Teacher connection and student perspective-taking and their association with school belonging among adolescents in 62 countries

ABSTRACT

Addressing worldwide declines in adolescents’ school belonging requires identifying factors to target in reversing this trend. We hypothesized that adolescents with higher levels of teacher connection and perspective-taking would have greater school belonging and that the joint association of both factors with school belonging would be more than additive. We analysed cross-sectional survey data from 15-year-olds in 62 countries (N = 392,096) collected in the 2018 Programme for International Student Assessment. Measures were expressed in standard deviation (SD) units. Based on regression models with both factors plus covariates, school belonging increased 0.12 (95% CI: 0.11, 0.13) units for a 1.00 SD increase in teacher connection and 0.09 units (95% CI: 0.08, 0.10) for a 1.00 SD increase in perspective-taking. The observed (joint) association of both factors with school belonging was 19.0% greater than the expected (additive) association. Efforts to increase teacher connection and perspective-taking could improve school belonging.

KEYWORDS:

• Adolescents
• students
• teachers
• school belonging
• perspective-taking
• empathy

Introduction

The need to belong is a fundamental human drive that is characterized by stable interpersonal attachments that influence one’s cognitive and emotional states (Baumeister & Leary, 1995). People feel a sense of belonging to a group when they perceive relational value – a sense that others in that group value their relationship with them as much as they do (Leary, 2021). A sense of belonging is essential when adolescents are developing the capacities and skills needed for making contributions with and towards others in an expanding social world (Fuligni, 2019). Learning how to balance self- and other-oriented motives and developing healthy intergroup relations are central to becoming a contributing member of society (Crone & Fuligni, 2020).

Adolescents can experience belonging across a variety of contexts, and school belonging is an important contributor to adolescent well-being. School belonging, commonly defined as how much students feel ‘personally accepted, respected, included, and supported by others – especially teachers and other adults’ in their school community (Goodenow & Grady, 1993, pp. 60–61) has been declining globally over the last decade (OECD, 2019). Additionally, adolescents have reported an increased sense of loneliness and more mental health problems (e.g. higher levels of anxiety and depression) that have only worsened during the COVID-19 pandemic (Farrell et al., 2023; Racine et al., 2021; Twenge et al., 2021; Wolf & Schmitz, 2023), with long periods of quarantine and reduced social and physical contact with others contributing to feelings of isolation (Brooks et al., 2020; Longobardi et al., 2020). Taken together, this suggests there is a global crisis of disconnection among adolescents. However, students with greater school belonging are psychologically healthier (e.g. fewer anxiety and depression symptoms), perform better in school (e.g. higher levels of academic achievement and motivation; higher levels of engagement; less absenteeism and dropout), are at lower risk for violence and substance abuse, and perceive their classrooms as being more fair and focused on mastery (rather than performance oriented) (Allen et al., 2018; Korpershoek et al., 2020; Rose et al., 2022; Shochet et al., 2006). Therefore, improving school belonging appears to be a pathway to addressing this disconnection.

Students who feel a sense of belonging in school experience a psychological state of attachment or connectedness to the school community (Wang & Degol, 2016), including experiencing psychological safety. Psychological safety in schools can be created through relationships with teachers that are safe, stable, and nurturing. Teachers have often been conceptualized as attachment figures for students in schools, allowing students to both safely explore challenges and receive emotional co-regulation (Roorda et al., 2017; Spilt & Koomen, 2022; Wentzel, 2016), with the importance of these relationships continuing through adolescence (García-Moya, 2020; Roorda et al., 2017; Spilt & Koomen, 2022). Creating and sustaining relationships requires relational capacities, such as an awareness and acceptance of self and others, in both the adolescent and teacher that together enable adolescents to feel safe, secure, and valued (Whitaker et al., 2022). It is therefore possible that school belonging could be increased through efforts to strengthen student-teacher relationships and improve students’ relational capacities, leading to greater psychological safety and connection within school communities.

Past research has identified positive student-teacher relationships as a strong predictor of school belonging (Allen et al., 2018), with these relationships generally characterized by students perceiving qualities such as warmth, support, empathy, interest, and closeness from the teacher (Centers for Disease Control and Prevention, 2009; García-Moya, 2020; Roorda et al., 2017). Responsive, caring behaviours and attunement from teachers are also important dimensions of the student-teacher relationship for adolescents (Whitehead et al., 2021). Several single country studies and two international studies using adolescent samples have documented positive associations between student-teacher relationships and school belonging (Ahmadi et al., 2020; Allen et al., 2022; Bakchich et al., 2022; Chiu et al., 2015; Cortina et al., 2017; Kiefer & Pennington, 2017; Saroughi & Cheema, 2022; Waters et al., 2010; Wong, Konishi, et al., 2019; Wong, Parent, et al., 2019; Zimmer-Gembeck et al., 2006). However, there is also evidence that adolescent growth, learning, and development may be best supported by a more holistic approach to the student-teacher relationship, called a developmental relationship. A developmental relationship goes beyond an emotional connection to include teachers challenging students, sharing power with them, and expanding their possibilities (Li & Julian, 2012; Scales et al., 2020; Wentzel, 2016). As conceptualized by the Search Institute, a developmental relationship is comprised of five dimensions or elements (express care, challenge growth, provide support, share power, and expand possibilities) and is bidirectional in terms of the contribution from and benefit to each person. The relationship functions to ensure that students feel cared for, heard, and valued (Scales et al., 2022). In a U.S. sample, Scales and colleagues showed that developmental relationships with teachers were significantly associated with students’ sense of belonging (Scales et al., 2020), but we know of no international study that has examined the association of school belonging with a measure of teacher connection that aligns with the five dimensions of a developmental relationship.

Past research has also identified individual factors, such as coping skills and self-esteem, that are associated with greater school belonging (Frydenberg et al., 2009; Sirin & Rogers-Sirin, 2004). Examination of these individual factors is important because these capacities are developing during adolescence. As it relates to school belonging, students’ perspective-taking is a developing relational capacity that is understudied. Also referred to as cognitive empathy, perspective-taking is the ability to adopt the psychological viewpoint of others (Davis, 1983; Hollarek & Lee, 2022), and ‘belonging to a group requires understanding the minds of others’ (Crone & Fuligni, 2020, p. 455). Perspective-taking, along with empathic concern (i.e. affective empathy), comprise the relational capacity of empathy and is likely necessary to develop and sustain safe, stable, and nurturing relationships with others (Whitaker et al., 2022). Warm and supportive parents who model the capacity by acknowledging their child’s perspective and responding to their emotional needs can shape the development of cognitive and affective empathy during childhood and adolescence (Barnett, 1987; Boele et al., 2019). Additional social influences, such as those from peer and teacher relationships as well as from the classroom environment (i.e. one in which students are encouraged by teachers to share their opinions), have also been shown to affect perspective-taking. The relationship between these social influences and perspective-taking is bidirectional, meaning that one’s perspective-taking capacity can influence as well as be influenced by these social relationships (Hollarek & Lee, 2022). While higher levels of perspective-taking have been shown to be associated with better quality friendships (Smith & Rose, 2011), we identified only one study that examined perspective-taking as a predictor of school belonging. In this study of 1,990 Slovenian adolescents, perspective-taking was positively associated with school belonging (Veldin & Kozina, 2022). Further examination beyond a single country study of the association between perspective-taking and school belonging is needed.

Using cross-sectional data from the 2018 Programme for International Student Assessment (PISA) collected from 15-year-olds living in six continents, we determined whether teacher connection (assessed through a developmental relationships framework) and perspective-taking were associated with school belonging. In addition to examining individual associations between school belonging with teacher connection and perspective-taking, we also examined the joint association of teacher connection and perspective-taking with school belonging. We hypothesized that adolescents with higher levels of teacher connection and perspective-taking would have greater school belonging and that the joint association of both factors with school belonging would be more than additive.

Methods

Study population and survey design

Data for this study were obtained from PISA 2018, a school-based survey of youth that occurs every three years among Organisation for Economic Co-operation and Development (OECD) countries and non-member partner countries. The overall aim of the survey is to assess the knowledge and skills that 15-year-olds possess to meet the challenges of our modern world. Detailed survey methods are described elsewhere and summarized here (OECD, n.d.-a). Each survey iteration focuses on one of three subject areas, and the 2018 PISA focused on reading. In addition to assessing the application of content knowledge, a student questionnaire allows adolescents to provide their views on their family, personal, and school experiences. PISA uses a two-stage, stratified sampling strategy in which schools are sampled, followed by the sampling of students within schools. Eligible students must be enrolled full- or part-time in public or private, academic or vocational schools and have completed at least six years of formal schooling. Each participating country/economy designates a national project manager who assists in developing and validating the assessment items and facilitates a high quality implementation. Thirty-seven OECD countries and 42 partner countries/economies participated in PISA 2018 (OECD, 2020, n.d.-b). We used data only from those countries/economies that included student questionnaire items to measure teacher connection, perspective-taking, and school belonging, leaving a sample of 465,210 adolescents across 62 countries/economies.

Measures

Teacher connection

We used five items from the student questionnaire to construct our measure of teacher connection (PISA, 2017). Each item was selected to align with a dimension or element of the developmental relationships framework created by the Search Institute (Search Institute, 2021): 1) ‘I felt that my teacher understood me’ (express care); 2) ‘The teacher tells me in which areas I can still improve’ (challenge growth); 3) ‘The teacher made me feel confident in my ability to do well in the course’ (provide support); 4) ‘The teacher listened to my view on how to do things’ (share power); 5) ‘The enthusiasm of the teacher inspired me’ (expand possibilities). All items, except for ‘challenge growth,’ asked students to think about their language teacher and their past two language lessons and indicate their level of agreement with each item on a Likert-type scale, ranging from 1 (‘strongly disagree’) to 4 (‘strongly agree’). The ‘challenge growth’ item asked students to indicate how often each thing happened in their language lessons on a Likert-type scale, ranging from 1 (‘never or almost never’) to 4 (‘every lesson or almost every lesson’). We included only those with complete data on all five items, with higher scores indicating greater teacher connection. The internal consistency (Cronbach’s alpha) of the teacher connection score items in our sample was .80. To facilitate interpretation of the results, we standardized the teacher connection score.

Perspective-taking

We used the standardized, PISA-constructed measure (PERSPECT) to assess perspective-taking, which consisted of four items from the perspective-taking subscale of the Interpersonal Reactivity Index (IRI) (Davis, 1983, OECD, n.d.-d; PISA, 2017), a widely used measure of empathy. The items used in PISA were worded the same as those from the IRI, although the response scale differed slightly. In PISA, students were asked to indicate how well each statement described them, on a Likert-type scale ranging from 1 (‘very much like me’) to 5 (‘not at all like me’): 1) ‘I try to look at everybody’s side of a disagreement before I make a decision’; 2) ‘I believe that there are two sides to every question and try to look at them both’; 3) ‘I sometimes try to understand my friends better by imagining how things look from their perspective’; 4) ‘Before criticizing somebody, I try to imagine how I would feel if I were in their place.’ All items were reverse-coded so that higher scores indicated greater perspective-taking. In our sample, the internal consistency (Cronbach’s alpha) of the perspective-taking score items was .82.

School belonging

We used the standardized, PISA-constructed measure (BELONG) to assess school belonging in students, which included six items (OECD, n.d.-d; PISA, 2017). Students were asked to indicate their level of agreement with the following statements on a Likert-type scale, ranging from 1 (‘strongly agree’) to 4 (‘strongly disagree’): 1) ‘I feel like an outsider (or left out of things) at school’; 2) ‘I make friends easily at school’; 3) ‘I feel like I belong at school’; 4) ‘I feel awkward and out of place in my school’; 5) ‘Other students seem to like me’; 6) ‘I feel lonely at school.’ Three items were reverse-coded so that higher scores indicated greater school belonging. In our sample, the internal consistency (Cronbach’s alpha) of the school belonging score items was .79.

Covariates

We included five covariates, each considered a potential confounder in the association of school belonging with teacher connection or perspective-taking: age; gender; index of economic, social, and cultural status (ESCS); perceived emotional support from parents; and country. Age (calculated in years from date of birth) and gender (male [1] or female [0]) were reported by adolescents on the student questionnaire. The standardized, PISA-constructed index of ESCS is a composite score based on 33 items from the student questionnaire used to determine highest parental education (two items for mother, two items for father), highest parental occupation (two items for mother, two items for father), and home possessions (25 items) (OECD, n.d.-d). The standardized, PISA-constructed measure of perceived emotional support from parents (EMOSUPS) consists of three items that asked students to indicate their level of agreement with the following statements on a Likert-type scale, ranging from 1 (‘strongly disagree’) to 4 (‘strongly agree’): 1) ‘My parents support my educational efforts and achievements’; 2) ‘My parents support me when I am facing difficulties at school’; 3) ‘My parents encourage me to be confident.’

Statistical analysis

The pragmatic goal of our study was to establish the unconfounded association of school belonging with two factors that might ultimately be modifiable by school-based interventions. Therefore, our analytic approach was designed to investigate the independent and joint associations of school belonging with teacher connection and student perspective-taking, after accounting for possible confounding variables. Our analytic sample included 392,096 of the 465,210 (84.3%) respondents after excluding 73,114 who had missing data on either teacher connection, perspective-taking, and/or school belonging. Of the 392,096 respondents included in this analysis, 32,529 (8.3%) were missing data on one or more covariates, so we first imputed these missing values with sequential regression imputation (Raghunathan et al., 2001) to create 20 imputed data sets. We ran our regression models on these imputed datasets and reported model parameters aggregated across the datasets (Graham et al., 2007). Stata/MP version 15.1 (Stata Corp) was used for analyses. We used the Stata ‘svyset’ command with two PISA variables (international school ID [CNTSCHID] and student weight [W_FSTUWT]) to account for the complex sampling design (OECD, n.d.-c). The threshold for statistical significance was p < .05.

Sample descriptive statistics were tabulated, and Pearson correlations were used to assess the bivariate associations among the study variables. Standardized measures (z scores) were used for school belonging, teacher connection, perspective-taking, the ESCS index, and emotional support from parents. We used multivariable linear regression to examine the associations of students’ school belonging (dependent variable) with teacher connection and perspective-taking (key independent variables), while controlling for all five potentially confounding covariates (age, gender, emotional support from parents, ESCS index, and country). We also examined the joint association of teacher connection and perspective-taking with school belonging. An interaction term (teacher connection score x perspective-taking score) was added to our model to evaluate the interaction (synergy), as departure from additivity, between these two factors (Knol et al., 2007). We then used regression-based margins, standardized to the distribution of the covariates in the study population, to estimate covariate-adjusted school belonging scores for five groups defined by teacher connection score (−2.0, −1.0, 0.0, 1.0, and 2.0) and perspective-taking score (−2.0 to 2.0 at 0.5 increments). In secondary analyses, we used the full model (teacher connection, perspective-taking, interaction term, and covariates) to examine associations between school belonging with teacher connection and perspective-taking in each of the 62 countries/economies.

Results

The mean (SD) age of the 392,096 survey respondents in the analytic sample was 15.8 (0.3) years, and 48.9% of the adolescents were male (Table 1). The mean (SD) standardized scores for school belonging, teacher connection, and perspective-taking were −0.12 (0.90), 0.08 (0.96), and 0.06 (0.98), respectively; these mean values varied across the 62 countries/economies (Supplementary Table 1). The magnitude of the correlation was similar between school belonging and both teacher connection (r = 0.148, p < .001) and perspective-taking (r = 0.151, p < .001), while the correlation between teacher connection and perspective-taking was lower (r = 0.116, p < .001).

Table 1. Correlations between study variables (pre-imputation).

Study variables	1	2	3	4	5	6
1. School belonging	—
2. Teacher connection	0.148**	—
3. Perspective-taking	0.151**	0.116**	—
4. Emotional support from parents	0.268**	0.204**	0.167**	—
5. Index of ESCS	0.102**	−0.020**	0.073**	0.132**	—
6. Age,y	0.012**	−0.003	0.016**	0.004*	−0.003	—
7. Gender (female = 0; male = 1)^a	−0.003*	−0.018**	−0.122**	−0.086**	0.010**	−0.002
n	392,096	392,096	392,096	360,410	391,058	392,096
Mean	−0.12	0.08	0.06	−0.06	−0.85	15.81
SD	0.90	0.96	0.98	1.00	1.24	0.29

Note: Unless noted otherwise, all measures are standardized scores expressed in standard deviation (SD) units. ESCS: economic, social, and cultural status.

*p < .05, **p < .001.

^aPercentage (95% confidence interval) male is 48.9 (48.2, 49.6); N = 392,096. Point biserial correlation used to assess bivariate associations between gender and other study variables.

Independent associations of school belonging with teacher connection and perspective-taking scores

Before adjusting for covariates, teacher connection and perspective-taking scores were each significantly associated with the school belonging score (Table 2, Model 1). These significant associations persisted after controlling for emotional support from parents and other covariates (Models 2 and 3). A 1.00 SD increase in teacher connection score was associated with a 0.12 SD unit (95% CI: 0.11, 0.13) higher school belonging score, and a 1.00 SD increase in perspective-taking score was associated with a 0.09 SD unit (95% CI: 0.08, 0.10) higher school belonging score (Model 3).

Table 2. Associations of school belonging with teacher connection and perspective-taking scores.

	Model 1^a			Model 2^b			Model 3^c
	β	t	p	β	t	p	β	t	p
Teacher connection (TC) score	0.14	33.18	<.001	0.12	32.33	<.001	0.12	31.84	<.001
Perspective-taking (PT) score	0.12	33.32	<.001	0.09	25.80	<.001	0.09	25.83	<.001
TC score x PT score	—			—			0.04	10.78	<.001
Model R^2	.046			.134			.135

TC: teacher connection; PT: perspective-taking.

^aModel 1 is unadjusted and contains no covariates (N = 392,096).

^bModel 2 contains the following five covariates: age; gender; emotional support from parents; index of economic, social, and cultural status; and country.

^cModel 3 contains the interaction term (TC score x PT score) and the following five covariates: age; gender; emotional support from parents; index of economic, social, and cultural status; and country.

Joint association of teacher connection and perspective-taking scores with school belonging

Figure 1 illustrates the positive interaction or synergy by showing the joint association of perspective-taking (shown at 0.5 intervals) and teacher connection (shown at five different levels) with school belonging. The magnitude of this synergy can be interpreted using the standardized regression coefficients and comparing the observed versus expected effects of teacher connection and perspective-taking (Table 2, Model 3). The observed (joint) effect of teacher connection and perspective-taking on school belonging was 0.25 SD units (sum of β = 0.12 for teacher connection, β = 0.09 for perspective-taking, and β = 0.04 for interaction), while the expected (additive) effect (without the contribution of the interaction) was 0.21 SD units. Thus, the synergy or interaction contributes to a 19.0% increase in the belonging score above the expected effect (i.e. interaction effect divided by expected effect = 0.04/0.21). Those with teacher connection and perspective-taking scores of 1.00 would have a 0.25 SD higher score on school belonging than those with teacher connection and perspective-taking scores of 0.00.

Figure 1. Joint association of perspective-taking and teacher connection scores with school belonging score.

Legend: Lines are derived from a linear regression model showing the association between perspective-taking score (at 0.5 SD intervals) and predicted school belonging score (95% confidence interval) at 5 different levels of increasing teacher connection score from -2.0 to 2.0. The regression model controlled for the following five covariates: age; gender; emotional support from parents; index of economic, social, and cultural status; and country.

Country-specific independent and joint associations of school belonging with teacher connection and perspective-taking scores

In secondary analyses, there was a significant positive association between school belonging and teacher connection in all countries/economies (adjusted β range from 0.04 to 0.23) (Supplementary Table 2). In all but two countries/economies, there was a significant positive association between school belonging and perspective-taking (adjusted β range from 0.03 to 0.19). The interaction term was significant in 32 of the 62 (51.6%) countries/economies.

Discussion

Key findings

In an international sample of almost 400,000 15-year-olds, we have shown, as hypothesized, that students reporting higher levels of teacher connection and perspective-taking had greater school belonging. We also found evidence of synergy between perspective-taking and teacher connection in so far as the observed (joint) association of both factors with school belonging was greater than the expected (additive) association. Although this synergy was apparent in the overall sample, it was statistically significant in just over half of the 62 countries/economies examined. The unique contribution of the study arises from the examination of two potentially modifiable factors that could contribute to school belonging, the international scope of the sample, and the use of five items to assess teacher connection that reflect each of the key dimensions of teachers’ developmental relationships with students.

Findings in context

Our study builds on existing research establishing a positive association between teacher connection and school belonging, including among PISA and non-PISA samples. We are aware of seven prior studies that utilized PISA data, clearly articulated items for their constructs, and found positive, significant associations between teacher connection and school belonging (Allen et al., 2022; Bakchich et al., 2022; Chiu et al., 2015; Cortina et al., 2017; Saroughi & Cheema, 2022; Wong, Konishi, et al., 2019; Wong, Parent, et al., 2019). Two of the studies used PISA 2018 data (Allen et al., 2022; Bakchich et al., 2022), although only for Australia or France, respectively. Two of the studies used a larger sample of PISA, with one using PISA 2003 data from 31 countries (Cortina et al., 2017) and another using PISA 2000 data from 41 countries (Chiu et al., 2015). Our study extends these findings by using the developmental relationships framework to assess teacher connection and examining its association with school belonging using PISA 2018 data from 62 countries/economies. Several studies outside of PISA also found positive, significant associations between teacher connection and school belonging (Kiefer & Pennington, 2017; Scales et al., 2020), with only one utilizing the developmental relationships framework (Scales et al., 2020). Among 534 adolescents in grades sixth through eighth from one U.S. middle school, Scales and colleagues (Scales et al., 2020) assessed student-teacher relationships from the student’s perspective using the Search Institute’s 20-item measure of developmental relationships. This measure included three to five items for each dimension of developmental relationships and was significantly associated with greater belonging at both the beginning and end of the school year.

To our knowledge, research examining the association between student perspective-taking and school belonging is limited. As previously mentioned, we identified only one study, which was conducted among Slovenian adolescents, that examined perspective-taking as a predictor of school belonging. That study used the IRI to measure perspective-taking and six items from PISA 2018 to assess school belonging. Results from structural equation modelling found a positive, direct association between perspective-taking and school belonging (Veldin & Kozina, 2022). Our study extends these findings using a large, international sample. We not only showed that higher levels of perspective-taking were associated with greater school belonging, but we demonstrated for the first time the joint association of perspective-taking and teacher connection with school belonging.

Limitations

The primary limitations of the study are related to its cross-sectional design, the measures utilized, and possible residual confounding. We can make no causal inference from the cross-sectional design. Even reverse causality cannot be excluded because it is possible, for example, that perceiving a higher level of school belonging could alter both teacher connection and perspective-taking. The measure of teacher connection is not validated. Although the items had high internal consistency in our sample, they have not been compared to a fuller measure of teacher developmental relationships, such as that created by the Search Institute (Scales et al., 2020). In addition, these teacher connection items only reference a student’s perception of a single teacher (the language teacher). It is possible that a given student could have other teachers with whom there was greater connection. Given the importance of even one strong connection with a teacher, the focus on only the language teacher might lead to an under-estimation of the association of teacher connection with school belonging. The items used to estimate perspective-taking are a subset of items used in the well-validated measure of cognitive empathy in the IRI (Davis, 1983), but do not assess the affective dimension of empathy (empathic concern). Thus, our measure of perspective-taking is not a full measure of empathy, and empathy may be the more complete latent construct related to school belonging. Residual confounding may have resulted in an overestimation of the strength of the association of school belonging with either perspective-taking, teacher connection, or the synergy of these two factors. For example, including in our models a more complete measure of parent-child connection and social position or a measure of peer connection, which was not available in the data set, could have weakened the observed associations of school belonging with teacher connection and perspective-taking. Lastly, PISA was designed to be representative of 15-year-olds attending educational institutions (OECD, n.d.-a), so findings cannot necessarily be generalized beyond those who were studied.

Implications

Our study showed positive associations between school belonging and both teacher connection and perspective-taking in a large, international sample. These data support the possibility that school-based efforts to modify these two factors could improve adolescents’ school belonging. However, we also suspect that for such efforts to be effective, they must be implemented within the context of a school community that recognizes the role of trauma (past and present) and its impact on the psychological safety of all community members (Whitaker et al., 2022). For example, students and teachers who have experienced trauma may have more trouble connecting and yet need it most. We speculate that teacher connection and perspective-taking work together to create a sense of psychological safety within adolescents. When adolescents feel ‘safely seen’ by others (teacher connection) and have the capacity to ‘see and understand’ others (perspective-taking), they can feel psychologically safe. With this safety, adolescents are better able to build authentic connections with others and experience school belonging. We believe that the ability to create this psychological safety calls for school leaders to make relationships a shared core value in the school community (Whitaker et al., 2022).

Caring and emotionally supportive relationships are foundational for positive student-teacher relationships. In developmental relationships with teachers, students feel cared for and supported and also share power with and are challenged by their teachers in ways that expand their possibilities, broaden their world, and support their development to become capable adults (Scales et al., 2022). Developmental relationships involve a holistic and genuine interest in the student that can contribute to the student being ‘safely seen’ by teachers in full view of their life experiences, cultural context, and evolving identity. Importantly, developmental relationships are not only for the benefit of adolescents. The reciprocal nature of developmental relationships enables adolescents and adults to learn and grow from each other over time through lasting connections (Scales et al., 2022). These relationships occur within an interconnected ecosystem of learning and development in which adolescents, families, teachers, administrators, and staff continually influence one another (Akiva & Robinson, 2022). In this interconnected ecosystem, adults need to experience developmental relationships with other adults, allowing them to build these kinds of relationships with adolescents (Pekel, 2017; Whitaker et al., 2022). The experience of psychological safety must be felt across the school community in order for belonging to be authentic and shared.

There is existing evidence that teacher connection (or the student-teacher relationship) is modifiable. For example, in a pilot study of Simple Interactions, a strengths-based programme that focuses on fostering connection, reciprocity, inclusion, and growth in adult-child relationships, adult participation in the programme was significantly associated with improvements in the social and emotional attunement between the adult and child (connection), balanced engagement between the adult and child (reciprocity), and inclusion of all children in high-quality interactions (participation) (Akiva et al., 2017). In a meta-analysis of randomized controlled trials and quasi-experimental studies to improve student-teacher relationships, the more effective programmes tended to use more direct, proactive teacher strategies, such as providing praise, expressing care, and using reflective and supportive listening. However, these programmes have rarely been rigorously assessed among adolescents (Kincade et al., 2020). Future research to strengthen student-teacher relationships should specifically target adolescents, given the continued importance of adult relationships during this period of development (Chhuon & Wallace, 2014; García-Moya, 2020) and the concerning decline in adolescent mental health (Farrell et al., 2023; Racine et al., 2021; Twenge et al., 2021; Wolf & Schmitz, 2023).

Given our study was the first to examine perspective-taking and school belonging beyond that identified in a single country study, further examination of this association is recommended. There is an extensive research base supporting the use of social and emotional learning (SEL) interventions in schools to improve students’ SEL skills, including perspective-taking, among others. However, these programmes appear to be more often implemented and assessed among elementary-aged rather than high school-aged students and are often multicomponent, making it difficult to discern the effect of perspective-taking alone (Durlak et al., 2011; Taylor et al., 2017). We suggest that future research and programmes aiming to study and strengthen perspective-taking (and empathy) among adolescents should consider the dynamic development of self- and other-oriented goals and thinking that occurs during this period (Crone & Fuligni, 2020). Adolescents’ thoughts on self and others and their ego development are influenced by broader social norms related to the distinctions between self and other (Bauer, 2008). We hypothesize that adolescents with more of a quiet ego, characterized in part by greater awareness and integration of others into one’s self-identity and living more interdependently versus individualistically (Bauer & Wayment, 2008), could experience greater school belonging. Future research to better understand this hypothesized association could complement studies of perspective-taking and school belonging in adolescents.

Conclusions

We have provided evidence, from data collected across many countries, that adolescents with higher levels of teacher connection and perspective-taking have greater school belonging. Future research and efforts to deepen our understanding of these associations and to strengthen teacher connection and perspective-taking in adolescents could be a potential antidote to counter the increasing sense of disconnection among adolescents worldwide. When an adolescent experiences a relationship that is developmental, there is a shared, sustained emotional attachment and mutual investment from the adult and adolescent; over time, there is greater complexity around what is done together and a shifting of power towards the adolescent (Li & Julian, 2012). As adolescents grow, they are also learning how to understand, balance, and respond to the needs and perspectives of others in ways that shape their contributions to and connections with social groups (Dunbar, 2018). Together these relationships and developing capacities can contribute to adolescents feeling increasingly seen, accepted, and valued within their school community.

Acknowledgments

The data used in this publication come from PISA 2018 (https://www.oecd.org/pisa/data/2018database). The views expressed here are those of the authors. They are not necessarily those of PISA.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data from the Programme for International Student Assessment is publicly available for download from the Organisation for Economic Co-operation and Development’s website: https://www.oecd.org/pisa/data/2018database/.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/02673843.2023.2254136

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.

Notes on contributors

Tracy Dearth-Wesley

Tracy Dearth-Wesley is a Senior Research Associate in the Columbia-Bassett Program.

Allison N Herman

Allison N. Herman is a Research Associate in the Columbia-Bassett Program.

Robert C. Whitaker

Robert C. Whitaker is a Professor of Clinical Pediatrics at Columbia University’s Vagelos College of Physicians and Surgeons and the Director of Research and Research Education in the Columbia-Bassett Program, a medical education track within the College of Physicians and Surgeons. Research in the program examines how relationships in healthcare, education, and human services can promote well-being by being more responsive to people’s life experiences.