Students’ Perspectives on Social Sustainability: Developing an Integrated Higher Education Framework

Abstract

It has been widely acknowledged that HEIs have begun to play a critical role in effecting substantial societal change. This study aimed to investigate the association between perceived international medical students’ social sustainability and their satisfaction with academic life, within an integrated conceptual framework. A cross-sectional design was employed to collect data from 271 international undergraduate medical students from Romania. The data was analyzed with SEM in WarpPLS. The study’s findings indicated that social cohesion emerged as the strongest predictor and that inclusion contributed moderately to the students’ academic satisfaction, while HEI resilience had a small effect on the students’ academic satisfaction. Social cohesion enhanced HEI resilience and inclusion, and HEI resilience was associated with inclusion. Overall social sustainability was found to be a moderately significant predictor of students’ academic life satisfaction. The results of the study provide a valuable theoretical contribution by validating a second-order model of social sustainability in an HEI. The practical implications highlight the necessity of focusing on both the components and the overall social sustainability for the successful development of student-centered HEIs.

1. Introduction

Higher Education Institutions (HEIs), also referred to as agents of change or catalysts for addressing sustainability issues (Žalėnienė & Pereira, 2021), play an essential role in fostering transformative societal change, particularly through their missions and activities that contribute to the advancement of a more sustainable future (Caeiro & Azeiteiro, 2020). Education for sustainable development in HEIs aims to provide learners with the knowledge, skills, attitudes, and values necessary to make informed decisions and take responsible actions that promote environmental, economic and social sustainability (Tafese & Kopp, 2025).

Over 600 HEIs have formally committed to sustainability, yet many face challenges in operationalizing these principles due to limited institutional prioritization, resources, and staff engagement (Leal Filho et al., 2023; Machado & Davim, 2023). Despite these constraints, sustainability research in HEIs is well supported in several European countries, notably Spain, Germany, Sweden, and Portugal (Tafese & Kopp, 2025). For instance, HEIs have increasingly integrated sustainability into their systemic programs, although efforts have predominantly focused on environmental and economic dimensions, with comparatively limited attention to social sustainability (Gamage et al., 2022; Tafese & Kopp, 2025). This imbalance may stem from the challenges in evaluating the outcomes of social initiatives, managing inequality, and the complexity of advocacy efforts (Ankareddy et al., 2025), as well as from difficulties in defining and measuring social sustainability in HE settings (Larrán Jorge et al., 2018).

In Europe, the operationalization of social sustainability in HE is shaped by the commitment to the United Nations Sustainable Development Goals (de la Calle et al., 2021). According to Claeys-Kulik et al. (2019), frameworks like the European Education Area and the European Higher Education Area (EHEA) reveal that social cohesion, equity, and inclusivity are key components of HE performance and long-term resilience. In Western Europe, diversity and inclusion initiatives, performance indicators that specifically encompass equity metrics, and targeted funding for underrepresented groups are frequently used for social sustainability (Bhopal, 2018; Vukasović & Stensaker, 2017), while in Northern Europe, social sustainability is reflected in social cohesion, trust, and institutional legitimacy (Jónsson et al., 2024). In Eastern Europe, social sustainability is becoming more prevalent, but it is not regularly incorporated into educational frameworks (Curaj et al., 2020; Stăiculescu et al., 2022; Rocchi et al., 2022).

Romania is one of the Eastern European countries which presents a very low level of sustainable education (Rocchi et al., 2022), and top Romanian HEIs are still raising awareness about the concept (Șimon et al., 2020). Examining the evolution of sustainability and, especially, social sustainability in Romanian HEIs provides insights on how they develop and implement the principles of social sustainability under constrained conditions and structural barriers such as regional disparities in educational access, student disengagement, and the underrepresentation of vulnerable groups (Santa & Fierăscu, 2022; Ionașcu & Pînzariu, 2023; Voicu & Muntean, 2023).

Broadly, social sustainability is defined as education that fosters a just, equitable society, where all individuals can participate fully in social and democratic life (Wolff & Ehrström, 2020). In HEIs, social sustainability may be considered a utopia, which needs strong principle guidelines to make HE genuinely transformative (Wolff & Ehrström, 2020; Barnett-Itzhaki et al., 2025; Ankareddy et al., 2025). Social sustainability in HEIs may be embedded across curricula and operations and in specific activities about justice, equity, inclusion, social cohesion, democracy, safety, and well-being (Barnett-Itzhaki et al., 2025).

Although social sustainability in HEIs is complex and confusing due to its value-laden and content-dependent nature (Wolff & Ehrström, 2020; Ankareddy et al., 2025), it is necessary that all stakeholders of an HEI, particularly staff, faculty, students, government, employers, suppliers and community, are involved for the transformative change to occur and to ensure the relevance of the sustainable initiatives (Leal Filho et al., 2019; Shih et al., 2025).

Students are a central stakeholder group in HEIs, not only due to their numerical significance and role in institutional academic activities, but also because of their willingness to engage in sustainable practices (Boarin et al., 2020). Students are recognized as active agents of change in universities, as their perspectives critically shape the effectiveness of these efforts (Leal Filho et al., 2023; Machado & Davim, 2023). However, the students’ effective participation can be limited by insufficient understanding of the concept of social sustainability (Zeegers & Clark, 2014) and by the perceptions that social sustainability is secondary to their academic and operational priorities (Shih et al., 2025). Promoting awareness on how HEIs contribute to societal well-being has been shown to increase student engagement and their contributions to sustainability (Maiorescu et al., 2020). At an operational level, these contributions may work as feedback loops (H. Young & Jerome, 2020). For instance, student advocacy for inclusive measures, mental health resources, and student participatory governance often make HEIs introduce peer mentoring programs, advisory councils, or improve accessibility initiatives (Hudler et al., 2019; Slimmen et al., 2025), embedding social sustainability principles into their institutional culture (Basheer et al., 2025). At the societal level, students who experience cohesive and inclusive HEI environments are more likely to rely on these values in their professional activities, promoting broader initiatives of social responsibilities, connected to diversity, inclusion, social justice, and disability rights (Hudler et al., 2019). As examples of good practices of European Universities initiatives in contributing to a more social sustainable society, Arnaldo Valdés and Gómez Comendador (2022) selected the EELISA HEI Alliance, which aimed to create communities that would solve social challenges by acknowledging the engagement (attitude and implication) of a student, his/her participation within the EELISA community, as well as his/her contribution to the process of solving social challenges. Another example selected by the same authors is the YUFE Alliance, which consists of a Student Forum, with the aim of ensuring that student’s perspective is always taken into consideration during the implementation stage of the initiative. Therefore, the assessment and implementation of socially sustainable initiatives in HEIs, can address the systematic problems of students across all facets of the institutions (Leal Filho et al., 2023).

A comprehensive evaluation of sustainability integration in HEIs requires assessment tools that encompass both academic and administrative metrics (Barnett-Itzhaki et al., 2025). In the absence of standardized tools and frameworks, HEIs often face difficulties in measuring the effectiveness of their sustainability initiatives and in identifying the dimensions that need improvement (Umar et al., 2024), in this case, social sustainability. A large proportion of HEIs have, to some degree, implemented sustainability frameworks (Ramirez-Montoya et al., 2024) and assessed the effectiveness of sustainability based on them. The most widely endorsed sustainability frameworks in HEIs include The Sustainability Tracking, Assessment and Rating System, the Campus Environmental Audit Response Form, The Sustainability Assessment Questionnaire, and the College Sustainability Report Card (Gamage et al., 2022). Yet, their focus remains uneven, with an emphasis on environmental and economic performance, with unclear assessment of social sustainability in HEIs (Machado & Davim, 2023) or the social sustainability is assessed with a mixture of items about education, research, operations, and community engagement (Aleixo et al., 2018).

This lack of conceptual clarity of social sustainability in HEIs draws attention to the need for more comprehensive approaches that move beyond fragmented or indirect measures (Machado & Davim, 2023; Barnett-Itzhaki et al., 2025). In this regard, the social sustainability framework developed by Ballon and Cuesta (2024), which encompasses inclusion, resilience, social cohesion, and process legitimacy, offers an important theoretical foundation. In this context, it may be assumed that HEIs may operate as dynamic social systems in which social cohesion, inclusion, and HEI resilience, collectively shape students’ perceptions of social sustainability in an HEI, and are reflected in their academic satisfaction lives.

Social cohesion refers to the strength of relationships and the sense of solidarity among community members (Fonseca et al., 2019). Social cohesion in an HEI refers to the degree to which students feel a part of the larger academic and social community, as well as the level of interactions with other students, teachers, and administrative personnel. According to Zhang and Qian (2024), students are more likely to have a sense of commitment to their academic goals and a sense of belonging when they perceive a strong and supportive network of peers and personnel in their HEI.

Inclusion in an HEI is reflected in ensuring equal opportunities for engagement in academic and social life, irrespective of socioeconomic background, ethnicity, gender, or ability of students or personnel (Tinto, 2017; Altes et al., 2024). It has been demonstrated that inclusive practices, such as equitable resource distribution, inclusive teaching, and focused assistance for underrepresented groups of students, increase their satisfaction and retention (Thai & Alang, 2024).

Organizational resilience is the ability of an institution to adjust and prosper in the face of instability, uncertainty, and change (Duchek, 2020). This refers to HEIs being able to respond to crises or unstable periods and adjusting to long-term issues like financial constraints, and digital transformation (Rapanta et al., 2020; Gligorea et al., 2022). For students, HEI resilience is experienced through the stability and accessibility of learning opportunities, the continuity of academic programs and inclusive learning environment, and adaptable support systems during times of uncertainty (de Oliveira Durso et al., 2021; Hassan et al., 2024).

Student satisfaction with academic life may be a result of social sustainability initiatives. It shows how well students’ expectations and experiences align with their expectations across peer relationships, support services, teaching quality, and the overall environment in an HEI (Gruber et al., 2010).

Collectively, the fragmented nature of existing social sustainability initiatives reflects an ongoing global shift toward embedding it in HEIs’ structures and cultures (Annarelli et al., 2024). Yet, research assessing social sustainability from students‘ perspectives, using an instrument for examining the effectiveness of social sustainability in HEIs, remains limited, particularly in contexts with constrained conditions. This study aims to address these gaps by investigating the extent to which social cohesion, inclusion, and HEI resilience—recognized as key components of social sustainability—predict Romanian students’ perceived satisfaction with their academic experiences.

The specific objectives of the study are: (1) to elaborate a comprehensive measurement instrument for assessing social sustainability in a Romanian HEI; (2) to evaluate the perceptions of the students regarding the implementation of overall social sustainability and on its components in a Romanian HEI.

2. Theoretical Background

2.1. Social Sustainability in Higher Education (HE) and the Conceptual Framework

Social sustainability in HEIs may be considered the ability of a university to elaborate and implement academic activities which focus on accessibility, diversity, inclusion, and well-being of students and staff or faculty members (Barnett-Itzhaki et al., 2025). Social sustainability in HEIs is a multifaceted concept, with key topics referring to the human and relational aspects of university life, access equity, the quality of social interactions, and long-term human development (Eizenberg & Jabareen, 2017).

Despite recent extensive discussion in the literature (Wolff & Ehrström, 2020; Barnett-Itzhaki et al., 2025), empirical frameworks for assessing the effectiveness of social sustainability remain fragmented or limited, particularly in HE. The absence of conceptual clarity regarding social sustainability in HEIs highlights the necessity for more comprehensive approaches that should also take into consideration the perceptions of students, who represent a core stakeholder group in HEIs, and may become vectors of change (Machado & Davim, 2023). Their perceptions about social sustainability, reflected in their academic satisfaction, may uncover HEI systematic problems that can be solved by implementing social sustainability initiatives (Leal Filho et al., 2023).

Student academic satisfaction often includes opinions about the social environment, academic quality, institutional services, and personal growth (Gruber et al., 2010). According to Elliott and Shin (2002), satisfaction is frequently used as a performance metric, but it can also serve as an indicator of overall effectiveness of institutional sustainability. Research revealed that students’ satisfaction in an HEI is positively associated with social integration, inclusive pedagogies, and perceived support from faculty and administration (Thomas, 2012; Kahu, 2013). Thus, student satisfaction is not only an indicator of academic quality, but also a reflection of social conditions—particularly those related to belonging, fairness, and stability of an HEI.

The conceptual framework for this study is based on the multidimensional understanding of social sustainability in HEIs from the students’ perspectives, and on previous research, which includes three dimensions of social sustainability—social cohesion, inclusion, and HEI resilience—as antecedents to student satisfaction with academic life (Boström, 2012; Leal Filho et al., 2019; Fonseca et al., 2019; Ballon & Cuesta, 2024). These dimensions represent interrelated components of a socially sustainable academic environment, as follows:

• Inclusion assesses how much students believe the university environment is equitable, respectful of diversity, and accessible to all students, regardless of their backgrounds (Tinto, 2017; Bhopal, 2018);
• Social cohesion expresses students’ perceptions regarding connectedness, trust, and mutual support, as well as the support provided by faculty and administrative staff (Chan et al., 2006; Kahu, 2013);
• HEI resilience refers to students’ perceptions related to the institution’s capacity to adapt, and support students during challenges, and maintain continuity in academic services (Duchek, 2020).

Based on the presented information, the following hypothesis was developed:

H1: 

The perceived social sustainability in an HEI is positively associated with students’ academic life satisfaction.

2.2. Social Cohesion and Students’ Satisfaction with Academic Life

According to Chan et al. (2006), social cohesion is the extent of connections, trust, and shared values that exist within a society. In HEIs, social cohesion is characterized by interpersonal trust, peer collaboration, faculty-student relationships, and a general sense of belonging (Kahu, 2013; O’Keeffe, 2013). Students who feel included in the academic community are more motivated, have a lower chance of dropping out, and are satisfied with their academic careers (Thomas, 2012; S. Wilson et al., 2018). More specifically, higher perceived social cohesion enhances the sense of belonging and engagement among students, leading to greater satisfaction with their academic experience (Thomas, 2012; Kahu, 2013).

Based on the presented information, the following hypothesis was elaborated:

H2: 

Perceived social cohesion in an HEI is positively associated with students’ academic life satisfaction.

2.3. HEI Resilience and Students’ Satisfaction with Academic Life

As HEIs deal with ongoing issues like digitization, enrollment fluctuations, financial limitations, and student mental health, resilience has emerged as a key characteristic even before the post-pandemic period (Shaya et al., 2022). While upholding its dedication to inclusivity and cohesion, a resilient HEI may face transformative adaptation, which involves changing its policies, procedures, and support networks to satisfy the emerging needs of its students in overcoming challenges and enhance their academic experiences (Duchek, 2020).

Building upon the information presented, the following hypothesis was developed:

H3: 

Perceived HEI resilience is positively associated with students’ academic life satisfaction.

2.4. Inclusion and Students’ Satisfaction with Academic Life

Inclusion in an HEI refers to the process of ensuring that all individuals—regardless of socioeconomic background, ethnicity, gender, disability, or other identity markers—have equitable opportunities to access, participate in, and benefit from academic life (Tinto, 2017; Bhopal, 2018). Inclusive initiatives have been shown to enhance students’ perceptions of fairness, institutional trust, and overall satisfaction (Elliott & Shin, 2002; Gruber et al., 2010; Filippou et al., 2025).

Based on the information provided, the following hypothesis was formulated:

H4: 

Perceived inclusion in an HEI is positively associated with students’ academic life satisfaction.

2.5. Social Cohesion, Inclusion, and HEI Resilience

An HEI’s ability to better resist and recover from challenges is enhanced when there is mutual support, trust, and strong interpersonal relationships among its members, which is a fundamental aspect that connects social cohesion and HEI resilience (Putra & Istiyani, 2022). In times of crisis, such as the COVID-19 pandemic, social cohesion has a significant impact on community resilience during crises, highlighting how institutions can better support their members and recover from disruptions, when social connections are stronger (Jewett et al., 2021).

Students’ social and educational experiences are enhanced by fostering student connection and building inclusive networks and interpersonal relationships (Nesterova et al., 2020). This supports the idea that inclusive education necessitates creating a setting in which every student feels a sense of belonging, even though it primarily refers to access (Bello & Chumba, 2025).

The capacity of an HEI to establish settings that promote diversity, encourage inclusivity, and improve the overall educational experience for all stakeholders is an essential component that connects HEI resilience and inclusion. This implies that HEIs are in a better position to adopt inclusive practices when they have high resilience. Fostering an inclusive educational environment requires the ability to effectively react to the needs of students, especially those from marginalized backgrounds (Moran, 2016). HEIs increase access and foster a more creative and flexible community by developing systems that satisfy a variety of expectations (Wong et al., 2022).

Building upon the information presented, the following hypotheses were developed:

H5: 

Perceived social cohesion in an HEI is positively associated with the perceived HEI resilience.

H6: 

Perceived social cohesion in an HEI is positively associated with the perceived inclusion in an HEI.

H7: 

Perceived HEI resilience is positively associated with the perceived inclusion in an HEI.

Figure 1 illustrates the conceptual framework for the proposed hypotheses.

3. Materials and Methods

3.1. Design

This study, which was conducted at the “Carol Davila” University of Medicine and Pharmacy in Bucharest, Romania, between May and June 2025, employed a cross-sectional design. To efficiently select readily available participants and permit rapid data collection within resource constraints, a non-probability convenience sampling method was used.

The study was divided into two phases: a pilot study was conducted in the first phase, and the conceptual framework and the suggested hypotheses were tested in the second phase.

3.2. Participants

In the first phase of the study, the sample consisted of 30 international undergraduate medical students. Eligibility criteria required that participants were enrolled in one of the English programs at “Carol Davila” University of Medicine and Pharmacy in Bucharest, Romania, had completed at least two years of study, and provided informed consent to participate. International undergraduate students offer a critical perspective on social sustainability in HE, as their experiences with inclusion, equity, and social cohesion, shaped by adapting to new academic systems, language barriers and social integration, highlight the effectiveness of institutional support and engagement initiatives (Glass et al., 2014; Aksay Aksezer et al., 2023).

In the second phase of the study, the sample comprised 300 undergraduate medical students. The criteria of Kisakye et al. (2022) and the “10 times rule” for PLS-SEM (Hair et al., 2014), state that the minimum sample size should be at least ten times the maximum number of structural paths pointing at any latent variable. In this proposed framework, the maximum number of paths pointing at a single latent variable is 3, indicating a minimum sample size of 30 participants. Additionally, an a priori power analysis using G*Power version 3 for a medium effect size (f² = 0.15), α = 0.05, and statistical power of 0.80, suggested a minimum of 55 participants (Faul et al., 2007). The inclusion criteria were the same as those mentioned in the first stage of the study. Out of 300 students, 12 did not agree to participate in the study, and 17 participants did not complete more than 30% of the questionnaire, leading to a final sample of 271 students.

The student profile consisted of 49.4% male participants and 50.6% female participants, with 51.3% from EU countries and 48.7% from non-EU countries (

Table 1. The profile of the participants.

Characteristics	Frequency	Percent (%)
Gender
Male	134	49.4%
Female	137	50.6%
Country
EU	139	51.3%
Non-EU	132	48.7%

Note: EU—European Union.

). The mean age of the participants was 21.08 years (±1.49 standard deviations).

3.3. Procedure

3.3.1. The Pilot Study

A pilot study was conducted with 30 international undergraduate medical students from “Carol Davila” University of Medicine and Pharmacy to ensure the instruments’ clarity, accuracy, and cultural relevance. Items were initially tested using both 7-point and 5-point Likert scales. Since 78.3% of participants reported that using both Likert scales was challenging to interpret and that 5-point Likert scale and 7-point Likert scales have similar results (Colman et al., 1997), 5-point scales were used in the final version of the instruments, reducing also the response ambiguity (Lozano et al., 2008). The internal consistency of the instruments was assessed using Cronbach’s alpha coefficients, which showed acceptable reliability (α > 0.70) (Nunnally & Bernstein, 1994).

3.3.2. Instruments

The instruments included in the questionnaire were adapted from validated scales in the literature and applied in English. Adaptation involved rewording: (1) items were carefully rephrased, whenever necessary, to retain their original meaning and (2) items were modified for relevance to the HE context and student population. The evaluation of the rewording was assessed by the students included in the pilot study. The questionnaire consisted of five sections (Appendix A). The respondents’ age, gender, and country were the sociodemographic details gathered in the first section of the data collection instrument. Items for assessing an HEI’s perceived social cohesion were included in the second section, while statements about perceived HEI resilience were included in the third section. The fourth section, on the other hand, referred to items of perceived inclusion within an HEI. The fifth section consisted of items measuring students’ satisfaction with their academic lives.

• Perceived Social Cohesion instrument

The perceived social cohesion was measured by a selection of items from the University Belonging Questionnaire (Slaten et al., 2017). The 11 items referred to university support and acceptance, faculty and staff relations, as well as university affiliation, on a 5-point Likert scale, where 1 represented strongly disagree’ and 5 represented ‘strongly agree’. In this study, the reported Cronbach’s alpha coefficient value of the instrument showed satisfactory internal consistency (α = 0.93).

• Perceived HEI Resilience instrument

The perceived HEI resilience was measured using 7 items, with a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). The items were adapted from previous literature for an HEI context (Kantur, 2015; Chen et al., 2021). In this study, the Cronbach’s alpha coefficient value of the instrument revealed satisfactory internal consistency (α = 0.85).

• Perceived Inclusion instrument

Perceived inclusion was assessed using 7 items on a 5-point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree). The items were adapted from existing literature for an HEI context (C. Wilson & Secker, 2015; Ho & Browers, 2019). In the present study, the instrument demonstrated satisfactory internal consistency, with a Cronbach’s alpha of 0.87.

• Satisfaction with the academic life instrument

The instrument used to measure students’ satisfaction with their academic lives was a selection of six items from earlier research (Gür, 2024; Khatri et al., 2024). A 5-point Likert scale, with 1 indicating “strongly disagree” and 5 indicating “strongly agree,” was used to evaluate the items. The instrument in this study has a Cronbach’s alpha coefficient value over 0.70, indicating acceptable consistency (α = 0.86).

3.3.3. The Main Study

Self-administered paper questionnaires were used for collecting data. Participants received an informed consent form and a description of the study’s aims and objectives before completing the questionnaires. All responses were collected and processed anonymously, ensuring confidentiality throughout data collection, processing, and analysis. No financial incentives were offered.

3.4. Statistical Analysis

A descriptive analysis using frequencies and percentages was performed in SPSS (version 24) to determine the participants’ profile, and the mean and the standard deviation were used for the quantitative variables. To assess the reliability of the instruments and to test the research hypotheses, a Partial Least Squares (PLS) model was analyzed, using WarpPLS (version 8.0). PLS-SEM, a variance-based structural equation modeling (SEM) technique, is suitable for theory development (Hair et al., 2019). WarpPLS was selected for its ability to evaluate complex model structures, accommodating both reflective and formative constructs (Kock, 2024). Before model testing, potential Common Method Bias was assessed using Harman’s single-factor test (Podsakoff et al., 2003) and by full collinearity Variance Inflation Factors (Kock, 2015). Harman’s single-factor test revealed that the first factor of an unrotated EFA accounted for 30.96% of the variance, being below the threshold of 50% (Podsakoff et al., 2003). Also, all VIFs values were below the threshold of 3.3 (Kock, 2015). These findings revealed that the potential biases were unlikely to affect the results.

After the validation of the measurement model, the structural model was evaluated as part of the two-stage WarpPLS analysis. Examining internal consistency and reliability, as well as convergent and discriminant validity, were all part of the validation process for measurement models. Cronbach’s alpha coefficients were used to measure internal consistency, and composite reliability (CR) was employed to assess the reliability of the constructs (Bagozzi & Yi, 1988). Convergent validity was assessed using the Average Variance Extracted (AVE) for reflective constructs, along with their loadings and cross-loadings, while weights were examined for the formative constructs (Kock, 2024). The AVE square roots, which are part of the Fornell–Larcker criterion (1981), were used to evaluate discriminant validity. In support of further discriminant validity analysis, the Heterotrait-Monotrait (HTMT) ratio was used (Henseler et al., 2015).

The structural model analysis focused on estimating the relationships between latent variables, which referred to β path coefficients and their statistical significance p-values, coefficients of determination (R²), predictive relevance values (Q²), effect sizes (f²), and overall model fit and quality indices with specific thresholds (Kock, 2024). If Q² has a value around 0.02, it indicates a small predictive relevance; if Q² has a value around 0.15, it indicates a medium predictive relevance; and if Q² has a value around 0.35, it indicates a large predictive relevance (Hair et al., 2019). According to Cohen (1988), f² values of 0.02, 0.15, and 0.35 represent small, medium, and large effect sizes, respectively. A statistical significance level of p ≤ 0.05 was applied.

4. Results

As the central construct of the proposed model is social sustainability, which is composed of the first-order variables social cohesion, inclusion, and HEI resilience, a formative second-order model was developed. To validate the second-order model, a preliminary analysis was conducted on the first-order measurement instruments, being associated with the dependent first-order variable, satisfaction with academic life.

4.1. First-Order Constructs Analysis

• Validation of the measurement model

Cronbach’s alpha coefficients and Composite Reliability (CR) were used to evaluate the internal consistency and reliability of the first-order constructs. Analysis of

Table 2. Reliability and validity of the first-order constructs.

First-Order Constructs	α	CR	AVE	VIF	1	2	3	4
1. Social Cohesion	0.93	0.94	0.59	1.38	0.77	0.21	0.30	0.51
2. HEI Resilience	0.86	0.90	0.56	1.12	0.21	0.75	0.19	0.32
3. Inclusion	0.87	0.90	0.56	1.28	0.30	0.19	0.75	0.46
4. Satisfaction with academic life	0.86	0.90	0.60	1.68	0.51	0.32	0.46	0.77

Note: α—Cronbach’s alpha coefficient; CR—Convergent Reliability; AVE—Average Variance Extracted; VIF—Variance Inflation Factor; The square root of AVEs is shown in italics (the Fornell-Larcker criterion).

revealed that all Cronbach’s alpha coefficients and CR coefficients had values over 0.70 (Nunnally & Bernstein, 1994; Fornell & Larcker, 1981), suggesting that internal consistency and reliability of the first-order constructs were achieved.

The Discriminant validity was assessed with the Fornell-Larcker criterion and the Heterotrait-Monotrait Ratio. Also, the Construct Validity (CV) was determined by both the Average Variance Extracted values (AVE) and the matrix of loadings and cross-loadings (Kock, 2024). The AVE square roots had higher values than any of the correlations between the first-order constructs, as illustrated in Table 2 (Fornell & Larcker, 1981), while the HTMT values were below the accepted threshold of 0.85 (Henseler et al., 2015) (

Table 3. The Heterotrait-Monotrait (HTMT) values of the first-order constructs.

First-Order Constructs	Social Cohesion	HEI Resilience	Inclusion	Satisfaction with Academic Life
Social Cohesion	1
HEI Resilience	0.24	1
Inclusion	0.33	0.22	1
Satisfaction with academic life	0.57	0.38	0.53	1

), revealing a satisfactory discriminant validity.

The Convergent Validity of the first-order constructs was initially tested by the AVE values (Table 2). All AVE values were greater than the threshold of 0.5, indicating a good Convergent Validity (Kock, 2024). The second step in testing the CV of the first-order constructs consisted of the loadings and cross-loadings analysis (Appendix B 

Table A1. The loadings and cross-loadings of the first-order constructs.

Items	Social Cohesion	HEI Resilience	Inclusion	Satisfaction with Academic Life
SC1	0.80	−0.01	−0.003	−0.01
SC2	0.78	0.006	0.09	−0.06
SC3	0.75	0.007	−0.03	−0.02
SC4	0.78	−0.01	−0.07	−0.03
SC5	0.76	0.08	0.02	−0.06
SC6	0.76	0.006	0.01	−0.02
SC7	0.75	−0.10	−0.04	0.09
SC8	0.80	−0.08	−0.02	0.13
SC9	0.76	0.09	−0.06	0.05
SC10	0.77	−0.02	−0.01	0.07
SC11	0.75	0.04	0.10	−0.14
HEIR1	−0.05	0.82	0.03	0.04
HEIR2	−0.02	0.77	−0.003	0.03
HEIR3	−0.03	0.76	−0.11	0.07
HEIR4	0.000	0.49	−0.001	0.04
HEIR5	−0.007	0.77	0.06	−0.000
HEIR6	0.06	0.78	−0.01	−0.12
HEIR7	0.05	0.79	0.02	−0.06
I1	0.03	−0.07	0.81	0.07
I2	−0.05	0.10	0.77	−0.11
I3	−0.10	−0.03	0.74	0.15
I4	−0.02	−0.01	0.76	0.07
I5	0.03	−0.05	0.71	−0.04
I6	0.03	0.14	0.72	−0.11
I7	0.08	−0.07	0.71	−0.05
SA1	0.01	0.14	−0.003	0.77
SA2	0.06	−0.02	−0.11	0.77
SA3	−0.01	−0.03	0.01	0.73
SA4	−0.08	−0.12	0.01	0.79
SA5	−0.04	−0.01	0.05	0.80
SA6	0.06	0.05	0.03	0.77

Note: SC—Social Cohesion; HEIR—HEI Resilience; I—Inclusion; SA—Satisfaction with the academic life; The loadings are emphasized in italics.

). The findings from Appendix B Table A1 reinforced that the first-order constructs had good CV because the loadings of the items on the designated factors had higher values than the loadings on other latent constructs and over the accepted threshold of 0.50, except HEIR4 with a slightly lower loading of 0.49. It was decided to keep this item because it did not significantly improve the CR or the AVE (Kock, 2024; Hair et al., 2019).

• Structural model analysis

Several fit and quality indices were examined to evaluate the structural model (Kock, 2024) (

Table 4. Model fit and quality indices of the first-order constructs.

Indices	Validation Criteria
Average path coefficient (APC) = 0.27, p = 0.001	p < 0.001
Average R-squared (ARS) = 0.20, p = 0.001	p < 0.001
Average Adjusted R-squared (AARS) = 0.20, p = 0.001	p < 0.001
Average block VIF (AVIF) = 1.09	It is acceptable if the obtained value is ≤5, ideally ≤ 3.3
Average full collinearity VIF (AFVIF) = 1.37	It is acceptable if the obtained value is ≤5, ideally ≤ 3.3
Tenenhaus GoF (GoF) = 0.34	It is acceptable if the obtained value is: - Small ≥ 0.1; - Medium ≥ 0.25; - Large ≥ 0.36.
Sympson’s paradox ratio (SPR) = 1.000	It is acceptable if the obtained value is ≥0.7, ideally 1
R-squared contribution ratio (RSCR) = 1.000	It is acceptable if the obtained value is ≥0.9, ideally 1
Statistical suppression ratio (SSR) = 1.000	It is acceptable if the obtained value is ≥0.7
Nonlinear bivariate causality direction ratio (NLBCDR) = 1.000	It is acceptable if the obtained value is ≥0.7

Note: the obtained values were computed in WarpPLS version 8.0 and followed the guidelines of Kock (2024).

). The model established direct, positive, and moderate associations, as indicated by the APC, which had a value of 0.27. The SC, I, and HEIR account for 20% of the variance in students’ satisfaction with their academic lives, according to the ARS coefficient value of 0.20. Multicollinearity was not a problem, as indicated by the model fit indices AVIF and AFVIF, which had acceptable values and were below the suggested criteria. The model’s overall quality was confirmed to be moderate to strong, as indicated by the GoF coefficient of 0.34. The model’s robustness was further confirmed by the values of 1.000 for SPR, RSCR, SSR, and NLBCDR, which showed no obvious problems with statistical suppression or causal directions.

Consequently, the R² of the constructs, their path coefficients (β), the effect sizes (f²), and their predictive relevance values (Q²) were taken into consideration. Figure 2 displays the path coefficients, determination coefficients, and their corresponding p-values. According to

Table 5. The hypotheses validation regarding the first-order constructs.

Hypothesis	Path Coefficient (β)	R2	Q2	f2	Status
H2: SC-SA	0.37 **	-	-	0.19	Accepted
H3: HEIR-SA	0.21 **	-	-	0.08	Accepted
H4: I-SA	0.30 **	-	-	0.14	Accepted
H5: SC-HEIR	0.22 **	-	-	0.05	Accepted
H6: SC-I	0.26 **	-	-	0.08	Accepted
H7: HEIR-I	0.23 **	-	-	0.06	Accepted
SA	-	0.43	0.43	-	-
I	-	0.15	0.15	-	-
HEIR	-	0.05	0.05	-	-

Note: SC—Social Cohesion; HEIR—HEI Resilience; I—Inclusion; SA—Satisfaction with academic life; R²—Coefficient of determination; Q²—Stone-Geisser coefficient; f²—effect size; ** p-value < 0.001.

, Social Cohesion predicted the students’ academic satisfaction (β = 0.37; p = 0.001), with a medium effect size (f² = 0.19). HEI Resilience predicted students’ academic satisfaction (β = 0.21; p = 0.001), with a small effect size (f² = 0.08), while Inclusion predicted the students’ academic satisfaction (β = 0.30; p = 0.001), with a medium effect size (f² = 0.14). Thus, the strongest predictors of student satisfaction with academic life are social cohesion and inclusion. Regarding other established relationships, Social Cohesion predicted both HEI Resilience (β = 0.37, p = 0.001, f² = 0.05) and Inclusion (β = 0.37, p = 0.001, f² = 0.08), having small effect sizes. At the same time, HEI Resilience predicted Inclusion (β = 0.37; p = 0.001; f² = 0.06), with a small effect size.

The coefficients of determination (R²) revealed that 43% of the variance of the Satisfaction of students with the academic life was explained by Social Cohesion, HEI Resilience and Inclusion, and had a moderate explained variance and a large predictive relevance (R² = 0.43; Q² = 0.43), Social Cohesion explained 5% of the variance of HEI Resilience and had a small predictive value (R² = 0.05; Q² = 0.05), while Social Cohesion and HEI Resilience explained 15% of the variance of Inclusion and had a medium predictive relevance (R² = 0.15; Q² = 0.15) (Table 5).

4.2. Second-Order Constructs Analysis

Since the first-order measurement constructs were validated, the multidimensional construct of Social Sustainability was built. The Social Sustainability construct was a formative second-order construct, which required measurement and structural validation. Firstly, the VIFs were evaluated and had values below the recommended threshold of 3.3 (Kock, 2015), suggesting that there were no multicollinearity problems between the first-order constructs of Social Sustainability (

Table 6. Second-order measurements.

Construct	First-Order Construct	VIF	Weight	SE	Loading
Social Sustainability (Formative)	Social Cohesion	1.13	0.50 **	0.056	0.82
	HEI Resilience	1.07	0.42 **	0.057	0.88
	Inclusion	1.12	0.48 **	0.056	0.84

Note: ** p-value < 0.001; VIF—Variance Inflation Factor; SE—Standard error.

). Secondly, the weights of the first-order constructs were analyzed, with the recommendation that they should have values higher than 0.50 and be statistically significant (Hair et al., 2019). All first-order constructs had significant weights, but only Social Cohesion had a value higher than 0.50, HEI Resilience and Inclusion had values slightly lower than 0.50. Although HEI Resilience and Inclusion had values lower than 0.50, they were retained because both weights were statistically significant (p < 0.001), had low VIFs, and their loadings on the second-order construct were high (0.88 and 0.84, respectively) (Becker et al., 2012; Hair et al., 2019) (Table 6).

The second-order structural model analysis referred to the following (Hair et al., 2019): (1) path coefficients (β) (Figure 3); (2) R²—the determination coefficient (Figure 3); (3) predictive relevance values (Q²), and (4) the effect size (f²). Social sustainability was a moderate predictor of student satisfaction with their academic life (β = 0.64, p = 0.001), accounting for 41% of the variance (R² = 0.41). The Q² value and the f² value were 0.40, suggesting that Social Sustainability had a large effect size and a strong predictive relevance. Thus, H1 was accepted.

Table 7. Model fit and quality indices of the second-order model.

Indices	Validation Criteria
Average path coefficient (APC) = 0.63, p = 0.001	p < 0.001
Average R-squared (ARS) = 0.40, p = 0.001	p < 0.001
Average Adjusted R-squared (AARS) = 0.40, p = 0.001	p < 0.001
Average full collinearity VIF (AFVIF) = 1.65	It is acceptable if the obtained value is ≤5, ideally ≤ 3.3
Tenenhaus GoF (GoF) = 0.47	It is acceptable if the obtained value is: - Small ≥ 0.1; - Medium ≥ 0.25; - Large ≥ 0.36.
Sympson’s paradox ratio (SPR) = 1.000	It is acceptable if the obtained value is ≥0.7, ideally 1
R-squared contribution ratio (RSCR) = 1.000	It is acceptable if the obtained value is ≥0.9, ideally 1
Statistical suppression ratio (SSR) = 1.000	It is acceptable if the obtained value is ≥0.7
Nonlinear bivariate causality direction ratio (NLBCDR) = 1.000	It is acceptable if the obtained value is ≥0.7

Note: The obtained values were computed in WarpPLS version 8.0, following the guidelines of Kock (2024).

displays the results of the analysis of the second-order model’s fit and quality indices (Kock, 2024). The second-order model’s variables had a direct, positive, and strong association, as indicated by the APC of 0.63. A moderate amount of the students’ satisfaction with their academic lives was attributed to social sustainability, as indicated by the ARS and AARS values of 0.40. The constructs were not associated, and the results may be regarded as credible, with an AFVIF value of 1.65. The GoF value of 0.47 suggested that the model fitted well overall, while the values of SPR, RSCR, SSR, and NLBCDR supported the validation of the second-order model’s robustness.

5. Discussion

This study extends the theoretical understanding of social sustainability in HE by developing an integrated framework, as being perceived by undergraduate students. While prior research on social sustainability analyzed it as a secondary objective of transnational HE alliances (Arnaldo Valdés & Gómez Comendador, 2022) or as an example of a living lab environment, where students, faculty, and staff collaborate on real-world challenges (Purcell et al., 2019), this study assesses the effectiveness of social sustainability in an HEI from the perspective of students and in association with their academic satisfaction. Moreover, this study uses an integrated framework which consists of three components- social cohesion, inclusion and HEI resilience. In doing so, it integrates constructs into a second-order model, which in the literature were studied in isolation, and it provides new insights into the theoretical topic of student satisfaction in sustainable universities.

Firstly, the findings emphasize that social cohesion was the strongest predictor of students’ academic satisfaction. Although there was a moderate association, the result is consistent with previous research and contributes to the development of theory by extending the understanding of social cohesion beyond social integration. The present finding aligns with the existing literature, which suggests that supportive peer networks, trust, and shared values are fundamental drivers of positive academic experiences (Poole et al., 2023), while research by Hussain et al. (2025) emphasizes the established relationship between social cohesion and students’ overall academic performance and emotional health. The study indicated that students who are socially engaged are less likely to feel isolated and more likely to succeed academically (Hussain et al., 2025). In the same vein, Rehman et al. (2024) asserted that social cohesion is closely linked to students’ sense of belonging and identity, both being important for academic success. Also, a sense of belonging has been demonstrated to enhance students’ emotional well-being, thereby encouraging active participation in academic and extracurricular activities (Gheorghe et al., 2024). This, in turn, serves to reinforce their satisfaction with academic life.

Secondly, the role of inclusion in predicting students’ academic satisfaction was moderate but extends the existing inclusion theories in HE. Although the concept of inclusion has been extensively researched in the context of equity and diversity, there is a limited body of literature that has examined its association with student satisfaction. The findings of the present study revealed that from students’ perspectives, inclusive practices are a measurable dimension of organizational effectiveness (Shore et al., 2010; C. Wilson & Secker, 2015), and implicitly, of social sustainability in HEIs. If students perceive a high sense of welcome and support, they will have an increased level of academic satisfaction (Sakız & Jencius, 2024).

Thirdly, while HEI resilience emerged as a predictor of students’ academic satisfaction, this relationship was small, which was in accordance with the existing literature, especially the one which focuses on COVID-19 and academic activities. Although students’ perceived HEI resilience may contribute to their satisfaction with academic life, the small established relationship suggests that resilience is one of many factors of satisfaction and it often acts in the background. For many students, more immediate academic activities are important in shaping their satisfaction. From a social sustainability perspective, these activities may be reflected in responsive teaching environments, active learning methodologies, and the existence of supportive faculty departments and structures (M. R. Young et al., 2023; Cai & Meng, 2025). Furthermore, HEI resilience may buffer negative experiences or crises but might not be visible in everyday aspects of academic life, unless those crises are present.

Moreover, the findings revealed that social cohesion can also be a predictor for HEI resilience and inclusion. However, these associations are small because social cohesion may act as a context and not be defined by concrete initiatives. For instance, social cohesion refers to feelings of trust, belonging and shared values among peers or community, but these do not always transform into concrete inclusive behaviors or HEI resilience policies (Manitsa et al., 2023). It is asserted that inclusion initiatives may exist in formal structures, but their effectiveness is amplified only when there are strong social ties and a sense of belonging (C. Wilson & Secker, 2015). Similarly, in resilience research, social cohesion does contribute to community or HEI resilience during times of crises, otherwise is more diffused.

Moreover, the present study revealed that HEI resilience may be a significant predictor of inclusion, but with a small effect. The concept of HEI resilience offers a theoretical framework for understanding inclusion as a dynamic and adaptable construct during a period of crisis. In contexts of change or crisis, resilient HEIs have shown the capacity to reconfigure structures and practices to maintain equitable access to resources, and support services for students (Probst, 2022).

The findings of the study also reveal that social sustainability is a moderate predictor of the students’ satisfaction with academic life, which is in accordance with the existing literature (Wolff & Ehrström, 2020; Barnett-Itzhaki et al., 2025; Ankareddy et al., 2025). Student satisfaction is multifaceted, encompassing academic quality but also experiences of social inclusion, campus climate, and the availability of supportive resources. When HEIs actively promote social sustainability, students are more likely to feel valued, engaged and connected to the campus community. These perceptions, in turn, may create a feedback loop (H. Young & Jerome, 2020); students who feel their needs and voices are listened to are more likely to advocate for further HEI improvements. Over time, this reciprocal process not only enhances organizational culture and campus life but also equips students with the values and skills to promote social sustainability principles beyond the HEI context (Hudler et al., 2019).

The findings of this study offer several practical implications for HEIs seeking to increase social sustainability initiatives, reflected in their turn, in the enhanced students’ academic satisfaction. By highlighting the medium and small predictive roles of social cohesion, inclusion, and HEI resilience, the results suggest targeted interventions that university policymakers can adopt to enhance student well-being.

Enhancing social cohesion may be considered a pivotal objective for HEIs. Social cohesion has been identified as the strongest predictor of students’ academic satisfaction, emphasizing the significance of cultivating trust, connectedness, and a sense of belonging within the academic community. To strengthen social cohesion, it is recommended that universities implement peer mentoring programs, collaborative learning environments, student learning hubs, and clubs, as well as community-building activities that foster mutual support among students (M. R. Young et al., 2023).

The integration of inclusive initiatives into learning environments can be achieved through a systematic approach. The considerable impact of inclusion on student satisfaction emphasizes the need for HEIs to shift from a largely compliance-driven approach to diversity policies to a proactive implementation of inclusion strategies. This may involve redesigning curricula in a manner that reflects a diversity of perspectives, implementing assessment practices free from bias, and ensuring equitable access to resources for all students, including those from marginalized groups.

The concept of HEI resilience can be strengthened to not only overcome crises but also promote student confidence in institutional stability. The capacity of resilience of a

HEI can be supported by engaging students in the co-creation of an adaptive approach, thereby involving them as active partners in promoting institutional sustainability (Duchek, 2020).

The findings of this study suggest that HEIs should adopt a holistic approach to social sustainability, given the multidimensional nature of student satisfaction. Nowadays, a limited focus on academic quality or employability outcomes is inadequate. Instead, HEI policymakers should recognize that relational and organizational factors, such as peer support, inclusivity, and resilience, also shape student satisfaction. The incorporation of these components into the sustainability initiatives of HEIs has been evidenced to enhance their long-term institutional reputation and competitiveness (Probst, 2022).

Research has highlighted the significance of participatory governance, whereby students actively engage in decision-making processes, as primary contributors to the effectiveness of European universities (Klemencic, 2014). In the case of Romania, where student participation in institutional governance is often characterized by formality rather than substance, social cohesion is frequently associated with informal networks and peer support structures (Curaj et al., 2020). Moreover, in the Romanian system, inclusion has been formalized through affirmative action and the provision of student support services, yet, persistent inequalities exist—particularly between students residing in urban and rural areas (Stăiculescu et al., 2022). The implementation of inclusive policies, in isolation, has been demonstrated to be inadequate in the absence of a concomitant institutional culture that offers support (Varga et al., 2021). Furthermore, in Romania, efforts to build HEI resilience are still in their early stages of development, often limited to digital infrastructure or financial contingency planning. However, EU-funded capacity-building projects indicate an emerging recognition of HEI resilience as a component of sustainability (Curaj et al., 2020).

The recommendations for HEIs to achieve social sustainability consist of the following practices:

• For ensuring inclusion, HEIs should provide equitable access for all students, including marginalized or underrepresented groups, and elaborate and implement inclusive curricula and accessible teaching methods;
• For ensuring Social Cohesion, HEIs should promote trust, collaboration, and a sense of belonging of students to the academic community, as well as encourage them to participate in mentorship programs and extracurricular activities;
• For ensuring HEI Resilience, it should be developed flexible academic teaching and administrative structures and robust support systems, to adapt to disruptions, while promoting inclusion and social cohesion;
• For ensuring long-term organizational social sustainability, HEIs should align the organizational sustainable goals to a broader social sustainability program of the EU or of a country. They should also ensure that the principles of equity, inclusion, and cohesion are integrated into their strategic planning and policies;
• For ensuring student satisfaction based on social sustainability principles, HEIs should engage students and staff members in the decision-making processes, policy development, as well as in the co-creation of institutional solutions. Based on evidence-based approaches, it is also recommended to measure and continually improve programs and interventions dedicated to social sustainability by promoting mental health, emotional support, and the inclusion of students.

Limitations and Further Research Directions

Despite its contributions, this study has several limitations that should be acknowledged. The present study relied on cross-sectional data, which captures relationships at a single point in time, which raises issues of generalizability. It is recommended that future research employs longitudinal designs to examine how these constructs evolve and also to provide more substantial evidence of causality (Podsakoff et al., 2012).

Moreover, the study was conducted within a single HEI, which may limit the generalizability of findings to other cultural, geographical, or institutional settings. A comparative study encompassing a range of countries and university types (public versus private, large versus small) could offer insights into how context influences the dynamics of social sustainability, HEI resilience, inclusion, social cohesion, and students’ academic satisfaction (Altbach et al., 2019).

The data were collected using self-reported questionnaires, which have been demonstrated to be subject to biases such as social desirability bias or standard method variance (Podsakoff et al., 2003). Although statistical examinations were conducted to mitigate these issues (e.g., Common Method Bias and VIF values), it is recommended that future studies employ a combination of self-reported data alongside objective or behavioral indicators (e.g., retention rates, academic performance, or institutional records) to enhance the validity of the research.

The convenience sampling method presents some limitations, as for instance selection bias and overrepresentation of certain groups of participants. It is recommended for further research to use larger sample sizes and mixed-methods approaches, which may offer the possibility of capturing several perspectives of the participants who are underrepresented in convenience samples.

The conceptual model demonstrated a moderate explained variance. While this finding is considered acceptable within the Social Sciences, it suggests that other relevant factors which influence students’ academic satisfaction should be included in the model; some of the factors are digital readiness, financial security, or academic self-efficacy (Aman et al., 2023). It is recommended that future research employs mixed-methods approaches, integrating qualitative methods such as interviews or focus groups, to uncover students’ understanding of how they perceive HEI resilience, social cohesion, and inclusion, or social sustainability in their academic environments (Creswell & Plano Clark, 2018).

6. Conclusions

This study highlights the essential role of social cohesion, HEI resilience, and inclusion as key predictors of students’ satisfaction with academic life, emphasizing their contribution to social sustainability in HEIs. Social cohesion emerged as the strongest predictor, supporting the importance of building an academic environment based on trust, collaboration, and a sense of belonging, with the aim of shaping positive student experiences. Inclusion and HEI resilience were also significant predictors, with resilience showing a smaller effect, indicating that both equitable initiatives and adaptable institutional structures are important for fostering supportive activities for students. To strengthen the development and implementation of this framework, future research should validate it across diverse HEIs and cultural contexts, ensuring its relevance in varied education settings. Integrating social coherence, inclusion and HEI resilience may contribute to the creation of socially sustainable and student-centered universities.