Attitudes and intentions of physical education teachers toward the inclusion of refugee students: The role of moral competence and contextual factors

DOI:

The migrant and refugee crisis constitutes one of the most significant social issues of the contemporary era (Almustafa, 2022). Increasing refugee flows have led to a substantial rise in the number of students with a refugee background in host-country schools, creating new challenges for educational practice. In this context, the adoption of teaching practices that promote inclusion and ensure the psychosocial safety of all students is considered essential. However, successful inclusion largely depends on teachers’ attitudes, their willingness to adapt their practices to the needs of all students, and their perceived ability to meet the demands of this new educational reality.

In this regard, Physical Education (PE) may play a crucial role due to its experiential, cooperative, and often non-verbal nature, as well as the opportunities it provides for meaningful interaction among students through group activities and games. International studies over the last fifteen years indicate that PE can function as a context for enhancing social interaction, psychosocial empowerment, and refugee students’ sense of belonging (Bartsch et al., 2021; Cseplö et al., 2022; Dundar, 2019; Genç, 2024; Grimminger, 2012; Lleixà & Nieva, 2020; Papageorgiou et al., 2021). Specifically, Dundar (2019), in a study conducted in Turkish public schools, reported that the systematic participation of Syrian refugee students in PE was associated with improved levels of social interaction and adjustment, as well as a reduction in the negative effects of traumatic experiences. Similarly, Cseplö et al. (2022), adopting a salutogenic approach and focusing on Sense of Coherence (SOC), highlighted that students perceive PE as a comprehensible, manageable, and meaningful environment that promotes well-being and supportive social relationships. These findings support the view that PE can serve as a protective and supportive factor within the school context.

Nevertheless, the effectiveness of inclusion depends not only on the characteristics of the subject itself but also on teachers’ attitudes, beliefs, and instructional choices. In this direction, Grimminger (2012) found that educational beliefs positively influence the intention to implement intercultural practices; however, intention does not always translate into actual teaching behavior, highlighting the well-known “intention–behavior gap.” Similarly, Bartsch et al. (2021) reported generally positive perceptions among PE teachers regarding the inclusion of students from refugee backgrounds, with variations related to gender and educational level. Within the Greek context, Papageorgiou et al. (2021) noted that although PE is recognized as a means of social inclusion, the lack of training and institutional support limits the implementation of inclusive practices. Likewise, Lleixà and Nieva (2020) emphasized the influence of cultural stereotypes and institutional constraints on teaching decisions, particularly regarding the participation of girls with a migrant background. Finally, the study by Genç (2024) is of particular interest, as it showed that empathy is positively associated with and significantly predicts PE teachers’ attitudes toward refugee students, suggesting that individual and psychosocial dimensions of their professional identity shape inclusive orientations. These findings can be interpreted through the lens of the Theory of Planned

Behavior (TPB; Ajzen, 1991), according to which intention is the most immediate predictor of behavior and is shaped by attitudes, subjective norms, and perceived behavioral control (PBC). In the context of including refugee students in PE, positive attitudes represent a necessary but not sufficient condition for adopting inclusive practices, as factors such as inadequate training and institutional constraints may reduce teachers’ PBC (Lleixà & Nieva, 2020; Papageorgiou et al., 2021).

Overall, the existing literature highlights both the supportive role of PE in the inclusion of refugee students and the importance of teachers’ attitudes, intentions, and perceived teaching competence. However, most studies (particularly within the Greek educational context) focus separately on attitudes or specific psychosocial factors, without adopting an integrated approach that combines demographic characteristics with the four components of the TPB (i.e., attitudes, intention, subjective norms, and PBC) regarding the inclusion of refugee children in school within a unified theoretical framework. At the same time, empirical investigation of the relationship between PE teachers’ moral competence and their attitudes, subjective norms, PBC, and intention to include refugee children in their lessons remains limited, especially in the Greek context. Therefore, the present study aimed to examine PE teachers’ attitudes, subjective norms, PBC, and intention regarding the inclusion of students with a refugee background, as well as the relationship between these variables and teachers’ moral competence index and years of teaching experience. Additional objectives were to explore whether gender constitutes a significant factor for the four TPB components and whether a set of participants’ contextual and experiential variables can predict their intention to include refugee children in their classes.

Methods

Participants

The sample of the present study consisted of 225 Physical Education (PE) teachers (119 males and 106 females) employed in public primary schools in Greece. Regarding teaching experience, 10 participants reported 1–3 years of service, 25 reported 4–9 years, 47 reported 10–15 years, 41 reported 15–20 years, 72 reported 21–25 years, 29 reported 26–30 years, and one participant reported 31 years of service. Furthermore, 60 PE teachers (26.7%)

reported that refugee students were currently enrolled in their school. In addition, 48 participants (21.3%) had received training in intercultural education, 52 (23.1%) had been trained in integration techniques, and 42 (18.7%) reported teaching experience in intercultural/minority schools (M = 0.60 ± 2.24 years).

Measures

Data were collected using a closed-ended self-report questionnaire consisting of three sections:

• a)    Contextual and experiential variables questionnaire . This section included items on gender, age, years of teaching experience, the presence of refugee students in the participants’ schools, previous teaching experience with refugee students or in intercultural/minority schools, and prior training related to the inclusion of children from minority and/or socially vulnerable groups in the school setting.

• b)    Theory of Planned Behavior questionnaire . A questionnaire based on the TPB by Ajzen and Madden (1986) was used. It included 6 items assessing attitudes (introductory statement: “For me, including refugee students in my PE lessons in the coming years is…” , followed by six bipolar adjective pairs such as “bad–good” and “undesirable–desirable” ), 3 items assessing subjective norms (e.g., “My school administration is supportive of the inclusion of refugee students” ), 3 items assessing PBC (e.g., “If I wanted to, I could include refugee students in my PE lessons” ), and 3 items assessing intention (e.g., “I intend to include refugee students in my PE lessons in the coming years” ). Responses were provided on a 7-point Likert scale. For attitudes, 1 indicated the most negative perception and 7 the most positive perception. For subjective norms, intention, and PBC, 1 corresponded to strongly disagree and 7 to strongly agree .

• c)    Moral competence : To assess participants’ moral competence, the modified Greek version of the Moral Competence Test (MCT; Mouratidou, Chatzopoulos, Goutza, & Karamavrou, 2003) was administered. The MCT is based on Kohlberg’s structural-developmental theory of moral development (Lind, 1978, 2003). It was selected because it is the only available instrument that evaluates morality as a whole rather than examining isolated components. The test includes two moral dilemmas. The first concerns workers who illegally enter a company’s administrative offices in order to obtain evidence for an allegation, while the second involves a doctor who assists a terminally ill patient in ending her life upon her request (assisted suicide). Participants are asked to rate 12 arguments for each dilemma (six in favor and six against) using a 9-point Likert-type scale ranging from -4 (“totally disagree”) to +4 (“totally agree”). Each argument corresponds to one of Kohlberg’s six stages of moral development. The main outcome measure of the MCT is the C-index, which reflects an individual’s ability to evaluate arguments based on their moral quality rather than on personal preferences or other psychological influences (Lind, 2000, 2003). The validity and reliability of the MCT require the fulfillment of four key criteria: preference hierarchy, quasi-simplex structure, cognitive-affective parallelism, and correlation with educational level. According to Mouratidou et al. (2003), all these criteria were confirmed for the Greek version of the test.

Procedure

Data were collected electronically via Google Forms using the snowball sampling method. On the first page of the questionnaire, participants were informed about the purpose of the study and the confidentiality of their responses. Participation was voluntary, and no personally identifiable information was collected. Written informed consent was obtained from all participating teachers before accessing the survey instruments.

According to the institutional regulations of Aristotle University of Thessaloniki (AUTH), prior approval by the University Research Ethics and Integrity Committee (EHDE) was not required for this non-funded questionnaire-based study involving competent adult participants. The study was conducted in accordance with the ethical principles of the Declaration of Helsinki and the applicable provisions of the General Data Protection Regulation (GDPR).

Data analysis

Gender differences in attitudes, subjective norms, PBC, and intention were examined using independent-samples t-tests. Associations among moral competence (C-index), TPB constructs, and years of teaching experience were assessed via Pearson’s r. Stepwise regression analysis was employed to determine whether contextual and experiential variables predicted PE teachers’ intention to include refugee children. In all analyses a significance level of p < .05 was used. Analyses were performed using SPSS 20.0 (SPSS Inc., Chicago, USA)

Results

The results of the descriptive statistics indicated that, across the total sample, (relatively) high mean scores were observed for attitudes (M=5.71±1), subjective norms (M=5.25±1.25), and intention (M=5.67±1.16) and moderately high mean score was found for PBC (M=4.88±1.57), regarding the inclusion of refugee children in PE classes. More specifically, mean scores by gender ranged from 5.61 (men) to 5.81 (women) for attitudes, from 5.62 (men) to 5.73 (women) for intention, from 4.79 (women) to 5.66 (men) for subjective norms, and from 4.77 (women) to 4.98 (men) for PBC.

In addition, as demonstrated by the results of t-tests, gender was not a significant factor for attitudes [ t (213) = -1.462, p > .05], for PBC [ t (223) = .987, p > .05], and for intention [ t (223) = -.769, p > .05]; however, this was not the case for subjective norms, as male participants exhibited a significantly higher mean score (M = 5.66±1.08) compared to female participants (M = 4.79±1.26) [ t (223) = 5.575, p < .01].

Furthermore, Pearson correlation analysis, conducted to examine the relationships among the four TPB variables, participants’ years of teaching experience, and their moral competence, revealed that only two parameters of TPB are significantly associated with PE teachers’ moral competence. Specifically, the following findings emerged: (a) PBC and subjective norms were moderately and negatively associated with participants’ C-index (r = -.242, p < .05 and r = -.248, p < .05 respectively); and (b) all four variables of TPB were strongly and positively associated with each other (see table 1 below).

Table 1. Pearson correlation analysis.

	1.	2.       3.	4.	5.     6.
Attitudes	1
Intention	.771**	1
Perceived behavioral control	.369**	.502**    1
Subjective norms	.246**	.405**  .460**	1
Years of teaching experience	-.004	-.069  -.010	-.050	1
Moral competence	-.073	-.144  -.242*	-.248*	-.048  1

Note : * p< .05. ** p< .01.

Finally, a stepwise linear regression analysis was conducted to examine whether five contextual and professional background variables predicted teachers’ intention. The initial set of predictors included: (a) presence of refugee students in participants’ schools, (b) prior training in intercultural education, (c) previous teaching experience with refugee students, (d) previous teaching experience in intercultural or minority schools, and (e) prior training related to the inclusion of children from minority and/or socially vulnerable groups. Using the stepwise method (entry criterion: p ≤ .05; removal criterion: p ≥ .10), only prior training in intercultural education was retained in the final model. The remaining variables did not contribute significantly to the prediction of intention and were excluded from the model. The final regression model was statistically significant, F(1,223) = 8.12, p =.005, explaining 3.5% of the variance in intention (R² = .035, Adjusted R² =.031). The multiple correlation coefficient was modest (R = .187), indicating a weak association between the predictor and the dependent variable. Prior training in intercultural education emerged as a significant negative predictor of intention (B = -0.531, SE = 0.186, β=-0.187, t=-2.85, p=.005). The effect size was small (f² = 0.036), suggesting that although the association was statistically significant, its practical magnitude was limited. The final regression equation was: Intention = 6.621 - 0.531 × (Intercultural Training).

Discussion

The present study aimed to examine Greek Physical Education (PE) teachers’ attitudes, subjective norms, PBC, and intention regarding the inclusion of students with a refugee background within the framework of the TPB. Additionally, it explored the relationship of these variables with moral competence and selected contextual factors

Overall, participants reported relatively high levels of positive attitudes, subjective norms, and intentions, and moderately high PBC concerning the inclusion of refugee students in PE classes. These findings are consistent with previous research indicating generally favorable orientations among PE teachers toward inclusion of students with refugee or migrant backgrounds (e.g., Bartsch et al., 2021; Papageorgiou et al., 2021). The strong positive intercorrelations observed among the four TPB variables further support the internal coherence of the model in this context. This finding aligns with the theoretical assumption that attitudes, subjective norms, and PBC are positively associated with intention. Notably, attitudes were strongly associated with intention, reinforcing the central role of evaluative beliefs in shaping teachers’ readiness to adopt inclusive practices. However, although teachers expressed positive orientations, PBC received comparatively lower scores, suggesting that confidence in their capability to implement inclusion may be more fragile than their attitudinal stance. This finding is consistent with previous evidence indicating that institutional constraints, limited training, and lack of structural support may weaken teachers’ perceived competence, even when their attitudes are favorable.

Gender was not a significant factor for attitudes, PBC, or intention, but male teachers reported significantly higher subjective norms compared to female teachers. This finding may indicate that male participants perceive greater institutional or social support for inclusion, or interpret professional expectations differently. Previous literature has reported mixed findings regarding gender differences in inclusive orientations (e.g., Bartsch et al., 2021), suggesting that contextual and cultural factors may shape how normative pressures are perceived. Given the absence of gender effects on intention, however, the practical implications of this difference appear limited.

One of the most intriguing findings concerns the relationship between moral competence (C-index) and TPB components. Contrary to expectations grounded in moral development theory (e.g., Lawrence Kohlberg), moral competence was not positively associated with attitudes or intention. Instead, it was moderately and negatively correlated with subjective norms and PBC. This pattern may suggest that teachers with higher moral competence rely less on external normative pressures when forming their professional judgments. In other words, individuals with greater ability to evaluate arguments based on moral quality rather than personal preference may adopt a more autonomous stance. Consequently, they become less susceptible to perceived social expectations (subjective norms). The negative association with PBC may reflect a more critical and reflective appraisal of institutional realities. Teachers with higher moral competence might be more aware of structural barriers and therefore less confident in their capacity to implement inclusive practices under current conditions. This finding aligns with prior research suggesting that higher levels of moral reasoning do not automatically translate into consistent moral action, particularly when contextual constraints are present. On the other hand, years of teaching experience were unrelated to TPB constructs and moral competence. This absence of association indicates that inclusive orientations and moral reasoning capacities are not necessarily strengthened through professional longevity alone. Instead, targeted educational experiences and professional development may play a more decisive role than accumulated years of service.

The regression analysis revealed that among the examined contextual and experiential variables, only prior training in intercultural education significantly predicted intention, and notably in a negative direction. Although the effect size was small, this counterintuitive finding warrants careful interpretation. One possible explanation is that teachers who have received intercultural training may develop a more realistic or critical understanding of the complexities and challenges of inclusion, which in turn moderates their expressed intention. Rather than reflecting reduced commitment, lower intention scores may represent a more cautious or context-sensitive stance. Alternatively, the content or quality of existing training programs may not sufficiently enhance teachers’ perceived competence, thus failing to strengthen behavioral intentions. The limited explanatory power of the regression model (3.5% of variance) further indicates that intention to include refugee students is influenced by additional psychological, organizational, and structural variables not captured in the present study.

From a theoretical perspective, the findings of the present study support the applicability of the TPB framework in understanding inclusive intentions within PE settings. At the same time, the negative associations between moral competence and certain TPB components suggest that moral reasoning operates in a more complex manner than a simple positive predictor of inclusive orientation. Practically, the results underline the need for professional development programs that go beyond raising awareness and instead strengthen teachers’ PBC through concrete pedagogical strategies, experiential learning, and institutional support. Enhancing structural conditions within schools may be essential to bridge the intention–behavior gap identified in previous literature.

Several limitations should be acknowledged. First, the use of a cross-sectional design precludes causal inferences. Second, data were collected through self-report measures, which may be subject to social desirability bias, particularly in morally salient topics such as inclusion. Third, the snowball sampling method limits generalizability beyond the present sample of Greek PE teachers. Finally, the study assessed intention rather than actual teaching behavior; future research should examine whether reported intentions translate into observable inclusive practices. Longitudinal and mixed-method designs could provide deeper insight into how moral competence, institutional conditions, and professional development interact over time. Moreover, qualitative exploration of teachers’ interpretations of intercultural training may clarify the unexpected negative association with intention.

Conclusion

The present findings indicate that Greek PE teachers hold generally favorable attitudes, subjective norms, and intentions toward the inclusion of refugee students, although their perceived behavioral control remains comparatively lower, pointing to a gap between positive orientations and confidence in practical implementation. The strong intercorrelations among the four TPB components support the applicability of the theory within this educational context, while the unexpected negative associations between moral competence and subjective norms/PBC suggest that moral reasoning shapes inclusive orientations in a more complex manner than initially assumed. Contextual and experiential variables showed limited predictive value for intention, with only prior intercultural training emerging as a significant, albeit weak, negative predictor, underscoring the need to reconsider the content and quality of existing training programs. Taken together, these findings suggest that fostering PE teachers' inclusive practices toward refugee students requires not only positive attitudes but also targeted institutional support and professional development that strengthen perceived behavioral control and help translate favorable intentions into sustained classroom practice.