Assessing Psychological Well-Being Among Adolescents: A New Multidimensional Instrument for Chilean School Students

Abstract

Adolescent psychological well-being is a key component of social and educational sustainability. Strengthening well-being during the school years supports healthy development, academic engagement, and resilience to contemporary challenges. Within the framework of the United Nations 2030 Agenda for Sustainable Development, fostering mental health and equity in education contributes to more inclusive, cohesive, and peaceful communities. This study aimed to develop and validate the Psychological Well-Being Scale for Adolescents in a Chilean school population. Drawing on Ryff’s multidimensional model, the scale included 32 items and was administered to 8638 high school students from all sixteen regions of Chile. Exploratory Factor Analysis identified a seven-factor structure in which self-acceptance was divided into two dimensions: one associated with self-esteem and achievement, and another with self-appreciation and body image. Confirmatory Factor Analysis demonstrated that the seven-factor model achieved the best fit. The scale showed high internal consistency, adequate convergent validity, and measurement invariance. The Psychological Well-Being Scale for Adolescents is a psychometrically sound and culturally adapted instrument validated with Chilean adolescents in school contexts. Current evidence supports its use for research and applied assessment within Chilean educational settings.

1. Introduction

Education for sustainable development requires incorporating not only curricular and cognitive dimensions, but also relational and socio-emotional aspects that strengthen students’ ability to face contemporary challenges. As UNESCO has pointed out, educational approaches that focus exclusively on academic performance run the risk of neglecting students’ emotional needs, compromising the construction of inclusive and resilient environments [1,2]. In this sense, the psychological well-being (PWB) of adolescents is a strategic resource for achieving sustainability goals, as it provides the emotional and cognitive stability necessary to learn, coexist, and actively participate in democratic life [3,4]. Thus, promoting autonomy, self-acceptance, and positive relationships, which are core dimensions of well-being models, is directly linked to developing civic and prosocial skills oriented towards the common good and sustainable education [4,5,6,7,8,9].

From this perspective, adolescent psychological well-being is a strategic resource for advancing the Sustainable Development Goals (SDGs), as it promotes mental health, social equity, and educational sustainability as interdependent dimensions of human development. Thus, from the perspective of SDG 3 (Good Health and Well-being) as a central focus for sustainability, empirical evidence shows that psychological well-being is a protective factor against anxiety, depression, and emotional dysregulation, and that it promotes healthy and resilient lifestyles, contributing to the development of socio-emotional skills that are fundamental to positive mental health [4,5]. In the school context, SDG 4 (Quality Education) maintains that psychological well-being directly impacts students’ participation, motivation, and academic commitment, promoting more meaningful and sustainable learning [2,6,7]. Moreover, recent evidence in sustainable education suggests that personal and social development occurs in school environments characterized by trust, participation, and collaboration. Studies on teacher resilience and self-efficacy [10,11] highlight that human strengths are nurtured by social support and perceptions of personal efficacy [12,13]. In this way, the inclusion of well-being as an indicator of educational quality represents a paradigm shift: schools are no longer exclusively instructional spaces but have become environments for comprehensive development, where learning also means feeling valued, safe, and connected to others [2,6].

Schools are also ideal places to promote values through an inclusive approach linked to SDG 5 (Gender equality) and SDG 10 (Reduced inequalities) [6,8,9], as psychological well-being is a means of promoting social justice and educational equity. Unequal access to learning and emotional care opportunities has a direct impact on young people’s mental health, especially in contexts of socioeconomic vulnerability [9]. In this context, strengthening psychological well-being in schools can reduce structural gaps, empower adolescents as agents of change, and generate healthy, equitable, and sustainable learning trajectories [1,6,7,8,9].

Finally, on a broader level, adolescent well-being is linked to SDG 11 (Sustainable cities and communities) and SDG 16 (Peace, justice, and strong institutions) due to its systemic effect, given that schools are fundamental spaces for social cohesion and local sustainability. In this sense, promoting emotional well-being, empathy, and student participation strengthens the construction of safer, more resilient, and more inclusive communities, consistent with the notion of Learning Cities promoted by UNESCO [14]. Recent literature expands on this approach by pointing out that incorporating sustainability into education is key. This is especially true when it is implemented from an early age through, for example, flexible, participatory, immersive methodologies with playful teaching methods that promote informed and responsible communities, positioning schools as fundamental pillars for the achievement of SDG 11 [15].

It also supports SDG 16 (Peace, Justice, and Strong Institutions) by promoting relationships based on respect, cooperation, and peaceful conflict resolution. In this way, education for well-being contributes to institutional strengthening and the consolidation of democratic and caring school cultures, in which adolescents can exercise citizenship and participation [6,8,9].

Therefore, translated to the adolescent context, this implies that measuring and strengthening PWB is not only an individual task, but also a collective effort involving schools, families, and communities as co-responsible agents. Placing adolescent PWB at the center of education for sustainable development enables us to catalyze empathy, solidarity, and cooperation, moving toward schools that are more equitable, inclusive, and aligned with the challenges of the 2030 Agenda.

From this perspective, adolescence constitutes a crucial developmental stage marked by profound biological, psychological, and social transformations, during which well-being represents not only a personal resource but also a shared societal value [16]. Given that adolescents spend most of their time in school, the educational environment plays a crucial role in shaping their psychological functioning and emotional resilience [17]. In this framework, understanding and accurately measuring PWB in this population not only allows us to describe their health and adaptive capacity but also to guide school interventions and policies consistent with education for sustainable development, promoting trajectories of flourishing during this formative period [1].

1.1. Psychological Well-Being as a Construct

Human well-being has been a subject of reflection since classical philosophy, with an emphasis on Aristotelian eudaimonia, understood as the “good life” oriented towards virtue and self-realization [18]. In contemporary psychology, this tradition was solidified in the last decades of the 20th century through conceptual frameworks that approached the understanding of health beyond the mere absence of disease, positioning well-being as a central axis of human development [19,20]. The emergence of Positive Psychology fostered a transition towards the systematic study of human strengths, flourishing, and the conditions that favor optimal functioning [21], helping to establish PWB as a central dimension of health and quality of life.

In this regard, psychological well-being is regarded as a crucial indicator of positive human functioning. It has been defined as a state in which individuals recognize their capabilities, cope with daily life challenges, develop meaningful relationships, and find meaning and purpose in their actions [22]. Thus, PWB constitutes a deeper framework than subjective well-being (characteristic of the hedonic tradition), as it does not focus solely on the presence of positive emotions or life satisfaction, but rather on processes of self-realization, personal growth, and vital purpose [23,24,25].

Regarding conceptual frameworks, two of the most influential models for understanding psychological well-being are Carol Ryff’s multidimensional model and Deci and Ryan’s Self-Determination Theory. Ryff [24] proposed a model comprising six interrelated dimensions that together reflect optimal psychological functioning: self-acceptance, understood as maintaining a positive attitude toward oneself; positive relations with others, centered on developing meaningful and trusting interpersonal bonds; autonomy, related to independent self-regulation despite social pressures; environmental mastery, involving effective management of life demands; purpose in life, linked to having direction and goals that give meaning to experience; and personal growth, which reflects openness to new experiences and continual self-development.

This model, which is widely supported empirically, highlights the importance of existential and relational factors in human development. For its part, Self-Determination Theory emphasizes the satisfaction of three basic psychological needs (competence, autonomy, and social relatedness) as essential for intrinsic motivation and well-being [23,26]. Together, these approaches show that psychological well-being is not a fixed state but a dynamic process of development and adjustment throughout the life cycle.

During adolescence, psychological well-being becomes particularly important. This stage is characterized by biological, cognitive, and social transformations that imply developmental discontinuities and necessitate the construction of a coherent identity [27]. From a positive development perspective, adolescence is a period that favors learning, the exploration of potential, and thriving in various contexts [28,29]. However, it is also a critical moment of vulnerability to psychosocial risks, which reinforces the importance of having resources and protective factors that strengthen well-being and promote flourishing trajectories, rather than discouragement and apathy [30].

PWB in adolescence has been linked to different psychological constructs. Previous studies have shown that higher PWB levels are associated with protective factors such as emotional intelligence [31], dispositional optimism [32], and resilience [33]. Regarding mental health, adolescents with higher psychological well-being tend to report less anxiety-depressive symptoms [34], better sleep quality indicators [35], and better overall mental health indicators [36]. Meanwhile, low PWB levels have been related to psychosocial outcomes, such as higher violence, greater perceived stress, suicidal risk, low self-esteem, and conflicts in interpersonal relationships [37,38].

However, the literature indicates that studies on psychological well-being in adolescents are fewer in number than those focused on adults [39]. One reason is that PWB involves abstract cognitive processes that can be more complex to evaluate at early ages [40]. This gap presents theoretical and methodological challenges for research, particularly in cultural contexts where resources for instrument validation are scarce. Particularly in the Latin American school context, the creation and validation of culturally relevant and psychometrically robust instruments remain an outstanding challenge [41].

1.2. The School as a Context for Promoting Psychological Well-Being

School is one of the most influential contexts for adolescent development, as it is a space where academic, interpersonal, and formative learning converge, directly impacting mental health and well-being [42,43]. Thus, in terms of Bronfenbrenner’s ecological model [44], the school institution is configured as a central microsystem that interacts with other levels of adolescent life, influencing the construction of identity, the acquisition of socioemotional competencies, and the strengthening of psychological resources. In this sense, psychological well-being, as defined by Ryff [24], assumes strategic value in the educational context, as its dimensions can be enhanced through daily interactions with teachers, peers, and the school community.

Research in adolescents suggests that the six dimensions of Ryff’s psychological well-being model are related to different aspects of school functioning. In general, it has been shown that all dimensions are positively associated with overall mental well-being [45]. More specifically, positive relationships and self-acceptance are linked to a motivational orientation towards goals, which favors commitment and academic direction [46]. Furthermore, the practice of structured activities promotes improvements in personal growth, environmental mastery, and self-determination, dimensions that impact academic success and personal development [47]. Thus, Ryff’s model not only constitutes a theoretical framework for evaluating well-being in adolescence but also offers practical guidelines for designing educational programs that promote psychological well-being.

This must also be contextualized by considering that adolescence in Latin America is marked by contextual factors, such as higher rates of socioeconomic inequality, educational gaps, and high levels of violence and social stress [48,49]. Such conditions can influence how young people experience and develop dimensions such as autonomy, environmental mastery, or positive relationships. Therefore, using instruments developed in Anglo contexts without proper adaptations can lead to biased or partial interpretations of the Latin American reality [50].

1.3. The Methodological Challenges of Measuring Well-Being

Despite the conceptual and empirical advances in measuring well-being, scientific literature shows notable limitations in assessing it during adolescence, particularly from the perspective of Ryff’s eudaimonic approach. A first gap refers to the imbalance between theoretical traditions. While subjective well-being has been widely studied in the adolescent population through life satisfaction or positive and negative affect instruments, psychological well-being remains little explored in this age group [41].

A second limitation lies in the lack of suitable instruments for adolescents. Questionnaires derived from Ryff’s theory, such as the Scales of Psychological Well-being, were initially designed for use with adults. Although they have shown factorial consistency in multiple international studies, their application to adolescents poses methodological challenges. Several items require abstract reflection and long-term projection skills that may not be fully developed at early ages [40]. This could explain, at least in part, the scarcity of psychometric studies focused on adolescence.

Cultural adequacy represents an additional challenge. Ryff’s model has been applied in multiple countries, revealing variations in its factorial structure that depend on age and sociocultural context [25]. In Latin America, adaptations have been limited and, in many cases, partial, which limits the comparability of results and makes it difficult to draw solid conclusions about adolescent psychological well-being in the region [51]. Considering the cultural heterogeneity and the profound social inequalities present in Latin American countries, the absence of relevant instruments is a critical limitation for both research and the implementation of educational and health policies.

1.4. The Present Study

Research on PWB in adolescence has grown in recent years; however, significant conceptual and methodological challenges remain, particularly in Latin American contexts, where structural inequalities and sociocultural heterogeneity demand culturally sensitive approaches [39,51]. Although Ryff’s multidimensional model of PWB has been widely applied across diverse populations, its instruments were initially developed for adults, and their application in adolescents raises questions regarding developmental appropriateness and psychometric robustness [40,41]. Furthermore, evidence suggests that school constitutes a critical environment for fostering well-being, making it essential to have valid and reliable measures adapted to the educational context [42,43].

Consequently, the general objective of this study was to develop and validate a culturally pertinent and psychometrically robust scale of psychological well-being for adolescents in the Latin American school context. The specific objectives proposed were: (1) To develop indicators aligned with Ryff’s multidimensional model of psychological well-being; (2) To determine the content validity of the items through expert judgment; (3) To establish the factorial structure of the scale as evidence of construct validity; (4) To estimate the reliability of the instrument through internal consistency indices; (5) To assess measurement invariance.

2. Materials and Methods

2.1. Participants and Design

The research employed a cross-sectional, non-experimental design with an instrumental scope, aimed at developing and validating a psychological well-being instrument. Participants were 8638 adolescents from schools across the 16 regions of Chile. A non-probabilistic cluster sampling strategy was applied, and the total sample was randomly divided into two subsamples to perform exploratory and confirmatory analyses.

Participants ranged in age from 10 to 19 years (M = 13.91, Md = 14, IQR = 12–16). In terms of gender, 51.7% identified as male, 44.8% as female, 1.2% as non-binary, and 2.3% preferred not to disclose their gender. Most students attended urban schools (89.37%), while 10.63% were enrolled in rural schools. Regarding school type, 41.83% attended municipal schools, 47.22% attended subsidized private schools, 1.25% attended fully private schools, and 9.70% studied in institutions administered by Local Education Services.

School Selection and Participant Recruitment

The invitation to participate in the study was initially extended to representatives of the intermediate level of the Chilean education system—Regional Ministerial Secretariats of Education (SEREMI, in Spanish), Provincial Departments of Education (DEPROV, in Spanish), Municipal Education Administration Departments (DAEM, in Spanish), and Local Public Education Services (SLEP, in Spanish)—who received a formal letter and an informational dossier about the project. These bodies acted as territorial coordination channels for disseminating the invitation to educational establishments.

Subsequently, educational establishments in Chile’s 16 regions were invited, prioritizing geographical, administrative (municipal, SLEP, subsidized private, and fee-paying private), and socioeconomic diversity to achieve a broad and heterogeneous sample. Interested schools confirmed their participation by means of a Letter of Institutional Commitment, signed by the management teams, and designated a teacher or professional responsible for coordinating the application.

Students were recruited from participating schools on a voluntary and anonymous basis, after obtaining informed consent from parents or guardians and informed assent from the adolescents themselves.

2.2. Instruments

Psychological Well-Being Scale for Adolescents (PWBS-A)

The PWBS-A was explicitly designed for this study. It is composed of 32 items with affirmations derived from Ryff’s multidimensional model [24,25]. Items were written to reflect six core dimensions (self-acceptance, positive relations with others, autonomy, environmental mastery, purpose in life, and personal growth). Experts reviewed them to ensure clarity and developmental relevance. All items were rated on an 11-point Likert scale, which combined sections on frequency (0 = Never to 10 = Always) and agreement (0 = Totally disagree to 10 = Totally agree). Following the recommendations of Cummins and Gullone [52], only the extreme points were labeled to facilitate comprehension and reduce potential response bias. This response format allows capturing greater variability in participants’ answers and has been previously used in studies assessing psychological well-being among Chilean adolescents, proving both understandable and practical for this population [53,54]. Evidence regarding the psychometric properties of the instrument is presented in Section 3.

2.3. Procedure

The initial pool of items was submitted to a panel of five expert judges with backgrounds in psychology, education, and social sciences. The five expert judges were selected through purposive sampling, considering their academic and professional backgrounds in areas related to psychological well-being, psychometrics, and adolescent education. The team consisted of three psychologists with experience in positive adolescent development, school well-being, and psychometric instrument validation, and two academics from the field of education with postgraduate training in school assessment and guidance.

All the experts worked at Chilean and foreign universities with active research lines in well-being and school coexistence. They had indexed publications and experience in projects funded by the National Research and Development Agency (ANID) or the Chilean Ministry of Education.

Their selection sought to ensure disciplinary diversity and territorial representativeness, including professionals from universities in the center and south of the country. The combination of specialization in developmental psychology, education, and research methodology ensured a solid evaluation of the theoretical relevance, semantic clarity, and cultural appropriateness of the items for Chilean adolescents.

Experts evaluated clarity, coherence, relevance, and sufficiency [55] using a 4-point scale (ranging from 0 to 3, with 0 = “insufficient” and 3 = “sufficient”). They provided written recommendations. Based on this input, adjustments were made before pilot testing.

A pilot study was then conducted to test item comprehension, scale usability, and the adequacy of the 11-point response format. Minor revisions were made, after which the final version was administered. School recruitment took place between March and October 2024, with school leadership invited to participate in virtual meetings to explain project aims, procedures, and ethical safeguards. Schools signed Letters of Commitment, and the project obtained approval from the Scientific Ethics Committee (CEC code: 046_22). Institutional consent, parental consent, and student assent were all collected. Data were gathered online using the LimeSurvey platform, with technical support and training provided to staff responsible for implementation.

2.4. Data Analysis

For content validity, mean ratings were calculated for each item, with values of 2 or higher considered acceptable. Inter-rater agreement was estimated using Aiken’s V, with ≥0.80 taken as adequate. Pilot data were also inspected to ensure variability across the 10-point response scale.

Exploratory analyses included an Exploratory Factor Analysis (EFA) performed on a Pearson correlation matrix, given that responses were treated as continuous variables. Sampling adequacy was assessed with the Kaiser-Meyer-Olkin (KMO) test (≥0.80) and Bartlett’s test of sphericity (p < 0.001). The number of factors was determined by parallel analysis, retaining factors with eigenvalues exceeding 95% of those from random data. Estimation used Maximum Likelihood (ML), and factor rotation was performed with direct Oblimin.

In addition, an Exploratory Graph Analysis (EGA) [56] was conducted. This method identifies latent dimensions through network analysis of partial correlations, applying the Glasso model for regularization and the Louvain algorithm for community detection. Item retention was guided by three (criteria: factor loadings ≥ 0.300, absence of cross-loadings, and theoretical congruence with Ryff’s dimensions. In the context of EGA, the latent dimensions identified are referred to as communities, which are conceptually equivalent to the factors extracted in an EFA.

The most parsimonious and conceptually coherent model was then tested using Confirmatory Factor Analysis (CFA), estimated with the Maximum Likelihood Robust (MLR) method, which accounts for potential deviations from normality [57]. Model fit was judged using conventional standards [30,58]: Comparative Fit Index (CFI) and Tucker–Lewis Index (TLI) ≥ 0.90, Root Mean Square Error of Approximation (RMSEA) ≤ 0.06 with 90% confidence interval, and Standardized Root Mean Square Residual (SRMR) ≤ 0.08. Standardized factor loadings ≥ 0.400 were required.

Measurement invariance was tested through a sequence of increasingly restrictive models: configural (same form; factor structure), metric (weak; factor loadings), scalar (strong; factor loadings and intercepts), and residual invariance (strict; factor loadings, intercepts, and residuals). Each step adds equality constraints to the previous model, evaluating whether the measurement properties remain equivalent across groups. Model comparisons were based on changes in fit indices, with invariance supported when ΔCFI ≤ 0.010 and ΔRMSEA ≤ 0.015 [59].

Reliability was examined with Cronbach’s alpha (α) and McDonald’s omega (ω), with ≥0.70 indicating acceptable internal consistency and ≥0.80 strong reliability. Convergent validity was assessed using the Average Variance Extracted (AVE), with values of ≥0.50 considered evidence of satisfactory convergence. All analyses were performed in RStudio, version 0.6-19 [60], using psych for descriptive statistics and EFA, EGAnet for EGA, lavaan for CFA and invariance testing, and semTools for reliability and AVE calculations.

3. Results

3.1. Content Validity

The refinement of the instrument was based on an expert panel evaluation. In contrast to other validation processes where item elimination was necessary, all 25 original items were retained in this study. However, several modifications were implemented to improve clarity and semantic consistency. Specifically, adjustments were made to the phrasing of inverse items to avoid misinterpretation (see Supplementary Material).

Additionally, temporal references were standardized by rephrasing all items into the present perfect continuous form (e.g., “I have been feeling”) to achieve uniformity across the scale. Two items were split into separate statements because they originally contained mixed content, which could have undermined construct validity.

Moreover, five new items were incorporated following judges’ suggestions, aimed at broadening the scope of the instrument to include relevant aspects not sufficiently captured before, such as self-acceptance in relation to physical appearance.

3.2. Exploratory Factor Analysis

Preliminary tests supported the adequacy of the data for factor analysis. The Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy reached an excellent value of 0.97, and Bartlett’s test of sphericity was significant, χ²(496) = 98,227.46, p < 0.001, confirming that the correlation matrix was suitable for factor analysis. However, Mardia’s test indicated that the assumption of multivariate normality was not met, as both skewness and kurtosis statistics were significant (p < 0.05).

Parallel analysis suggested the extraction of seven factors, which together explained 46.4% of the total variance (

Table 1. Exploratory factor analysis loadings.

	Item	F1	F2	F3	F4	F5	F6	F7
1	Me ha gustado mi forma de ser en general I liked my personality in general	0.036	0.096	0.277	0.041	0.073	0.074	0.402
2	Me he sentido seguro/a de mí mismo/a I have felt confident in myself	0.106	0.074	0.186	0.036	0.046	0.079	0.546
3	Me han gustado la mayoría de mis características personales I have liked most of my personal characteristics	0.103	0.066	0.235	0.019	0.056	0.086	0.524
4	Me he sentido conforme con cómo me veo físicamente I have been satisfied with how I look physically	0.084	0.019	0.181	0.083	0.005	0.050	0.554
5	Me he sentido orgulloso/a de quién soy I have felt proud of who I am	0.004	0.027	0.679	0.055	0.018	0.046	0.178
6	Me he sentido orgulloso/a de las cosas que hago I have felt proud of the things I do	−0.041	0.064	0.786	0.037	0.030	0.045	0.014
7	Me he sentido orgulloso/a de lo que he logrado en mi vida I have been proud of what I have achieved in my life	0.086	0.018	0.760	0.035	0.049	−0.013	−0.009
8	He sentido que me llevo bien con la gente I feel that I get along well with people	−0.003	0.089	−0.024	0.183	0.406	0.027	0.289
9	He sentido que entre mis amigos/as nos ayudamos cuando lo necesitamos I feel that my friends and I help each other when we need it	0.011	0.020	0.008	−0.018	0.815	0.001	−0.009
10	He sentido que puedo confiar en mis amigos/as I have felt that I can trust my friends	−0.011	−0.037	0.021	−0.035	0.872	−0.009	−0.027
11	He sentido que mis amigos/as saben que pueden confiar en mí I have felt that my friends know they can trust me	0.050	0.052	0.023	0.023	0.698	0.049	−0.036
12	He sentido que le caigo bien a las demás personas I feel that other people like me	−0.017	0.071	−0.071	0.241	0.393	0.056	0.317
13	Mantengo mis decisiones aun cuando los demás no estén de acuerdo conmigo I stand by my decisions even when others disagree with me	0.026	0.199	0.002	−0.007	0.013	0.456	0.121
14	Me preocupa lo que otras personas pueden pensar de mí I worry about what other people might think of me	0.006	0.023	−0.026	0.008	0.133	−0.046	−0.202
15	Tomo decisiones sin pedir la opinión de los/as demás I make decisions without asking for other people’s opinions	0.015	0.016	−0.037	0.001	−0.032	0.383	0.088
16	Digo lo que pienso, aun cuando los/as demás no piensen igual que yo I say what I think, even when others don’t agree with me	−0.004	−0.040	−0.001	−0.003	−0.012	0.920	−0.044
17	Expreso mis opiniones, incluso si son diferentes a las de la mayoría de las personas I express my opinions, even if they differ from those of most people	0.009	0.033	−0.007	0.006	0.025	0.785	0.012
18	Puedo hacerme cargo de mis labores y responsabilidades diarias I can take care of my daily tasks and responsibilities	0.050	0.747	0.063	−0.040	0.001	−0.027	−0.094
19	Soy capaz de resolver problemas de la vida diaria I am capable of solving everyday problems	−0.005	0.814	−0.029	−0.009	0.002	0.036	0.065
20	Busco soluciones cuando tengo dificultades I look for solutions when I encounter difficulties	0.003	0.691	−0.013	0.095	0.021	0.023	0.021
21	Tomo decisiones sobre lo que hago en mi día a día I make decisions about what I do in my daily life	0.014	0.766	0.047	0.010	0.015	0.024	−0.043
22	Soy capaz de manejar mis emociones en mi vida diaria I am able to manage my emotions in my daily life	0.016	0.427	−0.018	0.161	0.027	−0.013	0.238
23	Me ha gustado pensar en lo que haré en el futuro I enjoyed thinking about what I will do in the future	0.822	0.022	−0.041	−0.060	−0.003	0.004	0.021
24	He pensado que en el futuro cumpliré con mis metas I have thought that in the future I will achieve my goals	0.844	0.007	−0.002	0.001	−0.003	0.006	0.049
25	Me he sentido bien con las metas que tengo en mi vida I have felt good about the goals I have in my life	0.699	0.007	0.103	0.051	0.032	0.030	0.024
26	He pensado en cosas que quiero lograr en mi vida I have thought about things I want to achieve in my life	0.790	0.038	−0.044	0.027	0.011	0.013	−0.041
27	He sentido que tengo algo que me motiva a seguir adelante I have felt that I have something that motivates me to keep going	0.519	−0.059	0.084	0.203	0.051	0.025	−0.020
28	Me he sentido bien cuando pienso en las cosas que he logrado I have felt good when I think about the things I have achieved	0.199	0.029	0.227	0.448	0.050	0.017	−0.038
29	He pensado que cada día aprendo más sobre mí mismo/a I have realized that every day I learn more about myself	0.024	0.065	0.042	0.730	−0.016	0.059	0.010
30	He sentido que día a día me hago mejor persona I feel that I am becoming a better person every day	0.003	0.036	0.051	0.668	0.057	0.056	0.096
31	He querido aprender cosas nuevas I have wanted to learn new things	0.176	0.131	0.034	0.449	0.053	0.013	−0.123
32	He pensado que si me esfuerzo puedo lograr lo que quiera I have thought that if I try hard, I can achieve whatever I want	0.256	0.097	0.093	0.423	0.026	0.020	−0.078

Note: Bold values indicate the highest loading for each item within its factor, representing the strongest association in the Exploratory Factor Analysis.

). Among the 32 items, none exhibited cross-loadings, and only item 14 did not load on any factor. The seven-factor solution yielded a structure that was largely consistent with Ryff’s theoretical model of psychological well-being. Items reflecting purpose and motivation (e.g., having goals and feeling motivated) loaded on Purpose in Life, while items related to problem-solving and management of daily responsibilities corresponded to Environmental Mastery. Items addressing continuous learning and personal improvement clustered under Personal Growth, and those capturing trust, support, and positive interpersonal ties were defined as Positive Relations with Others. The Items that emphasize independence and assertiveness loaded on Autonomy. The main divergence from the original model was observed in Self-Acceptance, which split into two related but distinct factors: one reflecting self-worth and accomplishment (e.g., pride in oneself and one’s achievements) and another focusing on self-liking and body image (e.g., satisfaction with personality and physical appearance).

3.3. Exploratory Graph Analysis (EGA)

The analysis suggested a five-community solution. The resulting communities also showed strong convergence with Ryff’s theoretical model of psychological well-being.

Community 1 grouped items reflecting Self-Acceptance, including satisfaction with one’s personality, appearance, and personal achievements. Community 2 included items related to Positive Relations with Others, emphasizing trust, support, and positive social interactions. Community 3 reflected Autonomy, although item 15 (“I make decisions without asking for other people’s opinions”) showed only a weak association with this dimension. Community 4 was consistent with Environmental Mastery, clustering items on problem-solving, managing responsibilities, and emotional regulation. Finally, Community 5 integrated elements of Purpose in Life and Personal Growth, combining items on goals, motivation, continuous learning, and personal development (See

Table 2. Exploratory Graph Analysis Loadings.

	Item	C1	C2	C3	C4	C5
1	Me ha gustado mi forma de ser en general I liked my personality in general	0.373	0.059	0.027	0.024	0.032
2	Me he sentido seguro/a de mí mismo/a I have felt confident in myself	0.480	0.053	0.046	0.056	0.017
3	Me han gustado la mayoría de mis características personales I have liked most of my personal characteristics	0.516	0.042	0.054	0.033	0.018
4	Me he sentido conforme con cómo me veo físicamente I have been satisfied with how I look physically	0.406	0.030	0.009	0.017	0.017
5	Me he sentido orgulloso/a de quién soy I have felt proud of who I am	0.569	0.017	0.004	0.003	0.064
6	Me he sentido orgulloso/a de las cosas que hago I have felt proud of the things I do	0.428	0.022	0.013	0.028	0.069
7	Me he sentido orgulloso/a de lo que he logrado en mi vida I have been proud of what I have achieved in my life	0.366	0.031	0.000	0.007	0.145
8	He sentido que me llevo bien con la gente I feel that I get along well with people	0.112	0.403	0.017	0.036	0.020
9	He sentido que entre mis amigos/as nos ayudamos cuando lo necesitamos I feel that my friends and I help each other when we need it	0.017	0.501	0.012	0.019	0.028
10	He sentido que puedo confiar en mis amigos/as I have felt that I can trust my friends	0.003	0.532	0.000	0.000	0.001
11	He sentido que mis amigos/as saben que pueden confiar en mí I have felt that my friends know they can trust me	0.012	0.419	0.026	0.046	0.067
12	He sentido que le caigo bien a las demás personas I feel that other people like me	0.103	0.394	0.025	0.041	0.058
13	Mantengo mis decisiones aun cuando los demás no estén de acuerdo conmigo I stand by my decisions even when others disagree with me	0.064	0.019	0.341	0.091	0.024
14	Me preocupa lo que otras personas pueden pensar de mí I worry about what other people might think of me	−0.028	0.000	0.011	0.000	0.002
15	Tomo decisiones sin pedir la opinión de los/as demás I make decisions without asking for other people’s opinions	0.010	0.002	0.209	0.011	0.002
16	Digo lo que pienso, aun cuando los/as demás no piensen igual que yo I say what I think, even when others don’t agree with me	0.020	0.020	0.630	0.025	0.021
17	Expreso mis opiniones, incluso si son diferentes a las de la mayoría de las personas I express my opinions, even if they differ from those of most people	0.030	0.031	0.545	0.074	0.030
18	Puedo hacerme cargo de mis labores y responsabilidades diarias I can take care of my daily tasks and responsibilities	0.000	0.018	0.017	0.357	0.102
19	Soy capaz de resolver problemas de la vida diaria I am capable of solving everyday problems	0.025	0.023	0.077	0.557	0.047
20	Busco soluciones cuando tengo dificultades I look for solutions when I encounter difficulties	0.012	0.021	0.052	0.427	0.119
21	Tomo decisiones sobre lo que hago en mi día a día I make decisions about what I do in my daily life	0.019	0.021	0.055	0.503	0.088
22	Soy capaz de manejar mis emociones en mi vida diaria I am able to manage my emotions in my daily life	0.097	0.050	0.012	0.227	0.061
23	Me ha gustado pensar en lo que haré en el futuro I enjoyed thinking about what I will do in the future	0.000	0.000	0.001	0.022	0.403
24	He pensado que en el futuro cumpliré con mis metas I have thought that in the future I will achieve my goals	0.012	0.000	0.000	0.039	0.523
25	Me he sentido bien con las metas que tengo en mi vida I have felt good about the goals I have in my life	0.046	0.012	0.026	0.035	0.476
26	He pensado en cosas que quiero lograr en mi vida I have thought about things I want to achieve in my life	0.000	0.000	0.003	0.039	0.483
27	He sentido que tengo algo que me motiva a seguir adelante I have felt that I have something that motivates me to keep going	0.018	0.018	0.006	0.004	0.379
28	Me he sentido bien cuando pienso en las cosas que he logrado I have felt good when I think about the things I have achieved	0.125	0.029	0.005	0.054	0.370
29	He pensado que cada día aprendo más sobre mí mismo/a I have realized that every day I learn more about myself	0.051	0.006	0.023	0.089	0.423
30	He sentido que día a día me hago mejor persona I feel that I am becoming a better person every day	0.107	0.092	0.014	0.064	0.319
31	He querido aprender cosas nuevas I have wanted to learn new things	−0.002	0.023	0.002	0.070	0.340
32	He pensado que si me esfuerzo puedo lograr lo que quiera I have thought that if I try hard, I can achieve whatever I want	0.025	0.010	0.014	0.069	0.391

Note: Bold values indicate the highest loading for each item within its community, representing the strongest association in the Exploratory Graph Analysis.

).

No items were cross-assigned to more than one community. However, item 14 (“I worry about what other people might think of me”) did not cluster with any dimension, suggesting poor representativeness. Overall, the EGA solution replicated the core structure of Ryff’s model, except that Purpose in Life and Personal Growth were merged into a single dimension.

3.4. Confirmatory Factor Analysis (CFA)

To evaluate the dimensionality suggested by the exploratory analyses, two confirmatory factor analyses (CFAs) were conducted. Because Mardia’s test indicated that multivariate normality (p < 0.05) was not met, all models were estimated using robust maximum likelihood (MLR), which applies the Yuan–Bentler correction [61].

The seven-factor model, based on EFA, demonstrated an excellent fit to the data: CFI = 0.946, TLI = 0.939, RMSEA = 0.055 (90% CI [0.053, 0.056]), and SRMR = 0.042. All standardized loadings were statistically significant and above 0.40, ranging from 0.42 to 0.89. The latent correlations were moderate to strong (0.55–0.90), indicating both conceptual distinctiveness and theoretical coherence. The structure reproduced Ryff’s six dimensions, with Self-Acceptance divided into two related but distinct factors (self-worth/accomplishment and self-liking/body image).

The five-factor model, derived from EGA, also provided a reasonable but comparatively weaker fit: CFI = 0.914, TLI = 0.905, RMSEA = 0.070 (90% CI [0.069, 0.072]), and SRMR = 0.049. Standardized loadings were all significant and generally high (ranging from 0.68 to 0.87). This structure grouped Purpose in Life and Personal Growth into a single factor, while retaining Environmental Mastery, Positive Relations with Others, Autonomy, and a unified Self-Acceptance dimension.

Comparing the two models, the seven-factor solution achieved a superior fit according to incremental and absolute indices (CFI, TLI, RMSEA), suggesting that it provides the most accurate representation of the underlying structure of adolescent psychological well-being in this sample. This model was prioritized for confirmatory analysis because it aligns more closely with Ryff’s multidimensional framework. The division of the self-acceptance dimension into two distinct factors refines an existing theoretical domain, whereas the five-factor structure derived from EGA would have merged conceptually distinct constructs. Therefore, the seven-factor model demonstrates both stronger theoretical consistency and clearer empirical interpretability.

3.5. Reliability and Convergent Validity

Reliability analyses revealed adequate to excellent internal consistency across the seven dimensions, with coefficients ranging from 0.779 to 0.906. Convergent validity was also supported, as indicated by AVE values ranging from 0.501 to 0.764 (see

Table 3. Reliability and convergent validity indicators.

	F1	F2	F3	F4	F5	F6	F7
α de Cronbach	0.891	0.822	0.906	0.894	0.876	0.779	0.899
Ω de McDonald	0.893	0.836	0.906	0.898	0.864	0.790	0.902
AVE	0.625	0.550	0.764	0.645	0.581	0.501	0.693

Note. AVE = Average Variance Extracted.

).

3.6. Measurement Invariance

Prior to testing measurement invariance, only groups with sufficient sample sizes were retained to ensure adequate statistical power. For gender, analyses included adolescents identifying as male and female. For school type, the analyses were conducted among municipal, subsidized private, and Local Education Service schools. These groups met the minimum power criterion of approximately n ≥ 320 per group (equivalent to 10 responses per item for the 32-item scale), minimizing the risk of unstable estimates and convergence issues in the multi-group analyses.

Measurement invariance analyses supported the equivalence of the seven-factor model of the PWBS-A across key demographic and contextual groups. Full residual invariance was achieved by gender, school location, and school type, with minimal changes in fit indices across nested models (ΔCFI ≤ 0.010; ΔRMSEA ≤ 0.015). For educational level, the model reached scalar invariance, indicating comparable factor loadings and intercepts between primary and secondary students, while the residual level was not achieved, which is acceptable for mean comparisons (see

Table 4. Measurement invariance by gender, educational level, and geographic context.

Gender (Male/Female)
Model	CFI	TLI	RMSEA	SRMR	∆CFI	∆RMSEA
Configural	0.949	0.942	0.052	0.040	-	-
Metric	0.948	0.942	0.052	0.042	−0.001	0.000
Scalar	0.941	0.937	0.054	0.042	0.007	0.002
Residual	0.939	0.937	0.054	0.042	0.002	0.000
Educational level (Primary/Secondary)
Configural	0.946	0.938	0.055	0.042	-	-
Metric	0.945	0.940	0.054	0.043	−0.001	−0.001
Scalar	0.944	0.940	0.054	0.044	−0.001	0.000
Residual	0.933	0.932	0.057	0.044	−0.011	0.003
School location (Urban/Rural)
Configural	0.945	0.938	0.054	0.041	-	-
Metric	0.947	0.942	0.053	0.042	0.002	0.001
Scalar	0.947	0.943	0.052	0.042	0.000	−0.001
Residual	0.946	0.945	0.053	0.044	−0.001	0.001
School type (Municipal/Subsidized private/Local Education Service)
Configural	0.947	0.940	0.054	0.042	-	-
Metric	0.947	0.942	0.055	0.043	0.000	0.001
Scalar	0.946	0.943	0.055	0.043	−0.001	0.000
Residual	0.945	0.943	0.056	0.044	−0.001	0.001

Note. CFI = Comparative Fit Index; TLI = Tucker–Lewis Index; RMSEA = Root Mean Square Error of Approximation; SRMR = Standardized Root Mean Square Residual; Δ = difference between nested models.

).

4. Discussion

The development of a new psychological well-being scale adapted to the Chilean context responds to the need for culturally and evolutionarily relevant measures that reflect the realities, values, and challenges of adolescence in Latin America. Most of the instruments derived from Ryff’s model were designed for adults in Anglo-Saxon contexts, which creates conceptual and linguistic limitations when applying them to Spanish-speaking adolescents. In Chile, factors such as socio-educational inequality, academic pressure, territorial differences, and the importance of the school environment as a space for socialization significantly influence the experience of well-being. These aspects require instruments that capture idiosyncratic dimensions of Chilean youth culture, such as body image, personal agency, and interpersonal relationships in diverse school contexts [41].

In addition, recent public policies aimed at students’ social and emotional well-being, such as the Política Nacional de Convivencia Educativa y la Estrategia Nacional de Salud Mental Escolar [62] require locally validated psychometric tools that allow progress to be monitored and evidence-based interventions to be guided. In this sense, the PWBS-A not only provides theoretical and methodological rigor but also responds to an institutional demand from the Chilean education system for instruments that are sensitive to adolescent development and comparable with international frameworks for sustainability and educational well-being.

The results of this study furnish compelling evidence for the validity and reliability of the Psychological Well-Being Scale for Adolescents (PWBS-A), presenting a significant theoretical, methodological, and practical contribution. The findings substantiate the relevance of Ryff’s multidimensional model within adolescent populations [24], while revealing a novel insight that self-acceptance is bifurcated into two interconnected yet distinct dimensions: self-worth/accomplishment and self-liking/body image. This refinement enhances the understanding of the construct and posits that, during adolescence, self-concept encompasses both achievement-related dimensions and physical appearance, a notion that has been only superficially explored in prior research [40,63]. This distinction aligns with research highlighting the significance of body image during adolescence, especially in Latin American contexts characterized by sociocultural and media pressures [48].

This differentiation within the self-acceptance domain can be interpreted in light of developmental theories of the self, which emphasize that adolescence is a period of increasing differentiation in self-concept and domain-specific self-evaluations [64,65]. During this stage, evaluations related to achievement and physical appearance become particularly salient and operate as partially independent sources of self-assessment and well-being. Longitudinal evidence indicates that both achievement-related self-perceptions and body-related self-evaluations contribute uniquely to adolescents’ psychological adjustment and self-esteem [66]. Moreover, contemporary perspectives on self-esteem recognize that self-acceptance comprises multiple evaluative components whose salience may vary across developmental and sociocultural contexts [67]. Within this framework, the emergence of the Self-Worth/Accomplishment and Self-Liking/Body Image dimensions in the PWBS-A represents a developmentally sensitive adaptation of Ryff’s model, capturing the complexity of adolescent self-evaluation processes.

From a methodological perspective, the integration of exploratory factor analysis (EFA), confirmatory factor analysis (CFA), and exploratory graph analysis (EGA) constitutes a robust and novel approach. EGA is a new tool that combines network psychometrics with dimensionality estimation [56], thereby strengthening traditional factor-analytic methods [30]. The evidence of measurement invariance by gender enhances the utility of the PWBS-A, as it guarantees that the observed differences between boys and girls are not mere artifacts of the measurement process [59]. This represents significant progress, considering the need to equitably compare subgroups in the formulation of school policies and interventions. Reliability indices were consistently robust, and convergent validity was established, underscoring the internal coherence of the scale.

Unlike prior studies that modified adult-oriented instruments for adolescent cohorts [40,51], this research created a tool specifically tailored for adolescents and culturally rooted in Chile. This corresponds with demands for contextually pertinent instruments in Latin America, where cultural diversity and socioeconomic disparities can influence adolescents’ perceptions of well-being [68]. The results align with recent studies that emphasize schools as pivotal environments for enhancing socioemotional skills and well-being [42,43,45]. Nonetheless, they emphasize that body image-related factors are particularly significant among Latin American youth, highlighting the need to incorporate both universal and culturally specific dimensions in the assessment of adolescent well-being [69].

The PWBS-A is a reliable tool for school-based assessment and program implementation, so the applied implications are clear. It can be utilized for institutional diagnostics, the formulation of socioemotional interventions, and the oversight of educational policies. Adding well-being indicators to academic performance may help people see the quality of education in a more complete way [29,70]. The PWBS-A could be useful for determining the effectiveness of school-based programs that address bullying, resilience, or social-emotional learning. These are all areas that have been linked to adolescent well-being [38,71].

The results obtained reinforce the need to understand adolescent psychological well-being not only as an individual construct, but also as an indicator of social and educational sustainability. The validation of the PWBS-A contributes directly to SDG 3 (Good Health and Well-being) by providing an empirically sound tool for monitoring youth mental health from a positive development perspective. Likewise, its application in school contexts allows progress toward SDG 4 (Quality Education), incorporating the socio-emotional dimension as an essential part of comprehensive learning and educational assessment [1,4,8].

The fact that the scale has demonstrated gender invariance and cultural appropriateness is relevant to SDGs 5 and 10, as it promotes equity in the measurement of well-being, recognizing the diversity and structural inequalities that characterize the adolescent experience. Furthermore, its use in schools across different regions of the country aligns with the principles of territorial justice and equal opportunities in accessing valid assessment tools [2,6,8].

In terms of public policy and sustainability, the instrument provides useful evidence to strengthen SDG 16 (Peace, Justice, and Strong Institutions) by enabling educational institutions to identify factors that protect well-being, coexistence, and social cohesion. Recent literature indicates that education for sustainability must be integrated from an early age in order to promote urban transitions based on knowledge, participation, and social co-responsibility, consistent with the framework of SDG 11 [15].

Likewise, schools that use well-being measurement as a management tool contribute to building safer, more inclusive, and more participatory school communities, in line with SDG 11 (sustainable cities and communities) and the notion of “learning cities” promoted by UNESCO [6,7,14] and the United Nations [8,9].

Recently, the findings of the Secretary-General’s Report on the Sustainable Development Goals 2025 [1] reaffirm the urgency of strengthening mental health, educational equity, and social cohesion as pillars of sustainability. In this context, the PWBS-A directly contributes to SDGs 3, 4, 5, 10, 11, and 16 by providing an empirical, culturally relevant, and psychometrically sound measure of adolescent psychological well-being. Its application in schools allows for monitoring progress toward a more inclusive, equitable education that promotes positive mental health, in line with the global priorities of the 2030 Agenda. Thus, the assessment of well-being becomes a strategic component of the educational and social transitions necessary to ensure more just, resilient, and sustainable societies.

Subsequent research ought to investigate the temporal stability of the PWBS-A via longitudinal methodologies, facilitating the examination of developmental trajectories of well-being [41]. It is also necessary to replicate the study in other Latin American countries to enhance cross-cultural comparability. Finally, incorporating the PWBS-A into intervention trials and policy evaluation studies would enhance its utility as a scientific and practical tool for promoting adolescent mental health.

It is important to recognize several limitations. First, the cross-sectional design inhibits causal interpretations and limits the examination of developmental changes in psychological well-being. Although the sample was large and covered all 16 regions of Chile, it was obtained through non-probabilistic cluster sampling based on voluntary school participation. Consequently, the schools and students included may not proportionally represent the full diversity of Chilean adolescents in terms of geographic distribution and socioeconomic background, which limits the generalizability of the findings to the national population.

Second, the reliance on self-report instruments may have introduced certain response biases, such as social desirability, acquiescence, or individual differences in interpreting item content. Self-report data reflect participants’ subjective perceptions, which can be influenced by mood, situational context, or social norms related to emotional expression [72]. Although items underwent meticulous examination for linguistic clarity, cultural and contextual subtleties may still have impacted responses. This limitation highlights the need to augment self-reports with multimethod approaches, including teacher or peer evaluations, behavioral indicators, or qualitative assessments, to triangulate findings.

Overall, these limitations suggest that the present findings should be interpreted as preliminary evidence of construct validity within the Chilean educational context. Cultural and contextual specificities therefore bound the generalizability of the results, and causal inferences should be avoided. Nonetheless, the large and diverse sample provides a solid empirical foundation for future replication and extension studies across different regions and populations.

5. Conclusions

This study provides a new multidimensional instrument to assess psychological well-being in Chilean adolescents, designed with both developmental appropriateness and cultural adequacy in mind. The PWBS-A represents a significant advancement over previous measures by capturing dimensions particularly relevant to adolescence—such as the dual structure of self-acceptance—and by demonstrating robust psychometric properties, including gender invariance.

Its applicability is broad, spanning institutional diagnostics, the design of preventive and promotive school-based interventions, and the monitoring of national and international educational and mental health policies. In a context where adolescent well-being is increasingly prioritized globally, the PWBS-A emerges as a novel and pertinent tool that bridges research and practice.

Thus, the development of the Psychological Well-Being Scale for Adolescents (PWBS-A) constitutes an empirical and conceptual contribution aligned with the 2030 Agenda for Sustainable Development, by providing a culturally relevant and psychometrically robust measure to assess adolescent well-being. This instrument contributes to SDG 3 by enabling the monitoring of mental health in school contexts; to SDG 4 by integrating socioemotional assessment into educational quality; and to SDGs 5 and 10 by promoting gender equality and reducing inequalities through inclusive measurement tools.

Likewise, its potential application in programs focused on school coexistence and socioemotional learning strengthens SDG 16 by fostering more peaceful and collaborative school environments, and SDG 11 by consolidating sustainable educational communities integrated into territorial support networks. Overall, the PWBS-A represents a bridge between research and practice, contributing to educational and health systems that are more equitable, resilient, and sustainable.

Future steps should include longitudinal and test–retest studies to capture developmental trajectories and ensure the temporal stability of the PWBS-A. Further efforts are needed to address cross-cultural validation in other Latin American regions, as well as to examine concurrent and predictive validity with key indicators such as academic performance, physical health, and psychosocial adjustment. In addition, it is essential to conduct sensitivity analyses to evaluate the scale’s responsiveness to school-based interventions and to develop normative reference values that enable the contextualized interpretation of scores across diverse populations. Finally, future research should strengthen the explicit connection between adolescents’ psychological well-being and the Sustainable Development Goals (SDGs), consolidating the PWBS-A as a scientifically sound and socially relevant instrument that contributes to more holistic educational systems and more equitable mental health policies in the region.