The Child Protection System in Portugal and Catalonia (Spain)

Comparing child protection systems is always a sensitive issue, not only at the methodological, but also at the conceptual level given that terms and denominations are often derived from the historical, cultural and socioeconomic context of each territory/jurisdiction, and therefore, not always equivalent. That said, comparisons can also contribute to deepening the study of phenomena if one manages to identify the common, as well as the differentiating factors, which improve our understanding and help to put the feasibility of proposals for improvement and change into context.

Portugal and Spain are two neighbouring countries, sharing not only territory in the same peninsula, but also historical roots, cultural and religious practices, and Romance languages, which could imply similarities in their child protection systems. Catalonia is an autonomous region in the north-east of Spain. The reality is that they were indeed very similar until the 1980s, but that is no longer the case. Broadly speaking, one can say that, until the end of the twentieth century, the protection systems in both countries were characterized by the existence of residential institutions, which were larger in Spain than in Portugal. These institutions were close to centres linked mainly to the Catholic Church with a charity-based approach, providing care for the poorest children, often entrusted to them by the birth families themselves, which were large and had difficulties supporting them (Calheiros et al., 1993; Casas, 1998; Martins, 2006). The people in charge of these centres had little professional training and there were no professional diagnostic and intervention teams working with the biological families. The legislation did not provide for family foster care but only for adoption, which is decided in a private act before a judge.

From the 1980s onwards, with the fall of the dictatorial regimes and the establishment of democracy, both countries began to develop different child protection systems more in line with what was already being carried out in other democratic countries. According to Rodrigues et al. (2013) the term “residential care” was chosen in the first Spanish legislation (1987 and 1996 Laws), distancing itself from historical aspects linked to institutional charity, while Portuguese legislation (1999) opted for the term “institutional care”. Residential care in both countries underwent important structural and functional changes, some like each other, and others less so. These included giving priority to foster care over residential care, ensuring the temporary nature of both out-of-home placements, professionalizing the sector and regularizing the adoption process. Hence, residential care, in fact, has now become just another part of the protection system instead of being the only option.

However, important differences have existed in the trajectory of family foster care within both protection systems. In Portugal, foster care was legally recognized in 1979, before Spain (1987), but an amendment to the legislation in 2008 left kinship care out of the system and non-kinship care was contemplated as the only alternative. This meant that almost a thousand children who had been in kinship care until then were no longer considered to be covered by the protection system (Delgado, 2015). This is a clear example that, despite the cultural and socioeconomic importance that the family has in both countries, the legislation may determine a different route, as reflected in the statistics in Table 1. Thus, while foster care has become one of the main pillars of the protection system in Spain, in Portugal it is not regarded as a protective measure of out-of-home placement, which explains the great difference in percentages of residential care. Moreover, although the 2015 legislation indicated that foster care should be the priority for under 6 years-old, in 2016 only 18 (1.7%) of the 1072 children in this age group were in foster families (Instituto da Segurança Social, 2017). In this respect, according to Delgado (2015), great ambiguity and a lack of political definition have existed regarding the role social entities should play in the implementation of foster care.

Table 1 Percentage distribution of children in the child protection systems in 2016
Full size table

In Spain, kinship foster care was a priority option in the child protection services of most of the autonomous communities, although it was not included in the legislation until 1996. Non-kinship foster care was first contemplated under Spanish legislation in 1987, but despite great expectations for its implementation, it has failed to grow as intended and still makes up only a small part of the distribution of placement types. The recent 2015 State Law has been firmly committed to promoting it, laying down that children under 6 years of age should not be in residential care at all, but data on its impact are yet unavailable. Last official data in Spain (Ministerio de Sanidad, Servicios Sociales e Igualdad, 2017) were as follows: 40.2% in residential care, 38.1% in kinship care and 21.7% in non-kinship care (Table 1).

Law 142/2015 in Portugal introduced important challenges in favour of foster care and permanence. Nonetheless, even though non-kinship care is implemented more in Spain than in Portugal, both countries still have to do more in this matter (López et al., 2014). In this context, it is interesting to observe the transition made by the Irish protection system in just two decades, going from a system based on residential placements managed by large Catholic institutions to kinship and non-kinship foster placements currently accounting for more than 92% of the children in care in Ireland, one of the highest rates in the world (for more details, see Gilligan, 2019).

Why a Study on Subjective Well-Being?

The subjective well-being (SWB) emerges as a psychosocial (i.e., non-material) component of quality of life, that is life satisfaction, perceptions, values and aspirations that children—in this case—have in relation to the main aspects of their lives,, such as interpersonal relations, health, education and leisure time (Campbell, Converse, & Rodgers, 1976), believing that the subjective measures of social reality can be as useful as the objective measures in taking decisions and promoting social policies. However, research on the SWB of children in the protection system is still in its infancy in the international sphere. Studying SWB allows us to understand children’s viewpoints about different aspects of their lives, the evaluations they make and the perceptions they have about their foster family or the residential home where they are living, the schools they go to, how they use their time, or how they perceive their future, to give just a few examples. Moreover, it seems opportune to study the subjective well-being of children under the two child protection systems, in order to verify this kind of effectiveness. These studies assume that children are fully-fledged citizens (Casas, 1998). Therefore, in general, not only should their opinions be heard, but they should also be able to influence professional interventions and the design of childhood and youth policies, and actions for improvement that directly affect them should be undertaken.

As a starting point and reference, a few studies do exist. Llosada-Gistau et al. (2017) compared the SWB of children in residential and foster care in Catalonia, identifying greater levels of well-being among children living in kinship and non-kinship care, which were similar to levels among children in the general population, and significantly higher than among children in residential care. Also known is a study focused only on the residential care population in Brazil (Schütz et al., 2015), and another in Chile and Peru (Ortúzar et al., 2019), as well as a study on foster care in the United Kingdom (Selwyn et al., 2016). The present research was aimed at addressing the issue of the SWB of adolescents in the child protection system, which has not been dealt with before in Portugal, and at comparing data for children in both residential and foster care in Catalonia. Adolescents from the age of 11 can answer questionnaires with a deeper level of understanding and reflection.

However, our aim was not only to know and understand the situation of children in both protection systems in order to propose suitable changes and improvements, but also to carry out for the first time a comparison between both territories, that have similarities but also a big difference: the expanded placement of residential care in Portugal. Without having previous comparative studies between these territories regarding the questions we raise, it does not seem appropriate to raise hypotheses a priori, but only to seek answers to them.

Specifically, the objectives of this study were: (1) to compare the level of SWB perceived by children, relative risks and odds ratios, taking into account type of care, jurisdictions, children’ sex and age; (2) to assess children’s happiness with the placement; (3) to assess children’s relationship with carers; (4) to assess children’s satisfaction with school; (5) to assess their perceptions about themselves and their lives; and (6) to analyse the correlations between SWB measures and the variables related to children´s satisfaction.

Methods

Samples

In Portugal, a cross-sectional study was carried out in 2018, following the same method that had been used in Catalonia before. The study population comprised children in out-of-home care aged between 11 and 15 years old. The sample of children in foster care included all the children (44) hosted in four Portuguese districts. In Portugal, increasingly fewer children are in foster care. The districts were chosen because they were known to have enough children to make up a sample, which turned out to be the total population at that age. In addition, five of the children had great disabilities and were unable to answer the questionnaire. Thus, the sample was finally made up of 39 children with an average age of 13.27, SD = 1.42. The sample of children in residential care included a total of 145 respondents with an average age of 13.51, SD = 1.26. This sample was obtained by randomly choosing residential homes among the four Portuguese districts where the children in foster care were found and asking all the children to respond to the questionnaire. We had the collaboration of our Social Security Institute to have access to those adolescents and families. All the process of contact with residentials’ director, professional teams and foster families followed were the same as those performed in Catalonia and are described in the next paragraph.

In Catalonia, a cross-sectional study was conducted in 2013. The study population comprised children in residential care and foster care aged between 12 and 15 years old. A list of all children in the selected age-range in residential and family foster care was used. The sample was ad-hoc and included all the children who responded. We compared the sociodemographic characteristics of the sample regarding gender, place of birth and education with the overall population in care and found no significant differences. The questionnaire was sent by post in the name of the children in family and residential placements through the Catalan Government’s Observatory on Children’s Rights, together with a letter describing the study. All the directors of the children’s homes and the family support teams were contacted previously to inform them about the study and data collection process, and to ensure the questionnaires could be completed by the children individually and voluntarily with informed consent and that no incentive would be given for participating.

The sample of children in foster care included a total of 41 children with an average age of 13.29, SD = 1.03. The sample of children in residential care included a total of 326 respondents with an average age of 13.17, SD = 0.96.

There were more boys in out-of-home placement in Catalonia, and more girls in Portugal. In Portugal, only 6 children (4.1%) in residential care were born abroad in 6 other countries, while in Spain there were 18.4% (60) of children in residential care and 9.8% (4) in foster care who had been born abroad in 24 other countries (Table 2).

Table 2 Children characteristics
Full size table

Questionnaire

The data were obtained by administering the International Survey of Children’s Well-being questionnaire (ISCWeB: www.isciweb.org), which had been adapted and pre-tested in both jurisdictions. This questionnaire included questions grouped into 11 thematic sections: personal information, relationship with their parents, home and the people they live with, the things they have, friends and other people, the area where they live, school, how they use their time, more about them, how they feel about themselves, and their life and their future.

Variable Measures

The questionnaire was self-administered and included the version of the Personal Well-being Index-School Children (PWI-SC), developed by Cummins et al. (2003). It was originally created to measure the SWB of adults and was later adapted by Lau and Cummins (2005) for children, consisting of seven items. Nine items were used in our study, the last two being added by Casas et al. (2012), which related to use of time and life as a student. The scale ranged from 0 (Not satisfied at all) to 10 points (Totally satisfied). In this study, the reliability of PWI-SC9 was very good in Portugal, α = 0.85, and good in Catalonia, α = 0.77.

The general question in this scale was “How satisfied are you with each of the following things in your life”: (1) The things you have; (2) Your health; (3) How secure you feel; (4) The things you know you do well; (5) Your relationships in general; (6) The things you do outside your host house; (7) What could happen later in your life; (8) Your use of time; and (9) Your life as a student. The mean score in this scale was used to represent it.

Other variables included in the study were those related to the specific objectives of the study, as following: jurisdiction (Portugal and Catalonia); type of care (residential care and foster care); sex (girls and boys); age; happiness with the placement (three-point scale); relationship with carers (mean of three questions answered on a Likert-scale: Portugal, α = 0.78, and Catalonia, α = 0.81); frequency of activities with carers (mean of three questions answered on a Likert-type scale: Portugal, α = 0.85, and Catalonia, α = 0.74); three questions, answered in a scale 0–10: how satisfied are you with (1) your school experience, (2) your self-confidence, (3) your life as a whole; two questions answered in Likert-scale: how much do you agree with this sentence: (1) I feel alone, (2) I'm optimistic in relation to my future; and one question answered in a Likert-type scale: how often do you usually do extracurricular activities?

Data Analysis

Data analysis was carried out by using IBM.SPSS.23. Statistical analysis included descriptive—mean, standard deviation, percentages—and inferential statistics—independent chi-square test, odds ratio, relative risk, Student’s t-test, Mann–Whitney U test, Pearson’s correlation, and Spearman’s correlation.

Ethical Issues

Throughout the data collection process in both jurisdictions, we provided an explanation of the study to children, families and professionals, ensuring the confidentiality and anonymity of responses, as well as the non-obligation to respond to the survey or, the possibility of not responding to any question if they felt unprepared or uncomfortable. In Portugal, we obtained the consent of all institutions—Portuguese Social Security, residential homes, schools and foster families—and all the children who responded to the survey—Informed Consent Form. In Catalonia the study was approved by the department of the Catalonian Government responsible for the Child Protection System (DGAIA). The survey was sent by the Welfare Department to the addresses registered on their database. Children participated in the study voluntarily, only under consent and without any kind of incentive.

Results

Children’ Sex, Age, Type of Care, and Jurisdiction

Regarding SWB measured by PWI-SC9, Table 3 shows that there were slightly higher mean scores in Portugal, but without statistical significance. However, there were statistically significant differences between SWB in residential care and in foster care both in Portugal, t = 4.142, p < 0.001, g = 0.56, and Catalonia, t = 3.396; p < 0.01, g = 0.57. We consider that the mean differences in both territories present a medium effect size with confidence intervals that ensure a non-overlap of values with the averages of the other group under comparison. We used Hedges’g to measure the effect sizes because the samples sizes are very different (Hedges, 1981), and Cohen’s rule of thumb for interpreting results: small effect ≥ 0.2; medium Effect ≥ 0.5; and large Effect ≥ 0.8 (Cohen, 1977). However, this rule must be used in the context of the subject under analysis.

Table 3 Comparison of PWI-SC9 scores
Full size table

Analysing the statistically significant differences between Portugal and Catalonia concerning the nine items of PWI.SC9 scale, it was discovered that Portuguese children in residential care showed more satisfaction in relation to the things they had, t = 2.83, p < 0.01, g = 0.28; how secure they felt, t = 2.469, p < 0.05, g = 0.25; and the things they knew they did well, t = 2.054, p < 0.05, g = 0.20. In foster care, the same occurred in relation to how secure they felt, t = 3.332, p < 0.01, g = 0.74; and to what could happen later in their lives, t = 1.82, p < 0.05, g = 0.41.

Table 4 shows that boys in residential care in both jurisdictions displayed higher SWB scores than girls, with statistical significance: Portugal, t = − 2.366, p < 0.05, g = 0.04; and Catalonia, t = -3.532, p < 0.001, g = 0.39. Additionally, we aimed at comparing younger with older children. The children’s mean of age in Portugal is 13.47, and in Catalonia 13.19 years old. Thus, we divided children in two groups using the using the rounded average of 13. However, we tested different groups (11–12 vs. 13–14-15) and we obtained the same results pattern. It seemed that children in residential care had less SWB when they were older, being this result statistically significant in Catalonia, t = 2.008, p < 0.05, g = 0.23.

Table 4 PWI-SC9 mean values by sex and age
Full size table

Analysing the relative risks and odds ratio of children having more or less than average SWB on the PWI.SC9 measure in both jurisdictions (Portugal: 8.0344; Catalonia: 7.741), one can conclude that, as far as sex is concerned, there was a 1.4 more chance or relative risk of girls in residential care in Catalonia having a lower than average SWB than boys, p < 0.05. In this jurisdiction, children with less than average SWB had 1.885 more chance of being a girl than children with higher than average SWB, p < 0.05. In foster care in Catalonia, odds did not have statistical significance, since children’s SWB was independent of their sex. In residential care and foster care in Portugal, no odds were statistically significant. Nonetheless, the same pattern as in Catalonia existed.

Analysing the relative risks and odd ratio of children having more or less than average SWB on the PWI.SC9 measure in both jurisdictions, if we consider age groups, we can conclude that there was a 0.788 chance or relative risk of children between 11 and 13 years old in residential care in Catalonia having less than average SWB compared to children aged 14 to 15 years old. In this jurisdiction, children with less than average SWB had only a 0.633 chance of being 11 to 13 years old compared to children with higher than average SWB. In foster care in Catalonia, the odds did not have statistical significance, since children’s SWB was independent of their age. In residential care and foster care in Portugal, none of the odds were statistically significant. However, the same pattern as in Catalonia existed.

About the Placement

Regarding the placement, there were more children that liked foster care than children that liked residential care in both jurisdictions with statistical significance, Portugal: χ2 = 45.2, p < 0.001 and Catalonia: χ2 = 23.23, p < 0.001 (Table 5). However, there was no statistically significant difference between the two jurisdictions. Additionally, we found that in Portugal, 37.2% of children in residential care and only 10.3% in foster care preferred another type of social intervention. In Catalonia, the percentages were similar: 38.3% in residential care, and 14.6% in foster care.

Table 5 Happiness with the placement
Full size table

About the Carers

Table 6 shows that there were statistically significant differences between Portugal and Catalonia concerning the relationship with carers at home: on average, there was a worse relationship with carers in Portugal in both residential and foster care. Similarly, there were statistically significant differences between Portugal and Catalonia as regards the frequency of activities with carers: on average, activities were less frequent in Portugal in residential care. In Portugal, the frequency of activities with carers was lower in residential care in relation to foster care with statistical significance. In Catalonia, the relationship with carers at home was worse and the frequency of activities with carers was lower in residential care in relation to foster care, with statistical significance.

Table 6 Variable scores according to jurisdiction and type of care
Full size table

About the School

In both jurisdictions, children from residential care expressed less satisfaction with their school experience. However, this difference was only statistically significant in Portugal (Table 6).

About the Children and their Lives

In Portugal (Table 6), children in residential care presented less self-confidence without statistical significance, and less satisfaction with their lives in relation to children in foster care with statistical significance. They also felt more alone and less optimistic about their future.

In Catalonia, children in residential care presented less self-confidence and satisfaction with their lives in relation to children in foster care with statistical significance. In contrast to Portugal, children in residential care in Catalonia felt less alone than in foster care and had fewer extracurricular activities. Statistically significant differences were only found between both counties in relation to satisfaction with their lives—higher in Portugal in both types of care—and feeling alone less in Portugal in both types of care.

Analysing the Correlations Between the SWB Measure and the Variables Related to Children´s Satisfaction

Table 7 shows that SWB was positively correlated with the relationship with carers at home, frequency of activities with carers, satisfaction with school experience, satisfaction with self-confidence, and satisfaction with life in both types of care and in both jurisdictions. The exception is the correlation with frequency of activities with carers for foster care in Portugal, r = 0.14.

Table 7 Pearson’s correlation between PWI-SC9 and other variables
Full size table

Table 8 shows that PWI-SC9 was positively correlated with carrying out extracurricular activities, and being optimistic about the future, and negatively correlated with how alone they felt. The latter variable was the only one with statistical significance in both types of care and both jurisdictions.

Table 8 Spearman’s correlation between PWI-SC9 and ordinal variables
Full size table

Discussion

The comparison between the Portuguese and Spanish protection systems reveals different patterns of evolution during the last few decades. In Spain, a legal reform was carried out throughout the 1980s and 1990s that prioritized foster care (Del Valle et al., 2009). However, it is still less developed than in most European countries (Eurochild, 2010). In the case of Portugal, despite the legal reform carried out in 2015, which has reinforced the importance of the family as the most desirable context for the development of the child, there isn’t yet a real impact in the number of children in foster care. In fact, the situation remains far from the reality of Spain and especially from the rest of European countries (Instituto da Segurança Social, 2017).

According to the overall aim of the study, the results show similarities but also differences between territories. Portuguese and Catalonian adolescents living in foster care reported higher SWB in all life domains than those in residential care. The mean score on the PWI-SC9 scale for children in care was 8.17 points out of 10. However, results showed much lower SWB among children in residential centres (7.74), than among those in foster care (8.6). The means for those in family foster care were very close in both Portugal (8.76) and in Catalonia (8.44).

One important result is that only half of the children in residential care said that were happy living in a residential centre in Catalonia and even fewer (31.3%) in Portugal. It is very important to remind that most children in Portugal are in residential care. It can be concluded that this adversely affects their SWB, when we consider the same indicators for foster care: 93.5% of children in Catalonia and 92.1% in Portugal reported being happy to be living in a foster family. Some studies, such as Schwartz et al. (2014) in Canada, have already found that children at risk prefer a foster care placement rather than a residential care.

Foster care should be a top priority in both contexts as it proves to be a more capable measure in promoting the well-being of children. This result is particularly significant in the case of Portugal, where the incidence of institutionalization among children is significantly higher than in Catalonia, which constitutes a big difference between the two territories. Moreover, in Portugal, it is somewhat difficult to distinguish the characteristics motivating entry into the protection system between children that are placed in foster care and those that are forwarded to institutions (López et al., 2014). This means that the main responsibility for the outcome of lower levels of SWB among children in residential care in Portugal might be attributed to the residential centres alone, contrary to what happens in Catalonia (Llosada-Gistau et al., 2017).

These findings are in line with results obtained by other—still few—studies that compared the SWB of children living in different out-of-home placements, in which those who were living in residential centres always presented lower SWB scores. In Brazil, Dell’Aglio and Siqueiro (2010) presented data on young people living in institutions and those living with their disadvantaged families. Those two groups shared risk factors, but it was found that the situation was more serious for institutionalized children and adolescents that presented a higher school drop-out rate, a higher number of siblings on average and a lower percentage of parents living together. More recently, in the same country, Schütz et al. (2015) compared the SWB of children in residential care centres versus those from the general population living with their families. The results indicated that all items were significantly better for children living with their families. The same occurred with the findings of Llosada-Gistau et al. (2015). These authors concluded that adolescents living in kinship and non-kinship foster care reported better SWB in all life domains than those living in residential care.

Nevertheless, it is important to stress that other key factors that determine lower or higher levels of SBW among children in residential care may exist. The children’s relationships, such as secure bonds with professionals, family and peers, may be crucial for them to experience stable contexts (Lima & Morais, 2018) and satisfaction with their quality of life (Bradshaw, 2015; Rees et al., 2011). Recently, results from the study carried out by Campos et al. (2019) showed that the adolescents in residential care in Portugal had higher levels of mental health problems and lower academic achievement, indicating an urgent need for early screening of mental health problems and socio-cognitive interventions for children in residential centres.

Our study revealed lower SWB in girls in residential care in both jurisdictions. This fact was reinforced by the relative risk and odds ratio analysis. In contrast, this did not occur with girls living in foster families. This situation had previously been reported by other authors such as Tomyn (2013), who observed lower welfare scores in Australia among girls than boys of the same age and situation. For further research it would be important to explore why and how sex has an influence on the scale responses for girls living in residential care. An in-depth comparison between girls living in residential centres and in foster families should be developed in order to look for characteristics that might explain the differences. Gender differences were also observed with children in residential care in Portugal (Campos et al., 2019). This situation raises a question that would be worth addressing in future research: could foster care act as a protective factor for the SWB of girls at risk? Additionally, according to Llosada-Gistau et al. (2017), children seem to have less SWB in residential care when they are older.

Limitations

This study has some limitations, besides those related to the use of a questionnaire and the comparison between different countries, namely the fact that there were fewer children in the foster care sample than in the residential care sample in both jurisdictions. This explains why some of the differences were not statistically significant between the two groups.

Other limitations concern the following facts: not all children were studied, but only those aged between 11/12 and 15 years old; the lack of other possible variables that may determine children’s SWB, such as the different reasons for their removal from home; and the two surveys have been carried out years apart despite being justified because one was based on the development of the other.

In addition, future research should also study how objective well-being variables, not included in this study, are related with SWB of children in care. Furthermore, the current data does not include the history pathway of children through protection system. This kind of information could enrich future discussion.

Finally, it will be interesting to study in the near future Portuguese adolescents in kinship care, comparing with the same group in Catalonia, as well as with adolescents from the general population, foster care and residential care in both territories.

Implications

The significant correlations between SWB and almost all the other variables show that they are closely related and determine crucial aspects that should be equated by professionals and policy decision-makers in both jurisdictions. Thus, the relationship with carers and friends, satisfaction with school, self-confidence, and avoiding loneliness seem to be fundamental issues related to children’s perception of their well-being.

As consequences to put into practice in Portugal, to enhance SWB in residential care, it seems evident that it is necessary to take into consideration the voices of children at all stages, promote initial and continuous professional training for caregivers, limit the number of children in each residential care centre (less than 12), and develop effective programs that prepare young people to leave Residential Care and support the early years of independent living (Rodrigues & Barbosa-Ducharne, 2017). Moreover, most of the children who are in out-of-home care have learning difficulties and above average school drop-out rates, which worsen as they move from the first levels of education to the next (Jackson & Sachved, 2001). However, academic achievement is one of the main paths to social mobility as it is a way to abandon poverty and exclusion. Therefore, school and educational development must be effective priorities for out-of-home care, ensuring the support, resources and additional time needed to provide real opportunities for success (Montserrat et al., 2015). In this line, it would be interesting for further research to examine some objective measures like academic achievements, which are missing from this study.

In Catalonia, as Llosada-Gistau et al. (2015) observed, adolescents living in residential care have fewer opportunities to decide on the use they make of their time. Protection Services should commit decisively to offering and promoting quality leisure time for children in care as far as possible. This policy could work as a highly compensatory factor in preventing and combating the feeling of loneliness and exclusion. As mentioned earlier, psychological balance and self-esteem may be the greatest contributors to the SWB of these children.

Conclusions

On one hand, the results have reinforced the new and limited knowledge about the outcomes of different types of out-of-home care: in general, children that live with a family are more likely to feel happy with their lives. On the other hand, when we compare territories, children in Portugal said they were less happy living in residential centres than those in Catalonia, where the child protection system is less institutionalized than in Portugal, and this opens a new reflection.

The data highlight the need for both systems to continue to bear in mind the key factors associated with the quality of out-of-home care systems and to promote de-institutionalization strategies as indicated in the respective legislation of both countries.

It seems that foster care offers a better, more normalized and personalized living context. However, in both jurisdictions, many adolescents – half in Catalonia and one third in Portugal – were happy to live in residential centres, and this result, in itself, is very important because it confirms that it is possible to live in a residential centre and have a high level of SWB, although it is not a majority situation. More in-depth studies are needed to analyse the profile of this group, comparing it with the young people in residential care who manifest low SWB, trying to identify the factors that determine their evaluations. The absence of an adult role model in their life paths may justify these results, as well as long permanence in this type of care, or less positive previous experiences of integration in foster care. Nevertheless, in subsequent research it will be necessary to study what changes residential centres must make, namely in size, location and type, as Rodrigues et al. (2013) pointed out, to have a positive effect on children’ SWB and provide them with a measure that meets their needs. In parallel, foster care should be increased, starting for all the youngest children who are still institutionalized.

The collected data showed that, in both jurisdictions, the variables ‘frequency of activities with carers’ and ‘satisfaction with life as a whole’ presented lower scores in residential care in relation to foster care with statistical significance. In Portugal, this also occurred with ‘satisfaction with school experience’, while in Catalonia, this type of differences appeared in relation to ‘relationship with carers at home’, ‘satisfaction with self-confidence’, and ‘how often they do extracurricular activities’. These results are clues that should be further explored in future research. If they can be confirmed, they could help in guiding the development of residential care.

In light of the previous discussion, both child protection systems should, on one hand, provide as much support as possible to disseminate and promote family placements, whether in foster care or kinship care; on the other hand, ensure a continuous and effective way to select and train new foster families, in sufficient number to meet the needs of the systems. In Catalonia, it is essentially a question of reinforcing the support and supervision in training, educational and psychosocial areas. In Portugal, to promote a gradual but resolute de-institutionalization process, led by the government, with the joint contribution of Social Security Institute, front-line social workers, Catholic Church and Portuguese society.