What Kind of School Organizational Decisions Serve to Enhance Sustainable Personal and Social Growth?

Sustainable decisions in education are those which pursue long-term achievements for students. In many cases, the principals of the schools are those who must make these decisions. In this paper we explore some of the factors that influence the students’ long-term development by means of an analysis of the correlation between certain organizational aspects of the schools which directly depend on the principals and the scores that their students have obtained in the International Assessment PISA 2018 in mathematics, reading and science. The results point to a better long-term student achievement when the organizational decisions are designed to enhance interpersonal relationships (teacher-teacher, teacher-student and student-student), responding to the specific needs of each person and based on trust. Conversely, it seems that when the organizational decisions are merely made to control academic life, they either bear upon academic achievement in a negative way or do not significantly affect it. The results suggest that carefully attending to interpersonal relationships is the key factor behind all the educational decisions which generate sustainable socio-personal development.


Introduction
The objectives outlined by the UN in the 2030 Agenda for Sustainable Development [1] have paved the way for the political decisions in education throughout the world over the coming decades. Sustainable decisions in education are those which seek to achieve long-term outcomes in students, and as such they are decisions which are not only taken by the political institutions but also by the principals of the schools.
Research literature has already demonstrated that school principals' actions cannot be separated from student outcomes, from academic achievement to social and emotional development [2]. Moreover, school principals are critical actors in encouraging teachers to be involved in professional evaluation processes and in fostering their professional development [3], which indirectly influences students' development. Jarl et al. [4] show that school performance is the result of human behavior and relationships. It can be affected and changed by the actions of individual teachers and principals as well as by policy-makers. The authors explain how success is fostered by an organizational dynamic that both encourages and reproduces cooperation and support among colleagues as well as a shared vision of teachers and principals acting on behalf of the learning of all students. In this regard, Intxausti et al. [5] consider the action of school leaderships as crucial in achieving effectiveness in education.
A primary goal of this paper was to explore some of the factors that influence students' long-term development by means of an analysis of the statistical correlation between the organizational aspects of the schools which depend on the principal and the scores that they have obtained in the international assessments PISA 2018 (OECD's programme for International Student Assessment aimed to measure 15-year-olds' ability to use their reading, mathematics and science knowledge and skills to meet real-life challenges) of competence in mathematics, reading and science, as well as in global competence.
Global competence has been defined as the capacity to examine local, global and intercultural issues, to understand and appreciate the perspectives and world views of others, to engage in open, appropriate and effective interactions with people from different cultures, and to act for collective well-being and sustainable development [6]. The students' assessment in global competence in PISA 2018 was carried out by means of indicators of specific socio-emotional skills and attitudes towards global and intercultural issues, as well as by measuring some reasoning and argumentation skills related to these issues [7].
However, we have noticed that the answers to the students' PISA 2018 questionnaire about global competence have been significantly incomplete. Some participant schools have only partially completed them and others have not answered them at all. For this reason, we have decided not to include the students' assessments in global competence in our study. Even so, certain issues related to global competence have been considered in our study because, in the PISA 2018 school questionnaire addressed to the principals of the schools [8], they are asked about specific organizational aspects particularly relevant to global competence, such as the carrying out of practices related to multicultural learning in their school, the inclusion of some multicultural aspects in the formal curricula of their countries (communicating with people from different cultures, knowledge of these cultures, respect for cultural diversity, openness towards intercultural experiences, foreign languages, critical thinking skills, etc.) or the inclusion in the official curricula of topics such as climate change and global warming, global health, migration and international conflicts, hunger or malnutrition in different parts of the world, causes of poverty or equality between men and women in different parts of the world.
The questionnaire addressed to the school principals also includes questions about other issues related to the principals' way of understanding education, whether as a mere question of skills to be acquired ( [9,10]) or as a process which also includes personal and social development [11]; their perspectives on, for instance, how to approach the career guidance of their students, the use of digital devices of social networks in teaching, the criteria for grouping students into different classes (by ability or otherwise), how the assessments of students are used (to classify them into categories or to guide students' learning and adapt teaching to students' needs); the way that the school use the external or internal evaluations and so on.
The diverse perspectives of understanding education have been related by Orón et al. [12] to three models of understanding human development [13]: (1) the autonomous self model (AS), which conceives of human growth as driven by cognition, and achieved as a result of individual productive activity aimed at mastering the environment, according to the wishes of the individual; (2) the personal self model (PS), which also understands the action as driven by cognition, whilst also giving value to human relationships in the pursuit of individual growth; (3) the inter-processual self (IPS), a social model of human interpersonal development which considers that personal growth cannot occur without the individual attending to personal relationships, looking for others' growth at the same time as one's own development.
Following the IPS approach [13], acting for the common good does not necessarily occur as a consequence of merely focusing on skills acquisition. Rather, such socio-personal development occurs when human action proceeds from one's interiority and is oriented towards the overall growth of specific others as much as one's own growth [13,14]. Moreover, this model does not conceive of leadership in education as successfully applying specific methodologies but rather as knowing how to make sense of education and promote cooperation among colleagues [11,15] and students [12,16].
In order to delimit the study, we have a priori classified the questions addressed to the school principals in the PISA 2018 study into three categories: those about structural aspects of the school, such as its location or its public or private status, which the principals have no power to influence; objective questions about organizational aspects over which the principal has the power to make decisions, such as grouping the students in accordance with their ability, for instance; and finally, subjective questions about issues that the school principal cannot directly act on, such as teachers' perceptions or opinions. We have analyzed the answers which have been given to the entire second group of questions, as well as those to some questions in the first group, with the aim of identifying those variables which correlate with the students' achievement in mathematics, reading and science as well as determining which organizational decisions the principals could make in order to enhance sustainable student development.
The results point to a better long-term achievement on the part of the students when the organizational decisions are based on trust [17] and oriented to enhance interpersonal relationships (teacher-teacher, teacher-student and student-student) in response to the specific needs of each person. This attitude is inseparable from a growth mindset [18], on the part of both school principals and teachers, which believes in the possibility of the growth of all the educational actors, without classifying them in accordance with their performance. Conversely, it seems that the decisions designed merely to control academic life either bear upon academic achievement in a negative way or do not significantly affect it. On the whole, we find out that carefully attending to the interpersonal relationships is the key factor which is behind all the educational policies which generate personal and social development in a sustainable way.

Materials and Methods
The Program for International Student Assessment (PISA) is a research project, conducted by the Organization for Economic Co-operation and Development (OECD), which evaluates the knowledge and skills gained by 15-year-old students over three-year periods. For the analysis, we have used the answers which were given to the PISA 2018 questionnaires throughout the world [19]. The files available on [19] include the background questionnaires and the data files (from 2000 to 2018) which enable the processing of the data.
The analytical frameworks of the theory underlying PISA surveys, as well as the assessments, can be found in [20]. Since 2000, the design of the questionnaires has been progressively reviewed in order to improve the cross-country comparability of the measured issues [21][22][23]. This paper focuses on analyzing the correlation between the students' outcomes in mathematics, science and reading (dependent variables) and the answers given by the school principals (independent variables).
The questions addressed to the school principals [8] have been classified into three categories: those about structural aspects of the school; objective questions about organizational aspects over which the principal has the power to make decisions; and subjective questions about issues which the school principal cannot directly act on. For the sake of objectivity, we have only taken the responses given to the entire second group of questions, as well as those to some questions in the first group, as independent variables (a total of 128 variables), distinguishing between the quantitative variables (where the answer is a percentage, a number or an ordinal), shown in Table 1, and the qualitative ones (where the principal is asked to indicate which category the school belongs to), which are shown in Table 2. The first column in each table shows the coding and description of the questions, while the second and third columns show the items (regarded as independent variables) which are dealt with in each question (coding and description). These codes have been taken from the "School questionnaire for PISA 2018. Main survey version" [8].

SC003Q01TA
Average of number of 15-years-old students per classroom

SC164Q01HA
Percentage of students who leaves school without a certificate SC064 During <the last academic year>, what proportion of students' parents participated in the following school-related activities?

SC064Q01NA
Percentage of the parents who discusses their child's progress with a teacher on their own initiative SC064 During <the last academic year>, what proportion of students' parents participated in the following school-related activities?

SC064Q02NA
Percentage of the parents who discusses their child's progress on a teacher's initiative

SC064Q03NA
Percentage of the parents who participates in local school government

SC064Q04NA
Percentage of the parents who voluntarily participates in the school activities

SC156Q01HA
The school has its own written statement on the use of digital resources SC156Q02HA There are digital resources (for instruction)

SC156Q03HA
Use of digital resources for specific subjects

SC156Q04HA
There are meetings of teachers on the use of digital devices

SC156Q05HA
There is a specific programme to prepare students for responsible internet behavior

SC156Q06HA
There is a specific policy about using social networks in teaching and learning

SC156Q07HA
There is a specific programme to promote collaboration on the use of digital devices among teachers SC156Q08HA There is scheduled time for teachers to meet in order to share, evaluate or develop instructional materials and approaches that employ digital devices SC011 Which of the following statements best describes the schooling available to students in your location?

SC011Q01TA
How many other schools are available in your location? SC012 How often are the following factors considered when students are admitted to your school? (never/sometimes/always)

SC012Q01TA
A student's record of academic performance is considered for their admission

SC012Q02TA
Recommendation of feeder schools is considered for students' admission SC012Q03TA Parents' endorsement of the philosophy of the school is considered for students' admission SC012Q04TA Whether the student requires or is interested in a special programme is considered for their admission SC012Q05TA Admission preference is given to family members of current or former students SC012Q06TA Admission preference is given to students who reside in a particular area

SC012Q07TA
Other criteria for the admission of students SC042 What is your school's policy about this for students in <national modal grade for 15-year-olds>?

SC042Q01TA
Students are grouped by ability into different classes SC042Q02TA Students are grouped by ability within their classes Assessments of students are used to award certificates to students SC036 In your school, are achievement data used in any of the following <accountability procedures>?
SC036Q01TA Achievement data are posted publicly (e.g., in the media) SC036Q02TA Achievement data are tracked over time by an administrative authority SC036Q03NA Achievement data are provided directly to parents SC037 Do the following arrangements aimed at quality assurance and improvements exist in your school and where do they come from?

SC037Q01TA
Internal evaluation/Self-evaluation of the school SC037Q02TA External evaluation of the school SC037Q03TA Written specification of the school's curricular profile and educational goals SC037Q04TA Written specification of student performance standards SC037Q05NA Systematic recording of data such as teacher or student attendance and professional development SC037Q06NA Systematic recording of student test results and graduation rates SC037Q07TA Seeking written feedback from students (e.g., regarding lessons, teachers or resources)

SC037Q08TA
Teacher mentoring SC037Q09TA Regular consultation with experts over a period of at least six months (aimed at school improvement)  The school offers collaboration with local libraries

SC053Q16IA
The school offers collaboration with local newspapers SC150 Does your school offer any of the following options to students in <national modal grade for 15-year-olds> whose <heritage language> is not the <test language>?

SC150Q01HA
The students with a native language different from the official one receive instruction aimed at developing skills in the official language

SC150Q02HA
The students with a native language different from the official one attend a preparatory programme aimed at developing skills in the official language

SC150Q03HA
The students receive some instruction in school subjects through their own language before transferring to regular classes

SC150Q04HA
The students receive a significant amount of instruction in their own language aimed at developing proficiency in both languages SC150Q05HA Class size is reduced to attend to students with a native language different from the official one SC152 Does your school offer additional <test language> lessons apart from <test language> lessons offered during the usual school hours?

SC152Q01HA
Additional lessons in the official language are offered out of regular school hours The goal is to identify which of these variables bear on the students' outcomes in mathematics, science and reading (dependent variables) in order to know how sustainable improvement might be achieved by the principal of a school.
As dependent variables we have taken the mean values of the scores obtained by the school in the general skills evaluated in each competence (mathematics, science and reading [19]), which have been obtained as a result of computing the average of the marks achieved by the students in each of the ten values collected in the test for each discipline. After eliminating the schools with less than nine respondents, we have considered a total of 17,995 schools (some of them with questionnaires only partially completed) located in 79 countries (with an average of around 228 schools per country). Luxemburg is the country with the minimum number of participating schools (40), while Spain has the maximum number (1011). We have estimated that a total of 564,282 students have been considered in the study.
In order to identify the variables that might bear upon the marks obtained in the tests (in mathematics, reading and science) we carried out two kinds of analyses.
For each quantitative variable in Table 1, a Pearson correlation coefficient was computed (in Table 3) to measure the strength of a linear association between this quantitative variable and the mean values of the scores in each competence (mathematics, science and reading). Correlations between 0.1 and 0.3 will be regarded as "weak correlations" and correlations higher than 0.3, as "intermediate" correlations. These benchmarks are also taken as significant in other papers [24,25].
Each qualitative variable (in Table 2) classifies the schools into several categories (public or private, for instance). For each qualitative variable, a one-way ANOVA test was carried out to analyze if there are any statistically significant differences between the means of the scores in each competence corresponding to different categories (the null hypothesis being that there are no differences between these mean values). The large size of the sample enabled most of the analyses to accept that these differences are statistically significant even when there is a relatively low significance level. We decided to regard the influence of a variable as statistically significant whenever the differences between the averages corresponding to different categories of the variable were over 4% of the average score (around 420 points), i.e., over 20 points. These values correspond to an effect of the Cohen size of d = 0.2. These benchmarks have been recently established by Kraft [26], who argues that, in an educational context, it may be more justified than in others to interpret respectively d < 0.05, d < 0.20 and d > 0.50 as "small", "medium" and "large" effect sizes.
In this paper, the analysis was carried out with Statgraphics Centurion XVII software (2014, Statpoint Technologies, Inc., The Plains, VA, USA).

Results
The correlation between the marks obtained in each pair of disciplines is very high: 0.9407 between Mathematics and Reading; 0.9642 between Mathematics and Science; and 0.9725 between Reading and Science. Table 3 shows the Pearson correlation coefficients between the variables in Table 1 and the averages of the values obtained in Mathematics, Reading and Science. The positive correlations have been highlighted in orange. This means that the higher the variable value is, the higher the average of the marks is in the corresponding discipline. A light orange means that the correlation is weak (between 0. 1  The variables with a negative weak correlation with the three disciplines are: percentage of student dropouts and percentage of parents who participate in local school government. Other variables with a negative weak correlation with only two disciplines are: percentage of students with a native language different from the official one and percentage of parents who voluntarily participate in the activities of the school.
The variable with a negative intermediate correlation with all three disciplines is the percentage of students from socioeconomically disadvantaged homes. Table 4 shows the F-ratio of the ANOVA analysis which has been carried out for each qualitative variable. Each qualitative variable divides the schools into several categories. The higher the F-ratio is, the higher the differences are between the means in a discipline corresponding to the different categories. For instance, the differences between the means corresponding to private or public schools are statistically significant.
In Table 4 the qualitative variables which bear upon the average marks (in a certain discipline) have been colored in blue. The rest are colored in green. A dark blue has been used to highlight the differences between the average marks corresponding to the different categories of the independent variable when they are higher than 20 points (at least between two of the levels). Table 5 explains which category gives the best scores in each case (colored in blue).  The students whose native language is not the official one receive significant amounts of instruction in their native language aimed at developing proficiency in both languages The qualitative variables that bear upon all three disciplines, with differences higher than 20 points between categories, are: whether the school is public or private (with higher marks in the private case), whether the achievement data of the students are provided directly to parents (with better scores in the negative case), whether there are counsellors employed by the school for the career guidance of students, whether there are digital devices for a general use, whether there is a specific programme to educate students in responsible internet behavior, whether an exchange programme with schools in other countries is offered, whether the school hosts visiting teachers from other countries, whether group musical activities are offered, whether a school play or musical has been organized, whether there is a school newspaper or magazine, whether lectures or seminars are offered and whether the school has rooms where the students can do their homework. In all these cases, the positive answers to the question give higher marks in the three disciplines.
There are other qualitative variables which give lower differences between categories for the three analyzed subjects, such as the location of the school (the larger the town is, the higher the marks are), whether the school offers an exchange programme with schools in other countries, resulting in an improvement in marks, or whether the students are grouped by ability either into different classes or within their classes, with the lowest marks in the case of their being grouped for all the subjects. However, if they are grouped for only some subjects, the marks are even higher than in the case of no grouping at all. Another factor which gives differences between levels lower than 20 points are whether written feedback (regarding lessons, teachers or resources) is sought from the students compulsorily, and with less differences (vis-à-vis the case of no feedback) when the school carries out this feedback on their own initiative. Finally, whether there is a regular consultation with experts aimed at school improvement bears upon the marks of the three disciplines in a negative way, the same as whether the students whose native language is not the official one receive instruction in their native language aimed at developing proficiency in both languages or if there is not a standardized policy for reading subjects (including staff development). However, if this policy is an initiative on the part of the school, there appear intermediate differences in the marks corresponding to the three disciplines vis-à-vis the case of no policy for reading subjects.
Other variables bear upon the reading subject only (with differences of over 20 points between levels of the factor) in a negative way (i.e., leading to lower marks), such as all teachers sharing the responsibility for the career guidance of the students, or the students' assessments being used to make judgements about teachers' effectiveness. In the case of the students with a native language different from the official one, the only practice that has a positive influence on the marks (and this only in mathematics) is offering the students additional periods of instruction aimed at developing skills in the official language, at the same time as their attending regular classes.
Finally, whether the formal curriculum includes the topic of international conflicts is significant in improving the results in both science and reading.

Discussion
The analysis of the results aims to identify a common denominator for the factors which correlate with good academic scores in the three disciplines which have been considered in the study. Our thesis is that this common denominator is a careful attention to interpersonal relationships (teacher-teacher, teacher-student, student-student and also student-families) as opposed to a merely cognitive educational perspective or a desire to control the educational process. This means that relationships based on trust seem to result in improved academic scores. This interpretation is aligned with the study known as the McKinsey report, which analyses twenty educational systems around the world by means of the data provided by several international assessment tests (PISA, PIRLS, TIMSS, etc.). According to this report, systems with the same economic investment in education obtain diverse results. The conclusion of the report is that the differences in terms of quality are not attributable to the investment, but rather to the teachers: "The quality of an educational system cannot exceed the quality of its teachers" [27].
In order to enhance the quality of teachers, it is clear that attending to interpersonal relationships is one of the main objectives to consider. However, enhancing interpersonal relationships is not the same as working on socio-emotional skills. In fact, the complexity of the educational reality cannot be addressed by means of the inclusion of a programme of emotional skills in the school curricula, as some authors have suggested, arguing that the school is excessively focused on cognitive skills. As will be seen, the research into non-cognitive skills has produced diverse results.
On the one hand, a meta-analysis which reviews the papers published between 2013 and 2017 on this topic [28] asserts that the correlation between students' achievement in non-cognitive skills and their academic performance is relatively low and also diverse; working on non-cognitive skills has had a positive effect on some students, a negative effect on others, and no effect on the remainder.
On the other hand, another meta-analysis in 2017 [29] showed that socio-emotional learning is effective in the acquisition of certain socio-emotional skills and some particular attitudes. For instance, this analysis concluded that there is a correlation between working on these skills and attitudes during youth, and such factors as academic success or safe sexual behavior.
A third meta-analysis which has been considered [30] concludes that emotional intelligence is the third predictor of academic performance, after cognitive intelligence and conscientiousness, but only with a percentage of between 0.7 and 3.6 of the variance, which is regarded as a low significance.
Although there are unresolved questions regarding how to interpret and evaluate the influence of emotional factors on academic performance, what does seem to be clear is the impact of interpersonal relationships on academic results. Comprehensive reviews of this topic can be seen in [31,32]. This suggests that it is the interpersonal relationships, and not the emotional skills, which are decisive when it comes to offering an educational experience that enhances sustainable personal and social growth.
An analysis of the results obtained in our study supports this argument. Firstly, we find some specific non-personal factors which correlate with good academic outcomes, such as the number of data projectors or the number of computers for the teachers' use. Rather than presuppose that the impact on student achievement is due to their presence per se, we recognize that the impact in question might be attributable to a more fruitful interaction between the teacher and the students when a computer or a projector is being used in the classroom.
Similarly, although the socioeconomic situation of the families significantly bears upon the academic outcomes, it is surprising that other economic variables, such as the funding sources (parents, government or benefactors) or the area of the town in which the families reside, are not relevant in this regard. This suggests that the significant aspect is the family experience of the students, rather than the economic factor.
In addition, there are some striking organizational aspects which may be interpreted along the same lines. The average number of students per classroom is not significant for academic achievement (only weakly significant in the case of the scores in mathematics). However, grouping the students by ability (into different classrooms or within the same classroom) has diverse effects. If they are grouped for all the subjects, the results are lower than if they are not grouped. If they are grouped only for some subjects, however, the achievement is better. Better results are also obtained when students with a native language different from the official one attend regular classes at the same time as receiving additional periods of instruction aimed at developing their skills in the official language. What is relevant here is that all the specific needs of the students can be attended to without losing the natural reference group of the students, their classroom. The same occurs when it comes to grouping students by ability. The best results are obtained when this grouping is only carried out for some specific subjects.
The presence of full-time teachers is significant, preferably if they have a degree and the school provides them with effective resources for teaching. However, the correlation of these factors with the students' performance in the case of part-time teachers is either not significant or is only weakly significant. These data lead us to think that the presence of full-time teachers enhances a trust relationship with the students, close relations arising naturally. Moreover, a full-time contract allows the teacher to engage with the students in a deeper, more personal way.
In addition, the positive influence of an internal counsellor who is employed by the school (rather than an external service being contracted) supports the idea that personal interaction enhances career guidance.
As regards the use of digital devices, the results are inconsistent. This fact seems to suggest that it is not their use per se which is significant, but rather their role in the relationship experience.
It is notable that the existence of given programmes, policies and curricula per se does not guarantee a better performance, which reinforces the idea that all these structural aspects need to be implemented in conjunction with careful attention to personal relationships. This does not imply that suitable programmes, policies and curricula are not important, but rather that special attention should be paid to the way in which they are implemented in the classroom. For instance, we find that the existence of curricula related to multicultural topics (cultural diversity, perspectives on historical and social events, expression of diverse identities, planet conservation, equality between men and women or causes of poverty) does not bear upon academic results, while sharing experiences with people from other cultures (by means of exchange programmes with other countries, hosting foreign teachers or inviting external people to give lectures or seminars) does have a clear positive influence on results. In all cases, a personal face-to-face interaction, not by digital means, takes place. Another action which improves outcomes is that of generating living spaces for the students, such as bands, choirs, newspapers or magazines, or performances of plays or musicals, as well as providing the students with a room where they can do their homework.
Although it is surprising that not all extra-curricular collective activities bear positively upon academic achievement, it is not easy to say why some have a positive impact and others not. Perhaps it depends on the kind of relationships which the activities foster, although further research is needed to explore this issue.
Regarding the relationships with the families, there are two factors which weakly bear on academic results in a negative way, namely, whether the parents voluntarily participate in school activities or whether they take part in local school government. Another variable, whether or not the achievement data are communicated directly to the parents, has a significant negative influence on the students' academic performance. The only variable which has a positive influence on the students' achievement is the percentage of the parents who discuss their child's progress on the initiative of a teacher. These data may suggest that the relationships between teachers are parents are beneficial when they are oriented to cooperatively helping students during their learning process rather than merely giving the parents a "final sentence" on their child's achievement.
This kind of study should analyze not only the variables which bear upon academic achievement in a positive way, but also what correlates negatively or does not correlate at all with academic results. Observing what all these factors share, we suggest that decisions focused on a merely cognitive view or a desire to control the educational process hinder student performance. We find that the principals' day-to-day tasks usually address the legal and organizational dimensions in which the principal assumes the role of a manager trying to achieve certain standards of production and efficiency centered on skill acquisition [33]. In fact, the self concept of the principal as manager is what teaching staff typically criticize [34]. They tend to feel controlled by bureaucracy, which is behind the burn-out phenomenon [35]. Both the principals' behavior and the intentions behind them are decisive for both the relationships between principals and teaching staff and also between teachers and students. If these relationships are perceived as attempts of control it bears upon the educational process in a negative way [36].
It is surprising that the use which the school gives to the students' assessments or the inclusion of certain topics related to global competence in the curricula do not influence the academic performance, while other issues bear upon it in a negative way, such as grouping the students by ability in all subjects, a compulsory evaluation on the part of the students of both the teachers and the school, and using student achievements to evaluate the teachers or asking for external evaluations of the school performance. We can hypothesize that these actions are being experienced as control processes rather than as educational challenges aimed to enhance sustainable growth.

Conclusions
We have found that the attention to interpersonal relationships can be regarded as a common denominator in the organizational aspects of the schools which bear upon a sustainable achievement on the part of the students. This interpretation of the results leads us to recommend that principals of the schools make certain decisions related to the organizational aspects of the school. Dealing with personal and social aspects as a question of knowledge and working on socio-emotional skills is not sufficient for sustainable development to be enhanced. In particular, we find that the organizational aspects particularly relevant to global competence, such as working on critical thinking skills or the inclusion of multicultural aspects or topics such as climate change, global health or migration do not significantly bear upon student long-term achievement, while sharing face-to-face experiences with people from other cultures has a clearly beneficial impact. This finding indicates that personal and social issues should be dealt with in a living and experiential way.
In addition, other issues which reflect principals' way of understanding education, such as their visions of how to approach the career guidance of their students, the criteria for grouping students into different classes, or how the assessments of students and teachers are used, have a significant influence on students' long-term development, as is reflected in the PISA scores which have been analyzed. In particular, the results point to some decisions which the school principal could make in order to generate sustainable changes in their school, some of them not requiring more than an academic year to be implemented.
For instance, the specific needs of the students should be particularly attended to, but without their losing their natural reference group, their classroom. Student-student relationships should be enhanced also by the organization of collective activities, such as a play or a choir, which contribute to cooperation and trust. Moreover, in order to enhance teacher-students relationships, resources to employ full-time teachers should be allocated, preferably if these teachers have a degree and the school provides them with effective resources for teaching. In addition, it is beneficial to allocate specific resources for the employment of career guidance staff.
Regarding the relation of the school with the students' families, face-to-face interaction between teachers and parents should be promoted, but not at the moment of teachers communicating the students' achievement data, but rather during the process of learning, on the initiative of the teacher, with the goal of cooperatively helping the students' progress.
Finally, it is important that both teachers and students do not perceive the use of achievement data and students' feedback about teaching and school as an instrument of control, oriented to discover their weaknesses. Control of processes is necessary, but it should not ignore interpersonal relationships. A cooperatively designed process of evaluating and attending to teachers and students' needs, based on trust, offers a better option than that of appealing to the application of external control processes. The perspective of attending to personal relationships in education has also been considered in several educational treatises (which criticize an approach excessively centered on skills acquisition [16,37]) as well as in philosophical reflections [38], pedagogical discussions [39] and intervention proposals centered on interpersonal relationships [40][41][42].
Although only the influence of the organizational decisions on the marks in three disciplines has been analyzed, further indicators should be studied in order to establish clearly which organizational decisions foster a genuine personal and social growth. The students' assessment in global competence could be a way of evaluating the socio-personal growth of students in future editions of PISA, whenever the collected data are solid enough to be included in a serious study.  Institutional Review Board Statement: Ethical review and approval were waived for this study, due to the data were directly obtained from PISA data base.
Informed Consent Statement: Informed consent was obtained from all subjects involved in the PISA study whose data have been used in this study.