et al. (2012)} evaluated the emotional and behavioral problems of a group of children in residential care, finding that both residential caregivers and teachers observed higher levels of externalizing problems than those perceived by the children themselves.

Also, some research has focused on exploring the variables related to the psychological adjustment of children in residential care. Several studies have analyzed their association with certain sociodemographic characteristics, such as gender or age of the children. As for gender, there is no agreement among the different studies, so that some authors have found that boys show more problems in their psychological adjustment than girls (^{Fernández-Daza & Fernández-Parra, 2017}; ^{González & Cortés, 2015}; ^{Janssens & Deboutte, 2009}) and other authors point out the opposite (^{Jones et al., 2007}; ^{Rodrigues et al., 2019}). The only aspect in which there seems to be coincidence among the aforementioned studies is that girls present better prosocial behavior than boys. Regarding age, there is no consensus among different researchers either, as, on the one hand, Rodrigues et al. (2019) find that younger adolescents are those who show more difficulties in their psychological adjustment, while authors such as ^{Hoffnung-Assouline and Attar-Schwartz (2020)} as well as ^{Palacios et al. (2013)} do not find any relationship between age and psychological adjustment. In addition, other authors have associated the psychological adjustment of children in residential care to some variables related to their past and present trajectory in the CPS. Thus, a significant unfavorable impact of longer durations of institutionalization on children’s psychological adjustment has been pointed out (^{Delgado et al., 2012}; ^{Martín et al., 2007}; Palacios, 2003; ^{Woodhouse et al., 2018}).

In short, throughout the previous paragraphs, the negative and limiting consequences of residential care for children have been highlighted. Due to the fact that, despite this, the practice of institutionalization is still abundant in our country, for the reasons stated in the first paragraphs of this section, we consider it essential to research, disseminate and use programs such as Collaborating Families, developed in Andalusia, in order for children in residential care to benefit from them (^{León et al., 2019}, 2021). The Collaborating Families program is a complementary resource to residential care that consists of families who, in an altruistic and supportive manner, commit, with stable expectations, to share moments of leisure with children or adolescents from protection centers during different periods of time, such as weekends, vacations and/or holidays (^{Consejería de Inclusión Social, Juventud, Familias e Igualdad, 2022}a). This resource lacks specific regulation in state legislation, being only indirectly alluded to in the aforementioned Law 26/2015, in its article 172. In Andalusia, on the other hand, it has been explicitly included for the first time in the recent ^{Law 4/2021 of July 27}, on Childhood and Adolescence, in its article 110, under the figure of Social Collaboration.

The main purposes of the Collaborating Families program in Andalusia are based on allowing children in residential care to participate in alternative activities to those of the protection center; providing them with an environment where they can learn to relate in a healthy way and live together as a family; getting them to establish healthy emotional bonds with their collaborators; and expanding their social support network with significant adults who offer them individualized treatment (^{Consejería de Inclusión Social, Juventud, Familias e Igualdad, 2022}a; ^{Delgado & López, 2016}; ^{León et al., 2021}). To participate in the program, a series of technical criteria are required to be met, both by children susceptible to collaboration and by interested families (^{Consejería de Inclusión Social, Juventud, Familias e Igualdad, 2022}a). Regarding children and adolescents, they must meet the following characteristics: a) be in an abandonment situation; b) be in residential care with an intervention plan of long-term placement in a protection center; c) be older than 7 (except for very extreme situations of special needs); d) accept to participate in the program; e) benefit from the program (León et al., 2019).

Currently, there is very little research in Spain on the figure of Social Collaboration. One of the first studies in this field explored the experience of children with their collaborating families, finding that they rated the relationship with their collaborators very positively and expressed high satisfaction with the collaboration (^{Espinosa et al., 2022}; ^{Gallardo et al., 2020}). The data from this study also indicate that children in collaboration rate the time their collaborators spend with them, the fair treatment they receive from them and the possibility of talking to them when they need to very positively. In addition, they reported that they perceived the support of their collaborating families in any situation in which they need it. Likewise, ^{Cano-López et al. (2021)} interviewed representatives of different foster care and residential care collaboration associations in Andalusia and concluded that one of the main benefits of the figure of Family Collaboration is the individualized emotional support they provide to children, and the personalized accompaniment and increased support they have until adulthood, and even after they come of age. According to ^{Ferreira et al. (2020)}, referring, in general terms, to children in residential care, the social support they may perceive is identified as an enhancer of positive psychological functioning and as a buffer against psychological problems.

This work aims to continue expanding knowledge about the figure of Social Collaboration and what it means for the welfare of children in residential care. To this end, our first objective was to describe the psychological adjustment of the children and adolescents participating in the Collaborating Families program, both from their own perspective, that of their collaborating families, and that of their professionals of reference in the protection centers, comparing the information obtained from different sources of information. With respect to this first objective, our main hypothesis is that the professionals' assessment of the psychological adjustment of the participants is, on average, more negative than that of the rest of the informants, i.e., that they will rate a higher number of difficulties. As a second objective, we compared the psychological adjustment of the children participating in the Collaborating Families program with that of a group of peers in residential foster care, but who did not benefit from this program. This comparison was made from two perspectives, that of the children themselves and that of the professionals. In this case, our main hypothesis with respect to the second objective is that the average psychological adjustment difficulties of the participants in the comparison group will be higher than those of the group in family collaboration. Our third objective was to analyze the self-perception of psychological adjustment of participants with collaborating families in relation to some sociodemographic variables and variables related to their past and present history in the CPS. In reference to this third objective, while we do not pose any specific hypotheses regarding the gender or age of the children, we do expect that the level of psychological adjustment difficulties will be positively related to the time of institutionalization. Finally, the fourth objective of the present study was to explore the association between the self-perceived psychological adjustment of participants with collaborating families and different dimensions related to their experience of family collaboration: the social support they perceive from their collaborating families, their assessment of their relationship with them, and their satisfaction with the experience of family collaboration. As a hypothesis about the fourth and last objective, we expect that psychological adjustment difficulties will be less elevated as participants better value these aspects related to family collaboration.

Method

Participants

The sample consisted of 185 participants living in Andalusia, Spain: 37 children and adolescents in residential care, who at the time of the study were participating in the Collaborating Families program, their 37 reference professionals in the protection centers, their 37 collaborating families, as well as a comparison group composed of 37 children and adolescents in residential care who were not participating in the Collaborating Families program and their 37 reference professionals in the protection center (for more information, see ^{Espinosa et al., 2022} and ^{León et al., 2021}).

Table 1 shows the sociodemographic characteristics of the children with collaborating families and their peers in the comparison group.

Table 1. Differences between the characteristics of children with cooperating families and their peers in the comparison group. 

^Note.CF= Collaborating Families; CG= Comparison Group; AR= Adjusted Residual.

The participants with collaborating families had been in residential care for a mean of 8.1 years (SD = 2.89). Some 27.03% had been in only one protection center, while 37.84% had been in two, 29.73% in three and 5.40% in four. The average duration of their participation in the Collaborating Families program was over 4 years (M = 4.32; SD = 2.98).

On the other hand, both the professionals of the children with collaborating families and those of the comparison group had an average age of around 38 years (M = 38.94, SD = 8.65; M = 38.03, SD = 8.49, respectively) and had been working in their profession for an average of 10 years (M = 10.89, SD = 7.36; M = 10.60, SD = 7.26, respectively). Likewise, they had been working at the protection center for an average of 7 to 8 years (M = 8.25, SD = 5.78; M = 7.67, SD = 7.08, respectively).

In reference to the collaborating families, the male collaborators had a mean age of 46 years at the time of the study (M = 46.88, SD = 9.31), while the female collaborators had a mean age of 44 years (M = 44.78, SD = 8.91). The 87.4% percent of the female collaborators were working, as were 96.9% of the male collaborators. Likewise, 27% of the collaborating families were single-parent, while 73% of them were two-parent families. Of all the families, 81.1% had children.

Results

In response to the first objective, Table 2 shows the descriptive data of central tendency and dispersion related to the psychological adjustment of the children with collaborating families, evaluated by themselves, their referring professionals, and their collaborators. Table 2 also shows the distribution of the scores in the different ranges contemplated by the SDQ instrument (^{Goodman, 1997}), expressed in percentages.

As for the mean scores obtained (Table 2), from the perspective of the children themselves, these were within the normative range in all subscales and in the total scale of difficulties. On the other hand, from the data provided by the collaborating families, the mean scores were again within the normative range, both in the total scale of difficulties and in almost all subscales, the exception being the subscale of peer problems, which was in the borderline range. However, from the perception of the professionals, the mean scores were in the lower cut- off of the borderline range for the total difficulties and for the subscales of conduct problems and peer problems, while the rest of the subscales were within the normative range.

Table 2. Descriptive data and distribution of psychological adjustment scores of children with collaborating families. 

^Note.CF = Collaborating Families.

To evaluate the differences between the perceptions of the different informants, we ran contrasts of means for related samples, finding statistically significant differences and considerable effect sizes (^{Castro & Martini, 2014}; ^{Cohen, 1988}) in three subscales: conduct problems, peer problems and prosocial behavior. First, these significant differences were found between the assessment of the collaborating families and that of the professionals with respect to the conduct problems subscale (t(35) = -2.044, p = .048, d = .374). In this case, it was the latter who perceived more problems of this nature. In addition, with regard to peer problems, statistically significant differences were found between the perception of the children and that of the collaborating families (t(36) = -2.114, p = .042, d = .392), and between that of the children and that of their reference professionals (t(35) = -2.555, p = .015, d = .543). With respect to the latter differences, it was, on the one hand, the collaborating families and, on the other hand, the professionals who perceived more problems. In reference to prosocial behavior, we obtained statistically significant differences between the scores of the collaborating families and those of the professionals (t(35) = -3.686, p = .001, d = .898), with the collaborating families being those who perceived more prosocial behavior. Likewise, in this subscale of prosocial behavior, significant differences were found between the perception of the children and that of the professionals (t(35) = 3.494, p = .001, d = .771), in favor of the former.

As a second objective, we wanted to know whether there are differences between the psychological adjustment of participants with collaborating families and that of a group of peers in residential care who do not have collaborating families, both from their own perspective and that of their corresponding reference professionals. First, to respond to this second objective, Table 3 shows the descriptive data of central tendency and dispersion of the psychological adjustment of both groups, according to the children's own perspective and that of the professionals, as well as the distribution of the scores according to the ranges of the SDQ instrument, again expressed in percentages.

Table 3. Descriptive data and distribution of psychological adjustment scores of children with and without collaborating families and of reference professionals in both groups. 

^Note.CF = Collaborating Families; CG = Comparison Group.

With regard to the comparative analysis between the psychological adjustment of the participants with and without collaborating families, carried out by contrasts of means for independent samples, no statistically significant differences were found between the mean scores obtained from the children themselves, nor in those obtained from the professionals. However, the effect size was small, but considerable, for the comparisons made about emotional problems in the two groups of informants (d = .373, p = .113, in children; d = .410, p = .087, in professionals) and for total difficulties (d = .392, p = .095, in children; d = .318, p = .186, in professionals), in favor of the group with collaborating families (^{Castro & Martini, 2014}).

As for the third objective, we analyzed the possible relationships between the different subscales and the total scale of self-reported psychological adjustment of the children in collaboration with sociodemographic variables (age, gender and presence or not of disability or chronic illness) and their trajectory in the CPS (age of abandonment, time in the CPS, duration of institutionalization, age at the beginning of collaboration, and duration of collaboration). In no case were statistically significant differences or correlations found, obtaining a p > .05 in all cases.

Finally, as a fourth objective, we analyzed the relationship between the children’s self- perception of their psychological adjustment and their experience with the collaborating families. Regarding self-report about the support, they perceive from their collaborating families, we focused on three measures (The KIDSCREEN Group, 2004): the perception of fair treatment (M = 4.29, SD = 1.25), the feeling of acceptance (M = 4.80, SD = .47) and the perception of concern (M = 4.80, SD = .47). When correlating these measures and the children’s psychological adjustment, we found that those who presented fewer conduct problems also felt that their collaborating families had treated them more fairly (r = -.342, p < .05). In turn, we found that those who showed fewer peer problems also felt more acceptance from their collaborating families (r = -.380, p < .05) and perceived that their families cared more about them (r = -.380, p < .05).

In reference to the evaluation of the relationship with their collaborating families, they were asked to rate it in general (^{Cantril, 1965}), obtaining a mean score of 8.94 (SD = 1.83). When executing the correlations with the children’s self-perceived psychological adjustment, it was found that the better the rating they made of the relationship with their collaborating family, the fewer behavioral problems they presented (r = -.336, p < .05).

Finally, participants' satisfaction with their collaborating families was evaluated based on three measures from the interview: their assessment of the time spent with their collaborating family (M = 4.67, SD = .96), the feeling that their collaborating family was happy with them (M = 4.83, SD = .38) and the satisfaction, in general, with the collaboration (M = 4.58, SD = .94). When correlations were made between these measures and the children’s psychological adjustment, it was found that the fewer conduct problems they presented, the more they valued the time with their collaborating families (r = -.518, p < .01) and the more satisfied they were with them (r = -.506, p <.01). In addition, those who felt that their collaborating families were happier with them, also presented fewer conduct problems (r = -.400, p < .05) and total difficulties (r = -.396, p < .05).

Discussion

The first objective of our study was to describe the psychological adjustment of a group of children and adolescents with collaborating families, from the perspective of multiple informants. On the one hand, the children in our sample seem to perceive themselves as having few psychological adjustment difficulties, as their mean scores are within the normative range in all subscales and in the total difficulties scale. Moreover, considering the distribution of their scores, most of them are within this normative range, again in all subscales and in the total difficulties scale. These results are considerably more positive than those obtained by other authors who have assessed self-perceived psychological adjustment in residential care (^{Rodrigues et al., 2019}). As for the collaborating families, unlike the previous informants, they place the mean score of peer problems in the borderline range and point out that slightly less than half are in the normative range on this subscale. In a similar line, the reference professionals of these children again place the mean score of the subscale peer problems in the borderline range and also place the mean of conduct problems and total difficulties in this range. In the case of the professionals, it is also noteworthy that they rate approximately one third of the participants with difficulties in the clinical range, if we look at the total difficulties scale. In comparison with previous studies that include the assessment these professionals about the psychological adjustment of children in residential care, our data are similar or more positive (^{Cousins et al., 2010}; ^{Jiménez-Morago et al., 2015}; ^{Palacios et al., 2013}). In the study by ^{Jiménez-Morago et al. (2015)}, on children aged 4 to 10 years, the mean of total difficulties was higher, although it is still in the borderline range. However, in the work of ^{Cousins et al. (2010)}, where a sample between 10 and 15 years of age was evaluated, this mean exceeded 16 points, placing it in the clinical range.

When comparing the different informants, it seems evident that the children’s view of their own psychological adjustment is more positive than that of their collaborating families and, in turn, the assessment of these families is more favorable than that shared by their reference professionals. Specifically, the professionals highlight the greater presence of conduct problems, compared to the information provided by the collaborating families. A more negative view by professionals from protection centers has also been previously noted in other studies (^{Delgado et al., 2012}; ^{Gearing et al., 2014}; ^{Rodrigues et al., 2019}; ^{Sainero et al., 2015}). It is possible that professionals are more sensitive to behavioral problems than the families or the children themselves and, as they spend more time with these children and can compare with a larger volume of children, they can make a more refined assessment (^{Sainero et al., 2015}). On the other hand, it seems that the collaborating families and professionals coincide in a higher score on peer problems, compared to the self-assessment of children. Although it is true that some authors warn of a greater social desirability in adolescent self-reports, towards a more positive representation of themselves (^{Fan et al., 2006}), it is equally possible to make a positive reading of these results. Thus, it could be indicating that the collaborating families are being able to identify the problems that these children are having when it comes to relating to others. This is a very important starting point for them to work towards one of the purposes of this program, mentioned above, which is to give them the opportunity to learn to relate to each other in a healthy way (^{Consejería de Inclusión Social, Juventud, Familias e Igualdad, 2022a}).

Furthermore, in reference to the strengths of these children, our data point out that both themselves and their collaborating families score better on prosocial behavior, compared to the reference professionals. Although we insist on the idea that the latter are likely to have a deeper knowledge of the children in a wider variety of situations and contexts, it is beneficial for these children to have people who highlight their strengths and refer to them in a positive way. These more favorable data are in line with previous studies on these collaborating families, in which they indicated feeling, for the most part, very satisfied with the evolution of the children and with their adaptation to the family (^{León et al., 2021}).

Regarding the second objective of this study, the data obtained imply that we should speak of similarities rather than differences when comparing psychological adjustment between participants with and without collaborating families. This similarity occurs in other areas explored in the lives of these children (^{Espinosa et al., 2022}) and is to be expected if we consider that both participants are in residential care, and therefore share most of their time in the protection center under the care of their residential caregivers. However, our data indicate a tendency for the group with collaborating families to present fewer psychological adjustment difficulties. Thus, considering the effect size data, we would expect to find significant differences in psychological adjustment, in favor of the group with collaborating families, in the case of a larger sample size, both in the children’s and in their reference professionals’ reports. In fact, while the means and distributions in the group with collaborating families are more positive than in previous studies (^{Cousins et al., 2010}; ^{Jiménez-Morago et al., 2015}; ^{Rodrigues et al., 2019}), the data for the group without collaborating families are similar to the data of these investigations and even somewhat more negative when we refer to conduct problems (Rodrigues et al., 2019). Previous studies have highlighted the existence of more difficulties among those in residential foster care, compared to those who benefit from other family alternatives, such as foster care (^{Amorós & Palacios, 2004}; ^{Jiménez-Morago et al., 2015}; ^{Nowacki & Schoelmerich, 2010}).

With reference to the third objective, based on the lack of consensus found among previous studies, our results are related to those presented by ^{Hoffnung-Assouline and Attar- Schwartz (2020)} and ^{Palacios et al. (2013)}, who found no relationship between psychological adjustment and age. On the contrary, the results of our study do not coincide with those found in some previous research, as authors such as ^{González and Cortés (2015)} and ^{Janssens and Deboutte (2009)}, among others, find more difficulties in boys than in girls. Similarly, although we might expect variables such as duration in residential foster care to be positively related to the presence of psychological adjustment difficulties (^{Delgado et al., 2012}; ^{Martín et al., 2007}), our data point to a lack of relationship between these difficulties and different aspects of the trajectory of these minors in the CPS. As authors such as ^{Carrera (2020)} point out, the trajectory of children in the CPS is very varied and it is unlikely to find significant relationships when we make an isolated analysis of its different variables in relation to the adjustment dimensions of these children.

Next, the data obtained regarding the fourth objective of the study indicate that, in relation to the support perceived from their collaborating families, the participants rate very positively the fair treatment, concern for them and feeling accepted by their collaborating family. In addition, they give high scores when rating, in general, the relationship they have with their collaborating families. On the other hand, the data also show that the support they perceive from the collaborating family, the relationship they maintain with them and their satisfaction with the program are associated with some difficulties, especially conduct problems. We see that this problem, which affects a considerable part of the children in residential care (^{Bravo et al., 2022}; ^{Del Valle et al., 2011}; ^{Fernández-Molina et al., 2011}; ^{Palacios et al., 2013}), also seems to be an important aspect for the development of collaboration. In this case, the correlations performed force us to make a bidirectional interpretation of the results. On the one hand, it is possible that those who obtain greater support from their collaborating families and are more satisfied with the collaboration tend to present fewer difficulties. On the other hand, it may be that children with more difficulties are likely to feel less support from their collaborating families, have a poorer relationship with them and perceive that they are less satisfied with the program.

Finally, we would like to highlight the innovative nature of these findings, as this is the first time that psychological adjustment has been studied in a group of minors in residential care who have access to living in a family environment. Nevertheless, this study has a number of limitations. First, we have a small sample size, although it is representative of the population analyzed, taking into account the number of collaborating families active at the time of the study (^{León et al., 2021}). On the other hand, the novelty of this research implies, in turn, a limitation in the discussion of our results, since there are no previous studies that have evaluated children in family collaboration in reference to their psychological adjustment. Furthermore, the cross-sectional data do not allow us to make predictions about the directionality of the relationships we found. Our purpose is, in the first place, to be able to continue evaluating the participants of this program at different times of the collaboration. To this end, we hope to have a larger sample size, which will allow us to verify the trends observed in the data of the present work with respect to psychological adjustment, which point to existing differences between children who participate in the family collaboration program and those who do not, in favor of the former. In addition, it would be a great opportunity to also be able to count on the vision of other relevant figures for these children, such as their teachers in the school context.

To conclude, we can extract some of the main ideas provided by this study. First, it is possible to say that our group of children in residential care with collaborating families presents a better profile of psychological adjustment, compared to other samples belonging to this population. Second, we can affirm that there are differences in the psychological adjustment of these children when we compare the assessment made by themselves with respect to those of their reference adults, being fundamental to implement this multi-informant assessment when we study the residential care population (^{Rodrigues et al., 2014}). Third and finally, we hope that our data will contribute to fostering the image of the collaborating families as figures who promote the strengths of these children, without ignoring their difficulties. Previous findings of our research have facilitated the development and elaboration of the protocol for the assessment and training of the Collaborating Families program, among other aspects (see Protocolo del Programa de Familias colaboradoras of Andalusia, ^{Consejería de Igualdad, Políticas Sociales y Conciliación, 2022}). On this occasion, the results regarding psychological adjustment support a more positive view of children who participate in family collaboration, compared to those who do not. Moreover, they show that there is a relationship between some difficulties and the subjective experience of these children with family collaboration. This should encourage further development of better and more comprehensive training and support programs for these families, so that they can respond to more complex behavioral and emotional profiles.