When learning with SLD is a hard challenge: a complex perspective on teacher-student relationships in primary school

PMCID: PMC12546017

PMID:

Abstract

Introduction The growing body of research highlighting the importance of a positive and supportive student-teacher relationship underscores the need to investigate whether pupils’ needs and teachers’ educational stance may shape the potential protective role of the student-teacher relationship. In line with our hypotheses, we investigated whether the presence of a Specific Learning Disorder (SLD) diagnosis, together with pupils’ and teachers’ demographic characteristics, may affect how both teachers and pupils perceive the quality of their relationship. We also examined teachers’ beliefs and attitudes toward SLD diagnoses and their educational stance, expecting that more adultcentric and authoritarian approaches may increase relational challenges when managing the complexity associated with SLD diagnoses. Methods The present work addresses these aims, taking into account pupils’ and teachers’ perspectives on relational quality, teachers’ attitudes and beliefs regarding the rising number of SLD diagnoses in schools, and teachers’ agreement with the Adultcentric paradigm and Bleak Pedagogy. The study involved 66 Italian primary-school teachers and 110 Italian primary-school pupils, who completed validated questionnaires on student-teacher relationship quality, teachers’ attitudes toward SLD diagnoses, and educational stance. Data were analyzed through correlations, paired-sample t -tests, and linear mixed-effects models (LMMs) to examine associations between teachers’ beliefs, pupils’ reports, and perceptions of relational quality. Results Results show that the student-teacher relationship may be partly influenced by the presence of an SLD diagnosis. The perceived levels of “Closeness” and “Dependency” were significantly higher when teachers described relationships with a pupil diagnosed with an SLD, compared to their descriptions of the relationship with an undiagnosed pupil with poor school performance. Discussion Participants’ agreement with Adultcentrism and Bleak Pedagogy was associated with relationship quality and perceptions of SLD diagnosis, highlighting the substantial unsuitability of such educational stances with our evolving culture and society. Implications for intervention and future research are discussed.

Full Text

Over recent years, in Italy, within a general rise in neurodevelopmental disorders (Lucangeli and Perini, 2020), the number of students diagnosed with Specific Learning Disorder (APA, 2022) has notably increased. A report produced by the Statistical Office of the Ministry of Education and Merit (MIM, 2024) shows that the prevalence of Specific Learning Disorder (SLD) diagnoses rose from 0.9% in school year (SY) 2010/2011 to 6% in SY 2022/2023. Looking at the different types of SLD, the past decade has seen a steady increase across all categories: the number of pupils certified with dyslexia has more than doubled, while cases of dysgraphia, dysorthography, and dyscalculia have more than tripled. Considering only primary school, the percentage of pupils diagnosed with SLD increased from 0.8% in SY 2010/2011 to 3.2% in SY 2022/2023. This increase should also be considered in light of the fact that, starting in 2010 with the enactment of Law 170/2010, awareness of SLDs in Italy significantly grew, contributing to a broader recognition and reporting of such cases.
As a consequence of the increasing number of SLD diagnoses, the current expectation for teachers to manage a wide variety of special educational needs (e.g., Specific Learning Disorder – SLD) and to promote an inclusive classroom atmosphere (Cornoldi et al., 1998; Saloviita and Consegnati, 2019) requires them to continuously adjust and update their educational beliefs and methods to face the challenges of the current school context and the needs of individual pupils. The following section outlines the theoretical background and research context that motivate the present study, highlighting current educational challenges linked to the growing prevalence of SLD diagnoses, the potential effects of labeling, the importance of teacher-student relationships, and the role of educational stance, concluding with the aims and hypotheses of the study.
Given the challenges arising from the increasing complexity and broader acknowledgment of pupils’ varied needs, it becomes crucial to understand how the quality of the relationship between teachers and pupils can serve as a protective factor in promoting positive learning and developmental outcomes. In fact, the establishment of a supportive relationship between teacher and pupil is an important element of protection that is advantageous to the development of the child (Pianta, 2001a). The systematic review conducted by García-Rodríguez et al. (2023) highlighted that teachers’ attachment style and their availability in responding to student needs are associated with students’ school adjustment. The Student-Teacher Relationship Scale (Pianta, 1994, 2001b; Pianta et al., 1995; Pianta and Nimetz, 1991; Pianta and Steinberg, 1992), identified as “the most frequently used measurement instrument” (García-Rodríguez et al., 2023, p. 24), highlights various beneficial aspects of a relationship characterized by high levels of closeness and low levels of conflict and dependency within the theoretical framework of the attachment theory (Sabol and Pianta, 2012). Research consistently highlights the association of the quality of the teacher-student relationship with multiple components of pupils’ school experience, such as students’ engagement (Thornberg et al., 2022), self-efficacy in both the learning domain and academic success (Jederlund and von Rosen, 2023), concept-of-self, emotional regulation (García-Rodríguez et al., 2023), and cognitive abilities (Sankalaite et al., 2023; Semeraro et al., 2014). It is worth noting, however, that individual differences may also shape these relationships. From both teachers’ and pupils’ perspectives, some results in the literature on the quality of the pupil-teacher relationship suggest the presence of gender and age differences: females seem to build relationships characterized by greater closeness and less conflict (Koomen and Jellesma, 2015; Zee and Koomen, 2017; Zee et al., 2020). Therefore it is particularly important to highlight that, in the Italian context, these results could also be influenced by the prevalence of female teachers, especially in primary schools (La Grutta et al., 2023; Longobardi et al., 2016). Considering age differences, some studies showed a decrease in school attachment in the last years of primary school (Guidetti et al., 2017; Schneider et al., 2008; Zee et al., 2020). Consequently, it could be hypothesized that the overall quality of teacher-student relationships also declines during these final years, making age a relevant factor to consider among the other factors examined in this study.
Finally, previous studies suggest that higher workload and accumulated fatigue may hinder the ability to sustain consistent emotional attunement both with pupils in general and with those requiring more complex support (Bodenheimer and Shuster, 2020; Jennings and Greenberg, 2009; Maas et al., 2021). Therefore, among the factors that shape teachers’ perceptions, professional characteristics such as teaching experience and weekly working hours can be considered relevant, as they may influence the emotional resources and responsiveness available to build high-quality relationships with pupils.
Beyond teachers’ professional characteristics, the quality of teacher-student relationships may vary significantly when pupils face learning challenges or carry diagnostic labels, such as those related to SLD, which can influence teachers’ perceptions and attitudes. Building on the literature showing the protective role of the teacher-student relationship, it is important to examine how this role applies when students face significant learning challenges, such as SLD. Pennings et al. (2016) emphasize the association between a high-quality teacher-student relationship and teachers’ ability to respond to their students’ needs. More specifically, Zee et al. (2020) studied the influence of the teacher-student relationship in a sample of primary school pupils who were diagnosed with dyslexia, an autism spectrum disorder (ASD), or attention deficit hyperactivity disorder (ADHD). They found that the impact of the relationship quality seemed to be higher in pupils with behavior-related disorders rather than in pupils with learning disorders (Zee et al., 2020). Nevertheless, the multi-informant approach also enabled the detection of some differences in the quality of the relationship as perceived by teachers, classmates, and pupils with a dyslexia diagnosis (Zee et al., 2020). Although all informants agreed on the lower level of conflict in the relationship, pupils with a dyslexia diagnosis, unlike both teachers and classmates, perceived less closeness with their teachers compared to pupils without this type of diagnosis (Zee et al., 2020). These differences in pupils’ and teachers’ perspectives suggest the relevance of collecting both teachers’ and pupils’ points of view, as done in the present study. Moreover, the perception of lower emotional availability of teachers coming from children with learning disorders in comparison with pupils without a diagnosis, detected also in previous studies (Al-Yagon and Mikulincer, 2004; Murray and Greenberg, 2000), provides a first reason to explore the quality of the teacher-student relationship in a sample of pupils an SLD diagnosis. The second reason stems from the investigation of teachers’ perspectives. In their relationship with pupils having a Specific Learning Disorder, teachers reported a lower perception of emotional closeness (Al-Yagon and Mikulincer, 2004) and a higher dependency in the relationship (Pasta et al., 2013) in comparison with the quality of the relationship with pupils without this diagnosis.
The understanding of teachers’ attitudes toward pupils with an SLD diagnosis has been explored in the literature using both implicit and explicit measures. The use of implicit measures detected the presence of a “labeling bias” of teachers toward pupils with dyslexia (Hornstra et al., 2010), suggesting that the diagnosis itself may influence teachers’ perceptions beyond the observable learning difficulties. Therefore, distinguishing between pupils with an SLD diagnosis and low achievers without this diagnosis is crucial to disentangle the effects of SLD labeling from those of performance.
At a more explicit level of investigation, studies explored educators’ perceptions (Tosun et al., 2021) and misconceptions concerning the knowledge of dyslexia (Soriano-Ferrer and Echegaray-Bengoa, 2014; Soriano-Ferrer et al., 2016; Wadlington and Wadlington, 2005). The role of primary school teachers’ knowledge about SLD was found to be a predictor for the use of compensatory tools, and teachers’ beliefs and emotions about SLD can predict the use of dispensatory accomodations (Re et al., 2021), showing the connection between teachers’ representational framework and their educational practice. Moreover, Tosun et al. (2021) revealed a weak correlation between teaching experience and the negative perception of dyslexia. The explorative study conducted by Cornoldi et al. (2018) extended the focus on primary school teachers’ attitudes toward learning disabilities, including beliefs about the roles of legislation, teachers, specialized intervention, compensatory measures, integration in the classroom, the labelling effect and emotional challenges in the experience of their pupils with learning disabilities. Although some beliefs were shared across the three countries involved in the study (Italy, Spain, and the U.S.), significant differences emerged (Cornoldi et al., 2018), suggesting the relevance of investigating the multiple variables that could intervene in influencing teachers’ attitudes and beliefs that guide their educational practices.
In addition to diagnostic labels, another key factor to consider is the broader perspective of the educational stance that shapes teachers’ approaches to relationships with pupils and classroom management. To do so, it seems necessary to develop a deeper understanding of teachers’ ideas concerning an SLD diagnosis within a comprehensive and complex framework of beliefs and attitudes that contribute to defining the teacher’s educational stance (Florio, 2018). Adultcentrism and Bleak Pedagogy are two constructs that enable the exploration of specific beliefs and attitudes influencing teachers’ educational stance, as manifested in their methodological and classroom management choices. Indeed, as recognized by students, to define a “good teacher,” it is necessary to consider the interconnection among personal, relational, methodological, and management factors (Thornberg et al., 2022). Adultcentrism (Furioso, 2000; Goode, 1986) is an implicit paradigm of thought rooted in a sociocultural framework in which adulthood is regarded as the full realization of human development, in a dichotomous comparison with the perceived incompleteness of childhood (Florio et al., 2020a). Although adults intend to act in the best interests of children, the power asymmetry and the depiction of children as incomplete, immature, and incompetent individuals that characterized the adultcentric perspective could lead to inadequate recognition, understanding, and responses to children’s needs (Florio, 2018; Florio et al., 2020a; Mackay, 2003). Florio et al. (2022a) demonstrated that, in a sample of primary school teachers, an adultcentric perspective is a consistent predictor of the values and beliefs of an old-fashioned and authoritarian educational stance, namely the construct of Schwarze Pädagogik (Rutschky, 1977, 2015), which literally translates as Black Pedagogy (Florio et al., 2020b). It is necessary to clarify that this label does not pertain in any way to education involving black students and teachers, nor school programs including black studies (Johnson et al., 2014; Pitre et al., 2008): the label has been recently adapted, and the term currently used by this group of authors is Bleak Pedagogy (Florio et al., 2024a, 2024b; Florio et al., n.d.). Bleak Pedagogy comprises values, beliefs, and methods focused primarily on shaping the pupils according to ideal standards of educators and society, thereby undermining the pupils’ will (Florio et al., 2020b). The Bleak Pedagogy methods also include forms of both physical and psychological violence for disciplinary purposes (Florio et al., 2020b; Perticari, 2016). The sociocultural context also influences educators’ ability to recognize this detrimental educational stance, which – especially in its psychological forms – tends to remain implicit and, for this very reason, is often enacted in practice (Florio et al., 2022a). The first evidence (Florio, 2018) suggested that the agreement with Adultcentrism and especially Bleak Pedagogy could predict specific features of teachers’ representations concerning an SLD diagnosis, such as a critical attitude toward the role of society (e.g., medicalization, how children are raised nowadays), specialists, parents, and classmates in determining the presence and the school experience of pupils with an SLD diagnosis. These results underline the importance of investigating teachers’ attitudes and beliefs regarding their complex role and its impact on the quality of the teacher-student relationship, especially when dealing with a pupil diagnosed with an SLD or perceived as demonstrating low academic achievement.
Despite the growing body of research on teacher-student relationships and SLD, few studies have directly compared teachers’ perceptions of pupils with a formal SLD diagnosis and low achievers without a diagnosis. This comparison is particularly relevant for disentangling the potential effects of labeling from the underlying learning difficulties. Building on the literature on labeling bias (Hornstra et al., 2010) and regarding teachers’ attitudes toward SLD (Cornoldi et al., 2018; Tosun et al., 2021), our study aims to examine how teachers perceive and describe the quality of their relationships with these two groups of pupils. We also explore whether factors such as teaching experience and weekly working hours are associated with these perceptions, as suggested by previous findings on teacher fatigue and workload (Zee et al., 2020). Specifically, we ask whether the quality of teacher-student relationship differs between pupils with and without an SLD diagnosis, and how teachers’ professional characteristics—such as educational stance—might influence their views. These considerations form the basis of our study objectives and hypotheses, outlined below. Building on these premises, the present study integrates variables such as labeling effects, teacher workload, experience, and pupils’ age, to offer a multi-layered perspective that extends existing research on teacher-student relationships and SLD.
The first objective is to investigate perceptions of the quality of the teacher-student relationship with pupils who have an SLD diagnosis (Specific Learning Disorder diagnosis—SLD pupils). The first hypothesis (A) is that the presence of the diagnostic label of a Specific Learning Disorder could influence teachers’ and pupils’ representations of their relationship, even when compared with pupils who, in the eyes of teachers, have some difficulties in dealing with learning challenges (Without a Diagnosis – WD pupils) (Hornstra et al., 2010; Pasta et al., 2013; Re et al., 2021; Zee et al., 2020). The second hypothesis (B) is that pupils’ age and gender, as well as teachers’ experience and working hours, may influence the quality of the teacher-student relationship, as highlighted by previous literature (Guidetti et al., 2017; La Grutta et al., 2023; Longobardi et al., 2016; Tosun et al., 2021).
The second objective is to enrich the perspective on the quality of student-teacher relationship through the investigation of teachers’ beliefs and attitudes toward SLD diagnosis and the paradigm behind their educational approach, considering also the possible influence of teachers’ demographics. The first hypothesis (C) is that adult-centered and authoritarian educational practices are associated with difficulties in the educational relationship and greater teacher fatigue in dealing with the complexity that SLD diagnoses bring to the table. The second hypothesis (D) is that teachers’ beliefs and attitudes toward learning disabilities could influence both the pupils’ school experience (Re et al., 2021; Zee et al., 2020) and how teachers perceive the relational quality with pupils (Pasta et al., 2013).
Teachers and children belonged to 26 different primary schools located in the province of Bergamo, in Italy. All participants were treated in accordance with the Declaration of Helsinki (World Medical Association, 2013), as well as with the ethical guidelines for research provided by the American Psychological Association (APA, 2017) and the Italian Psychological Association (AIP, 2022). Participants were also provided with the researcher’s contact information to receive further clarifications if needed, and informed consent was obtained to proceed with the battery of questionnaires. Participants received clear information about the objectives and phases of the research and about their rights as participants, including the guarantee of anonymity and the possibility of dropping out of the study at any moment. The following subsections describe the characteristics of the two participating groups—teachers and pupils—as well as the procedures adopted for each group.
According to the aim of the study, measures investigating the quality of the teacher-student relationship were used. Both teachers’ perspectives (STRS – Student-Teacher Relationship Scale; Pianta, 1994, 2001b; Pianta et al., 1995; Pianta and Nimetz, 1991; Pianta and Steinberg, 1992) and pupils’ perspectives (STRQ – Student-Teacher Relationship Questionnaire; Murray and Greenberg, 2000; Tonci et al., 2012) were taken into account, and a measure to collect teachers’ attitudes and beliefs regarding SLD diagnosis (RADSA – Representations and Attitudes towards SLD diagnoses questionnaire; Florio et al., 2022b) was also included. In order to fulfill the second specific objective, the Bleak Pedagogy Scale (Florio et al., 2020b) and the Adultcentrism Scale for Teachers (Florio et al., 2022a) were used, allowing further exploration of the complexity of the educational relationship, including the paradigm of thought, values, and methods underlying teachers’ educational approaches.
Teachers were asked to indicate one pupil with an SLD diagnosis (SLD pupils), and one pupil they considered to have poor school performance but without any diagnosis (WD pupils). In particular, to investigate the quality of the teacher-pupil relationship they were asked to complete the Italian version of the Student-Teacher Relationship Scale (Pianta, 1994, 2001b; Pianta et al., 1995; Pianta and Nimetz, 1991; Pianta and Steinberg, 1992) twice. The STRS is composed of 22 items in its Italian version validated by Fraire et al. (2013), reporting good internal consistency for the subscales (“Closeness”: α = 0.85; “Conflict”: α = 0.92; “Dependency”: α = 0.69). The initial instructions of the STRS ask teachers to think of a particular pupil while answering the questions. Subsequently, items describing relationship characteristics are presented, and participants indicate how much each statement reflects their relationship with the pupil in question. Responses are set on a 5-point Likert scale of applicability (1 = Definitely does not apply; 2 = Does not really apply; 3 = Neutral, not sure; 4 = Somewhat applies; 5 = Definitely applies). The Italian version of STRS maintains the original three-dimensional structure, composed of “Closeness” (8 items), “Conflict” (10 items) and “Dependency” (4 items). The “Closeness” subscale provides information about the degree of warmth, affection and open communication experienced by the teacher with a particular student; the “Conflict” subscale provides an evaluation of the level of negativity and conflict present in the relationship; lastly, the “Dependency” subscale measures the extent to which the teacher perceives the student as overly dependent (Pianta, 2001b). A higher total score on the STRS indicates an overall positive relationship, which tends to be characterized by lower levels of conflict and dependency, and by higher levels of closeness (Pianta, 2001b, p. 12).
In addressing the complexities associated with SLD diagnoses, the Representations and Attitudes towards SLD diagnoses questionnaire was chosen (RADSA; Florio et al., 2022b) as the tool that best allows teachers to examine their own representations. This makes evident possible implicit obstacles that may affect the quality of the relationship, especially with SLD pupils. The RADSA questionnaire, validated on a sample of Italian primary school teachers (Florio et al., 2022b), is composed of 62 items with a 4-point Likert scale measuring the level of agreement (1 = Fully disagree; 2 = Slightly agree; 3 = Agree; 4 = Fully agree). RADSA is organized in four sections: “Increase of SLD diagnoses,” “Peer group,” “Parents’ attitudes towards SLD diagnosis,” “The moment when the teacher refers parents to an SLD specialist.” Each section is composed of three factors. Given the multidimensional structure of the instrument, Table 1 illustrates its sections and factors, along with example items. For a comprehensive overview of the RADSA and its characteristics, (see Florio et al., 2022b). As no internal consistency estimates were reported in the original validation of the RADSA, we computed Cronbach’s α on the full set of items. The resulting internal consistency was good (α = 0.84).
Teachers’ agreement with attitudes and methods typical of an old-fashioned and authoritarian educational stance was evaluated with the first section of the Bleak Pedagogy Scale (originally published as the Black Pedagogy Scale; Florio et al., 2020b). The first section is named “Bleak Pedagogy Observation” section (BPO), which consists of 24 items, with a Cronbach’s α of 0.87. Teachers are requested to express their degree of agreement with each item on a 4-point Likert scale. This section of the instrument is composed of three factors: “Values of Bleak Pedagogy” (α = 0.87; high values: agreement with the main educational values and objectives typical of the Bleak Pedagogy’s perspective), “Education of children over time” (α = 0.75; high values: nostalgic attitude towards old-fashioned educational practices as more effective and useful), and “Methods of Bleak Pedagogy” (α = 0.74; high values: agreement with Bleak Pedagogy disciplinary and educational methods, such as pedagogical beating, humiliation, coldness towards the child, etc.). The original instrument also includes a first item for temporal placement and a further a section assessing the perceived prevalence of Bleak Pedagogy methods in the past and today, which were not used in the present study. The Bleak Pedagogy Scale has been validated on a sample of Italian teachers (Florio et al., 2022a), following a first study (Florio et al., 2020b) conducted with a sample of Italian university students and parents.
The second measurement tool used is the Adultcentrism Scale for Teachers (ADT; Florio et al., 2022a), composed of 9 items (Cronbach’s α = 0.74) with a 4-point Likert scale based on agreement. High scores indicate that teachers tend to think of their pupils as disempowered and without responsibility, as an empty receptacle in need of being filled by adults with social and cultural values. The validation of this version of the instrument on the Italian teachers’ sample was conducted by Florio et al. (2022a) after the first validation study for the original scale (Florio et al., 2020a) with a sample of Italian parents and university students.
Students play an active role in constructing the teacher-student relationship and determining its quality and dynamics (Pianta, 2001a). Therefore, to avoid a partial representation of this relationship, the Student-Teacher Relationship Questionnaire (Murray and Greenberg, 2000) in its Italian version (Tonci et al., 2012) was presented to children, to collect data on how they experience their relationship with their teacher. The instrument, validated by Tonci et al. (2012) on a sample of 4th and 5th grade Italian pupils, consists of 22 items with possible answers based on a 4-point Likert scale (1 = Not at all true; 2 = Not very true; 3 = Quite true; 4 = Very true), loading on the following three subscales: “Closeness with teachers” (α = 0.86; high scores: positive evaluation of the relationship in terms of emotions, closeness and support), “Dissatisfaction with teacher” (α = 0.76; high scores: child’s negative opinions and feelings concerning the teacher), and “Bond with school” (α = 0.71; high scores: positive feelings towards the school environment and activities). In our sample, Cronbach’s α exceeded 0.76 for the younger group and 0.74 for the older group. The coefficients, ranging from 0.742 to 0.875, are generally considered acceptable to good (DeVellis, 2016; George and Mallery, 2003; Taber, 2018). For each STRQ subscale, mean corrected item-total correlations (r.drop) were also computed: coefficients ranged from 0.481 to 0.682 across age groups. These values are considered acceptable and ideal values (Nunnally and Bernstein, 1994; Hair et al., 2019), supporting internal consistency alongside Cronbach’s α.
The analyses were performed in R (Version 2025.05.0 + 496). Analyses were designed to address the specific hypotheses outlined in section 1.6 (Aims of the Study). The corresponding statistical procedures are briefly recalled below to guide the reader through the structure of the results. The absence of important violations of normality was assessed, as values of skewness and kurtosis between −1 and +1 are considered acceptable (Muthén and Kaplan, 1985). Thresholds of sk > 2 and ku > 7, values that would indicate a severe violation of normality, were not exceeded (Costello and Osborne, 2005; Kim, 2013; West et al., 1995). Regarding STRQ, internal consistency coefficients (Cronbach’s α) were computed separately for younger and older children, since the original validation of STRQ was conducted on 4th and 5th grade pupils only (Tonci et al., 2012). Values of α > 0.7 were considered acceptable, whereas α > 0.8 indicated good reliability (DeVellis, 2016; George and Mallery, 2003; Taber, 2018). We acknowledge that Cronbach’s α by itself provides only a partial index of scale quality and should not be interpreted as definitive evidence of dimensionality or construct validity; further analyses (e.g., item-total correlations or factor analysis) would be warranted to fully support the internal structure of the subscales (Taber, 2018).
However, we chose not to conduct factor analysis for several reasons: first, STRQ was used in its validated structure without modification. Second, as detailed communalities (h) were not reported in the original Italian validation of STRQ (Tonci et al., 2012), it is difficult to assess whether our sample size meets the conditions for stable factor recovery: according to Hogarty et al. (2005), when h are unknown or potentially low, sample to variable ratios (N/p) below 10:1—such as in our case (N/p = 6.5:1)—may be insufficient to ensure stability. Lastly, factor analysis was beyond the scope of the present study, which did not aim to revalidate STRQ on a younger sample. Instead, item-total correlations were computed using the “sirt” package (Robitzsch, 2024). The mean corrected item-total correlation (r.drop) is reported and interpreted following Nunnally and Bernsteof (1994), who suggest accepting values of r.drop ≥ 0.3, with values above 0.5 considered ideal (Hair et al., 2019).
To fulfill the first purpose of the study and to verify the related hypotheses (A and B), analyses considering teachers’ and pupils’ perceptions of the relationship involving SLD vs. WD students were carried out. These included descriptive statistics for STRS and STRQ subscales, correlations to detect associations between STRS subscales and STRQ scores, and paired t-tests to compare STRS and STRQ subscales scores. For paired-sample t-tests, Cohen’s d was calculated to estimate the magnitude of differences, in accordance with standard guidelines (Cohen, 1988). Correlations with teachers’ demographics and independent-sample t-tests have been conducted to detect significant differences in all variables based on each student’s school grade (i.e., younger: 2nd and 3rd grade, older: 4th and 5th grade) and gender, to verify the second hypothesis related to this first objective.
In line with the second aim of the study (hypotheses B and C), descriptive statistics for the other measures (i.e., RADSA and Adultcentrism/Bleak Pedagogy) were calculated. To verify the related hypotheses, correlations were performed among all the subscales and teachers’ demographics. Regarding Pearson product–moment correlations, Cohen’s guidelines (Cohen, 1988) were used to interpret their strength: values of r falling within the range of 0.50 to 1 are considered “large,” those between 0.30 and 0.49 “medium,” and those between 0.10 and 0.29 “small.” When significant correlations warranted further investigation, to account for the nested structure of the data (i.e., students nested within teachers and schools), Linear Mixed-effects Models (LMMs) were performed using the “lme4” package (Bates et al., 2015) for LMMs and the “lmerTest” package (Kuznetsova et al., 2017) to compute p-values via Satterthwaite’s approximation for degrees of freedom. When it was not possible to include both school and teacher as random intercepts due to convergence issues, only the school-level intercept was retained to account for the nested data structure. For these analyses, data were first restructured from wide to long format to allow for within-teacher comparisons, since each teacher completed STRS twice (once for a student with SLD and once for a student without a diagnosis). LMMs were also based on the relevant hypotheses: fixed effects included diagnostic status (SLD vs. WD), pupil demographics (gender, age group), teacher characteristics (experience, weekly hours), and theoretical variables such as Adultcentrism and Bleak Pedagogy scores. Random intercepts for school were included in all models, and random intercepts for teacher were added when multiple ratings per teacher were available. When teacher-level variance was negligible (singular fit), only school-level random intercepts were retained.
Intraclass correlation coefficients (ICC) from null models were computed to justify the multilevel approach and are presented in the Results section. ICCs were derived from the variance components obtained via the VarCorr{lme4} function (Devine et al., 2024). Values were interpreted as follows: ICC < 0.05 = poor, 0.05–0.75 = small to moderate, 0.75–0.90 = good, > 0.90 = excellent (Liljequist et al., 2019; Trevethan, 2017). However, it should also be noted that ICC values between 0.05 and 0.25 are quite common in social sciences (Snijders and Bosker, 2012). When the ICC was equal to 0, indicating no variance attributable to the grouping factor, the LMM approach was nonetheless retained if the model was statistically significant, to ensure consistency across analyses. To complement the information provided on each LMM, we computed marginal and conditional R2 values with “performance” package (Lüdecke et al., 2021). These quantify, respectively, the proportion of variance explained by the fixed effects alone (R2
m) – providing insight into the effect size attributable to the predictors – and the total variance explained by the full model (R2
c), accounting for both fixed and random effects (Johnson, 2014; Nakagawa and Schielzeth, 2013; Nakagawa et al., 2017).
Table 2 presents the descriptive analyses performed on teachers’ responses to the STRS measure, in its two administered versions. In line with hypotheses A and B, we explored whether the diagnostic label and teacher-related variables shaped teachers’ perceptions of the relationship and their educational stance.
STRS Closeness (SLD pupils) was negatively correlated with weekly teaching hours (r = −0.25, p = 0.04), indicating that the greater the number of work hours, the less closeness was perceived in the relationship with an SLD pupil. The same dimension of Closeness with a WD pupil was positively correlated with years of teaching (r = 0.24, p = 0.05), showing a marginally significant association. With WD pupils, the STRS Dependency dimension was positively correlated with hours of teaching per week (r = 0.25, p = 0.04). These three associations were further examined through separate LMMs, each including teaching experience (years of teaching) and weekly hours spent with pupils as fixed effects and school as a random intercept. Neither the effect of teachers’ work experience on Closeness (WD pupils), nor that of weekly hours spent on Closeness (SLD pupils) yielded significant results. Thus, only the LMMs that showed significant effects are reported in detail. Specifically, for Closeness with WD pupils, the number of weekly hours spent with the pupil was a significant predictor (cf. M1, Table 3), indicating that more contact time was associated with higher perceived Closeness in the relationship with a WD pupil. Finally, Dependency scores related to pupils without a diagnosis were also significantly predicted by the weekly hours spent with the pupil (cf. M2, Table 3), suggesting that teachers perceive undiagnosed pupils as more dependent when they spend more time with them.
Given these results, we explored through LMMs whether teachers’ endorsement of Adultcentrism and Bleak Pedagogy moderated the associations between time spent with the WD pupil and STRS-perceived Closeness or Dependency. This analysis further contributed to the investigation of hypothesis D. The only significant interaction emerged for BP’s methods subscale (cf. M3, Table 3), suggesting that the more teachers endorsed traditional methods of discipline and authority, the less they perceived increased contact time with undiagnosed pupils as indicating Dependency. The null model for this outcome, involving again the Dependency subscale of the STRS, also returned an ICC = 0, indicating no variance attributable to school-level clustering.
Regarding educational stance and its association with teacher-related variables, we found that BPO (r = −0.28, p = 0.034) and BP “Values of Bleak Pedagogy” factor (r = −0.26, p = 0.046) were negatively correlated with years of teaching. Conversely, BP “Education of children over time” factor, which reflects a nostalgic attitude towards old-fashioned practices considered effective and useful, was positively correlated with teachers’ working hours per week (r = 0.26, p = 0.042). To further test these associations, LMMs were fitted including a random intercept for school: results did not confirm the association between years of teaching and scores on BPO or BP’s values, but a positive association that approached significance was found between teaching hours per week and “Education of children over time” (cf. M4, Table 3).
Moreover, paired-sample t-tests highlighted two important differences. Firstly, teachers scored significantly higher (M = 29.7, SD = 5) on STRS Closeness when referring to an SLD pupil, than when considering a WD pupil (M = 26.2, SD = 5.9): t(63) = 4.708, p < 0.001 (95% CI = 2.18–5.39), d = 0.59. Secondly, teachers tended to perceive higher levels of dependency in the relationship when a pupil with SLD was considered (M = 7.6, SD = 2.86) compared to a pupil without a diagnosis but with poor school performance (M = 6.3, SD = 2.23): t(64) = 4.131, p < 0.001 (95% CI = 0.71–2.05), d = 0.51. To verify whether these dimensions of the pupil-teacher relationship varied depending on the student’s diagnostic status (hypothesis A), two LMMs were conducted: one for the Closeness subscale and one for Dependency. In both models, teachers and schools were treated as random factors to account for the nested structure of the data. Results showed that Closeness scores were significantly higher when teachers described their relationship with a pupil with an SLD compared to one without a diagnosis (cf. M5, Table 3). A similar pattern was found for the Dependency subscale, where higher scores emerged in the case of SLD pupils (cf. M6, Table 3).
In accordance with our first objective (hypotheses A and B), we examined children’s perceptions of the teacher-pupil relationship and potential differences based on gender, age group, and diagnostic status. Table 4 presents the distributions of children’s responses on each of the three STRQ subscales, together with internal consistency coefficients (Cronbach’s α, r.drop), which were satisfactory across all three STRQ subscales.
Distribution data by SLD/WD status and age group (younger/older) were not reported, as no differences were observed across these subgroups. Nonetheless, differences in scores were visible between males and females, considering the whole sample and the younger children subgroup (cf. Table 5).
To further examine whether children’s responses to STRQ varied as a function of their gender, three LMMs were estimated including random intercepts for teacher and school. No significant main effect of gender was found on the three factors of STRQ, namely Closeness (cf. M1, Table 6), Dissatisfaction (cf. M2, Table 6), and Bond with School (cf. M3, Table 6). Although gender did not emerge as a significant predictor on any STRQ dimension, the analyses provided valuable insights into the variance structure of the data. Children’s scores on the Dissatisfaction subscale appear to be influenced by contextual factors: the high conditional R2 (96.7%) and moderate clustering at the school (ICC = 0.69) and teacher levels (ICC = 0.28) suggest that children’s dissatisfaction reflects the environment they experience rather than their gender. For completeness, given the pattern found for teachers regarding weekly hours, we also tested whether pupils’ STRQ scores were linked to the time the teacher reported spending with the specific child, but no significant correlations emerged, either for SLD or WD pupils. This may be due to the fact that children completed the STRQ with reference to their entire group of teachers, so this analysis might prove useful in future studies with individually matched teacher-pupil data.
STRQ Dissatisfaction (SLD pupils) was positively correlated with STRS Conflict (SLD pupils) r = 0.4, p = 0.012, and with STRS Dependency (SLD pupils) r = 0.33, p = 0.025. To account for clustering by school, a LMM was fitted including a random intercept for school. ICC and R2 values are reported collectively, as the predictors were tested within a single model. A random intercept for teacher was not included, as each teacher contributed only one data point for the pupil with SLD, making it impossible to estimate teacher-level variance. Results showed that STRS Conflict significantly predicted higher levels of STRQ Dissatisfaction, while STRQ Dependency did not (cf. Table 7). These results should be interpreted with caution, as they reflect associations between aggregated teacher and student perceptions across schools, rather than matched dyadic data. Nonetheless, the conditional R2 of 63.1% and the ICC of 0.48 suggest that a considerable portion of variance in pupils’ dissatisfaction is linked to school-level factors, highlighting the role of context even when individual predictors are not significant.
Table 8 summarizes the descriptive analyses conducted on the results of the RADSA instrument, providing an initial overview of teachers’ attitudes and beliefs on SLD diagnosis. In pursuit of the second objective, aimed at deepening the understanding of teacher–pupil relationship quality by relating it – among other factors – to teachers’ beliefs about SLD diagnosis, we proceeded to explore the RADSA dimensions and their associations with teacher-related variables.
To test whether these correlations remained when accounting for school-level clustering, four LMMs were estimated including random intercepts for school. Results confirmed significant effects of teacher age on “Attention to the class emotions and individual needs” (cf. M1, Table 9) and “Roles and information” (cf. M2, Table 9). In addition, years of teaching predicted higher scores on “Fairness of evaluation” (cf. M3, Table 9) and lower scores on “Diagnosis as alibi” (cf. M4, Table 9). These findings also reveal that the explained variance is not negligible, particularly for “Roles and information” (R2c = 37.9%), where school-level differences (ICC = 0.33) play a significant role. This suggests that teacher-related variables interact with school contexts in shaping these specific beliefs about SLD diagnosis.
Table 10 illustrates descriptive statistics of the Adultcentrism paradigm and the Bleak Pedagogy educational stance in the present sample. To address hypothesis C, we investigated how adult-centered tendencies and authoritarian educational values and practices align with specific RADSA subscales.
The correlations observed between the RADSA instrument, the ADT scale, and BPO scale, were informative (cf. Table 11). Both Adultcentrism and Bleak Pedagogy positively correlated with two subscales of RADSA section concerning the increase of SLD diagnoses (“System-level causes” and “How children are raised nowadays”).
In order to test more rigorously the associations between RADSA dimensions and the constructs of Adultcentrism and Bleak Pedagogy, a series of LMMs were run (cf. Table 12): results of these analyses confirmed the pattern observed in the bivariate correlations. The ICC and R2 values further indicate that, while the associations between Adultcentrism, Bleak Pedagogy, and certain RADSA dimensions are mainly driven by individual teacher beliefs (low ICC and R2c), some dimensions appear to be more sensitive to school-level factors, as suggested by higher conditional R2 values.
Results showed that teachers’ agreement with “Diagnosis usefulness” dimension significantly predicted STRQ Dissatisfaction scores in SLD pupils (cf. M1, Table 13). Children’s Dissatisfaction was also positively predicted by teachers’ emphasis on the “Roles and Information” dimension (cf. M2, Table 13). Regarding Bleak Pedagogy, higher teacher endorsement of “Values of Bleak Pedagogy” predicted lower STRQ Closeness (cf. M3, Table 13) and weaker Bond with School in SLD pupils (cf. M4, Table 13). Similarly, higher Adultcentrism association with lower STRQ Bond with School scores was confirmed (cf. M5, Table 13).
Teachers’ perception of the relationship quality with SLD pupils was also affected by their educational stance and practice: higher scores on “Methods of Bleak Pedagogy” were associated with lower perceived STRS Closeness (cf. M6, Table 13), but agreement with “Values of Bleak Pedagogy” predicted higher scores on Closeness perceived in the relationship (cf. M7, Table 13). Interestingly, while BP’s values were associated with higher perceived closeness, this pattern diverged from what was found for BP’s methods. This prompted us to focus on BP’s second factor (“Education of children over time”), which also reflects aspects of educational practice—although framed in temporal terms. Among all variables, the only one directly linked to both this second factor and pupils themselves was the weekly hours spent with them. This observation guided our further analysis: we tested whether agreement with BP’s methods would predict higher scores on “Education of children over time” (a nostalgic attitude towards old-fashioned methods), assuming that the way teachers manage time with pupils would reflect a consistent pattern shaped by those methods. The LMM confirmed this hypothesis of association: a higher agreement with BP’s methods predicted a more nostalgic attitude towards those same methods (cf. M8, Table 13). However, the mediation analysis testing the role of weekly hours as a mediator between BP’s methods and “Education of children over time” did not yield significant results (ACME = −0.005, p = 0.86), while the direct effect of BP’s methods remained significant (ADE = 0.421, p = 0.001).
Finally, higher Adultcentrism was confirmed as a predictor of higher STRS Dependency considering both SLD and WD pupils (cf. M9, Table 13). These results should be interpreted within the framework of partially overlapping samples, as they illustrate meaningful associations between beliefs and perceptions at the group level, without assuming matched identities between individual teachers and students. Taken together, the relatively high school-level ICCs and R2c values observed for STRQ models (up to 65.8%)—compared to the minimal clustering in STRS—may suggest that pupils’ perceptions of relational quality are more context-dependent, whereas teachers’ self-perceptions appear comparatively stable. The recurring associations with Bleak Pedagogy-related predictors (M3, M4, M6; Table 13) and the potential influence of Adultcentrism (M5, M9; Table 13) could indicate that authoritarian educational stances and adult-centered attitudes play a role in shaping both relationship quality and pupils’ bonds with the school context.
This study investigated teachers’ perceptions of the quality of their relationships with pupils diagnosed with SLD compared to low achievers without diagnosis, by considering a set of teacher-related factors (teaching experience and weekly working hours) and pupil-related characteristics (age and gender). Furthermore, we examined how these elements interact with broader constructs, such as labeling effects and educational stance, to provide a deeper understanding of the dynamics underlying teacher-student relationships. The quality of such relationship appears to be partly influenced by the presence of an SLD diagnosis, since both the perceived levels of Closeness and Dependency were significantly higher when teachers described the relationship with a pupil with an SLD, rather than with a pupil without a diagnosis but identified by the teacher as having poor school performance was taken into consideration. Previous studies have mainly compared pupils with and without an SLD diagnosis, focusing on differences in teacher-student relationships (e.g., Al-Yagon and Mikulincer, 2004; Murray and Greenberg, 2000; Zee et al., 2020) and exploring teachers’ implicit and explicit attitudes (Hornstra et al., 2010). To our knowledge, very few studies (e.g., Pasta et al., 2013) have directly compared pupils with a formal SLD diagnosis to low achievers without diagnosis. By adopting this finer distinction, our study contributes a novel perspective that helps disentangle the effects of diagnostic labeling from those related solely to academic performance. Our results align with both the primary expectations of the present research and the findings of previous studies. In fact, the STRS Closeness dimension is considered the most representative of a positive relationship (Sabol and Pianta, 2012), and it has been found that the presence and type of diagnosis can be a factor that may influence it (Pasta et al., 2013; Zee et al., 2020). However, this comparison should be interpreted with caution, as the group of WD pupils may include heterogeneous profiles, and the definition of low achievement was based on teachers’ subjective assessments rather than standardized criteria. Therefore, it is important to note that the comparison group may have included pupils with diverse and possibly undiagnosed difficulties, which could intervene in challenging the quality of the pupil-teacher relationship, thus diminishing the perception of closeness in a similar way to what was found for pupils with difficulties that had a greater impact at the behavioral level (Zee et al., 2020).
Instead, the significant effect of weekly hours spent with the WD pupil was confirmed as a predictor of higher levels of Dependency (STRS) and Closeness (STRS) perceived by the teacher. This pattern suggests that increased contact time may be associated with both enhanced relational involvement and a stronger perception of the pupil’s reliance on the teacher. However, the moderation effect of Bleak Pedagogy’s methods indicates that this interpretation is not uniform: the more teachers endorsed authoritarian educational styles, the less they interpreted increased time as a sign of Dependency. This moderation effect may reflect not a lack of relational engagement, but rather a specific normative view of educational responsibility, where closeness is valued and cultivated, yet interpreted through a lens of directive guidance and control – a stance that, if accepted by the other member of the relationship (i.e., the child), may indeed foster dependency without necessarily generating Dissatisfaction, as also suggested by the absence of significant associations with this STRQ subscale. This interpretation is consistent with Maturana and Dávila’s (2006) claim that adult control in the educational relationship generates dependence in children. And, as suggested by Florio et al. (2023), teachers who endorse Bleak Pedagogy’s methods may be genuinely convinced that increased involvement and directivity serve the child’s best interests. In this view, spending more time with a WD pupil does not necessarily imply fostering autonomy or shared agency, but rather ensuring protection, order, and “proper” development. In sum, such time investment may enhance relational closeness while simultaneously fostering a perception of the pupil as dependent – unless such dependency is reframed, by some degree of adherence to authoritarian educational stance, as appropriate receptivity to adult guidance. These interpretations remain provisional, as the mechanisms linking contact time with the pupil, perceived dependency, and educational stance may involve bidirectional dynamics or contextual factors beyond the scope of this study. Longitudinal research would be valuable to examine how these relationships evolve over time and to disentangle the direction of these effects.
Regarding years of experience, previous research has shown that teaching experience can influence teacher-student relationship perceptions (Zee and Koomen, 2017). In our study, however, no significant association was found between teaching experience and the quality of teacher-student relationships, suggesting that this link may depend on contextual factors or other variables not captured in the present design.
The finding that teachers report greater Closeness and Dependency in their relationships with SLD pupils compared to peers without a diagnosis, albeit with equally poor school performance, suggests that the diagnostic label might activate a more empathic, protective, or affiliative stance. The result concerning Closeness appears consistent with the literature emphasizing the positive impact of diagnosis as a relational “framework” (Zee et al., 2020; Pasta et al., 2013), helping teachers to interpret students’ difficulties through a lens that reduces blame and facilitates supportive behaviors. Regarding Dependency, the higher scores in relation to SLD pupils might reflect teachers’ increased willingness to provide help and guidance to students officially recognized as struggling. However, this interpretation warrants caution: over-involvement and perceptions of excessive need may also emerge, especially in classroom environments where the diagnosis becomes a cue for permanent vulnerability rather than a signal of support potential (cf. Pasta et al., 2013). In fact, prior work suggests that a formal diagnosis may alter teachers’ attributions, leading them to view academic difficulties as less changeable and more reflective of the pupil’s inherent limitations which can ultimately reinforce low expectations (Frohlich et al., 2020; Woodcock and Vialle, 2011). It has to be underlined that the cross-sectional nature of our data does not allow for determining such temporal direction of the associations we found.
Some noteworthy findings emerged from the exploration of associations between pupils’ and teachers’ representations of their relationship. Before delving into these results, it is worth recalling that internal consistency analyses supported the reliability of STRQ subscales across both younger and older children. All three subscales showed acceptable to good internal consistency, as indicated by Cronbach’s α and by mean corrected item-total correlation (r.drop), in line with recommended thresholds (Nunnally and Bernstein, 1994; Hair et al., 2019). These findings strengthen the reliability of the instrument even among younger pupils, despite the original validation of STRQ being conducted primarily on 4th and 5th grade students (Tonci et al., 2012).
Correlational analyses between children’s and teachers’ responses revealed that when teachers express that they perceive more Conflict and Dependency in their relationship with SLD pupils, the same pupils report more Dissatisfaction in their relationship with the teachers. This relationship was also confirmed by LMM analysis, showing that especially teacher-perceived conflict seems to play a predictive role toward perceived Dissatisfaction among SLD pupils. These findings align with the extensive literature emphasizing the protective role of positive teacher-student relationships (García-Rodríguez et al., 2023; Haldimann et al., 2023; Jederlund and von Rosen, 2023; Thornberg et al., 2022), particularly for pupils with diagnosed vulnerabilities (Re et al., 2021; Zee et al., 2020). They also support a systemic perspective, which conceptualizes both student and teacher as active constructors of the relationship quality (Pianta, 2001a).
Differently from a previous study that explored teacher-pupil relationship quality using a multi-informant approach (Zee et al., 2020), no significant differences emerged in our sample between SLD and WD pupils in children’s STRQ scores. This result should be considered in light of the mismatch between STRS and STRQ measures, which may partly explain the lack of differences observed as children reported on teachers in general while teachers referred to a specific pupil. It should be noted, however, that Zee and colleagues employed a different instrument – the “Student Perception of Affective Relationship with Teacher Scale” (SPARTS; Koomen and Jellesma, 2015)—and therefore the two studies are not fully comparable. Furthermore, it is important to consider that in our study the grouping into SLD and WD pupils was based on teachers’ subjective assessment, and that the sample did not include children who were perceived by their teachers as having no learning-related difficulties.
On the other hand, in line with the findings of previous studies (Koomen and Jellesma, 2015; La Grutta et al., 2023; Longobardi et al., 2016; Zee and Koomen, 2017), gender differences were found in the perception of the teacher-pupil relationship: female pupils reported significantly higher levels of perceived closeness in the relationship and positive attachment to the school, whereas male pupils expressed greater Dissatisfaction. This pattern remained evident among younger children subgroup. These results are consistent with previous findings suggesting that girls tend to describe more positive relationships with teachers, and a stronger affective bond with the school environment (Murray and Zvoch, 2011; Hamre and Pianta, 2001). In contrast, boys may be more likely to experience conflict and emotional distance in the relational climate, since they tend to express more externalizing emotions (e.g., anger) during middle childhood (Chaplin and Aldao, 2013). Moreover teachers—especially female teachers, though not exclusively – have been found to tend to report lower levels of closeness and/or higher levels of conflict in their relationships with male pupils (Quaglia et al., 2013; Spilt et al., 2012). Such dynamics could make the relational exchange challenging for both parts: if the teacher perceives the relationship as tense, this may be expressed through reduced warmth or patience, and the pupil, in turn, may feel less understood or supported, reinforcing dissatisfaction and disengagement from school environment. This interpretation remains hypothetical and should be tested in future research specifically designed to explore these reciprocal dynamics, particularly since the significance of these differences in paired comparisons was not confirmed by the LMMs: a main effect of gender was absent when accounting for the nested structure of the data. This may suggest that the observed gender-based differences could vary depending on teacher or school specific factors, and that inter-individual variability at higher levels may attenuate these group-level effects when modeled appropriately. Additional classroom or school-level variables, which were not measured here, might further explain these gender-related differences.
According to the second aim of this study, the resulting pattern of correlations, confirmed by LMMs, delineates the profile of a teacher who might struggle when navigating the complex relational and interpretative framework captured by RADSA, especially in presence of strong agreement with Adultcentrism and/or Bleak Pedagogy constructs. A consistent finding is that higher scores on these scales are associated with the belief that the rise in SLD diagnoses is driven by system-level issues, or changes in child-rearing practices. Such associations may reflect a skeptical attitude towards the legitimacy or utility of the diagnostic process itself, since it seems to be perceived as the by-product of social or cultural transformations rather than of individual needs. A possible explanation of this significant association is that this belief system may resonate with what Kimura et al. (2024) define as “authoritarian nostalgia,” that is, “a collective political memory that is painful because the past is perceived in positive and even idyllic ways.” (Kimura et al., 2024, p. 389). In educational settings, this outlook may manifest in a preference for disciplinary structures perceived as lost.
Further differences emerged between the Adultcentric and Bleak Pedagogy perspectives with respect to the perceived role of parents. While Adultcentrism was positively associated with the belief that parents need more training about SLDs and the diagnostic process, BP’s methods subscale showed a negative relationship with this same belief. These diverging directions may reflect the fact that Adultcentrism, as a more theoretical and responsibility-focused stance, promotes the idea that adults should be better informed and proactive. In contrast, agreement with traditional authoritarian methods may discourage collaborative views and portray parental involvement and information as either excessive or unnecessary. This is in line with the skeptical attitude towards the legitimacy or utility of the diagnostic process described above: if the rise in diagnoses is perceived as the result of social or cultural changes, rather than of an improved understanding of learning disorders, then the value of parent training itself may be called into question altogether. On the contrary, the total score on BPO (especially in its values and convictions regarding educational change over time, i.e., first and second factors) and the score on ADT were both positively related to the belief that parents use their child’s diagnosis as an alibi for any weakness or problem. Only BPO, however, was associated with the belief that parents tend to react negatively to SLD diagnoses. When considered together with the previous results, these views appear to reflect a rather critical conception of parenthood: parents are implicitly blamed for their children’s difficulties, either for working too much or for transmitting inadequate values through education, and are seen as reacting defensively to the diagnosis – using it not as a tool for understanding and supporting, but as a form of social protection that shifts the focus onto individual pathology. In this framework, when there is agreement with Adultcentrism and/or Bleak Pedagogy, the reasoning around SLDs seems to revolve more around parents than around the children themselves. This consideration may align with the blaming process that has already been noted in other studies where children’s difficulties (e.g., behavioral, social, emotional) are ascribed to inadequate parenting—particularly to mothers—characterized by chaotic or inconsistent practices (Broomhead, 2013; Francis, 2012; Harborne et al., 2004; Peters, 2012; Rogers, 2007) which ultimately conveys the idea that children are seen as not having been properly raised or taught how to behave, indeed frequent reference are made to ineffective parenting and lack of discipline (Broomhead, 2013).
From a different angle, teachers scoring higher on BP’s values reported a stronger perception of closeness in the relationship with SLD pupils, pointing to a possible disconnection between teachers’ and pupils’ experiences. One possible interpretation is that, when a formal diagnosis is present, these teachers may attribute the difficulties to the diagnosis itself rather than to the child’s intentional behavior, thus reducing blame and making the relational space more accessible. In contrast, when no diagnosis is available, the same difficulties might be interpreted as signs of laziness, lack of motivation, or insufficient discipline: attributions that are likely to undermine the emotional quality of the relationship. This interpretation is consistent with previous studies showing that a clinical label can significantly recalibrate teachers’ causal attributions. Frohlich et al. (2020), for example, found that learning difficulties in students with an LD diagnosis were perceived by teachers as less stable and less controllable than those associated with ADHD, which in turn influenced their sense of self-efficacy and educational approach. Similarly, Woodcock and Vialle (2011) reported that preservice teachers were more likely to attribute poor outcomes in students with SLD to internal and uncontrollable causes, such as ability, rather than effort. Additional evidence indicates that the SLD label can activate stereotypical beliefs that alter teachers’ expectations and interpretations. Kashikar et al. (2023) demonstrated that the presence of a diagnostic label leads teachers to adopt more sympathetic and less punitive attitudes, and to frame difficulties as stable and internal, rather than as behavioral shortcomings. Likewise, Hornstra et al. (2010) observed that teachers tend to hold lower academic expectations for students with dyslexia, based on implicit beliefs associated with the label.
Several authors have pointed out that the way teachers conceive of their role in the educational relationship may shape their sense of indispensability and lead to an overestimation of their own centrality. Bingham (2004) critically observed that exercising authority is a relational activity, yet educators often behave “as if authority were something they owned, something they could choose to use or not,” whereas, in fact, “authority has us rather than us having authority.” (Bingham, 2004, p. 28). This implies that the teacher’s role may be guided less by the learner’s needs and more by historically internalized logics of control and dependency. In this view, students should be “strategic enough to be against the authority of the teacher while being for the authority of the text” (Bingham, 2004, p. 36), an invitation to decenter the adult as the sole mediator of meaning and to depersonalize educational authority, allowing learners to engage with knowledge on their own terms. These reflections align with the idea that the adult’s perceived overemphasized responsibility may not always respond to the child’s actual needs, but rather to deeper, internalized cultural models of educational power.
This idea also resonates with the work of Rinaldi (2021), which is rooted in the Reggio Emilia approach, which highlights the relational nature of education as a process involving mutual influence among children, educators, and families. In this framework, roles are conceived as interdependent, and educational dialogue is not seen as a form of trans

Sections

"[{\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref59\", \"ref4\", \"ref64\"], \"section\": \"Introduction\", \"text\": \"Over recent years, in Italy, within a general rise in neurodevelopmental disorders (Lucangeli and Perini, 2020), the number of students diagnosed with Specific Learning Disorder (APA, 2022) has notably increased. A report produced by the Statistical Office of the Ministry of Education and Merit (MIM, 2024) shows that the prevalence of Specific Learning Disorder (SLD) diagnoses rose from 0.9% in school year (SY) 2010/2011 to 6% in SY 2022/2023. Looking at the different types of SLD, the past decade has seen a steady increase across all categories: the number of pupils certified with dyslexia has more than doubled, while cases of dysgraphia, dysorthography, and dyscalculia have more than tripled. Considering only primary school, the percentage of pupils diagnosed with SLD increased from 0.8% in SY 2010/2011 to 3.2% in SY 2022/2023. This increase should also be considered in light of the fact that, starting in 2010 with the enactment of Law 170/2010, awareness of SLDs in Italy significantly grew, contributing to a broader recognition and reporting of such cases.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref16\", \"ref90\"], \"section\": \"Rising SLD diagnoses and educational challenges\", \"text\": \"As a consequence of the increasing number of SLD diagnoses, the current expectation for teachers to manage a wide variety of special educational needs (e.g., Specific Learning Disorder - SLD) and to promote an inclusive classroom atmosphere (Cornoldi et al., 1998; Saloviita and Consegnati, 2019) requires them to continuously adjust and update their educational beliefs and methods to face the challenges of the current school context and the needs of individual pupils. The following section outlines the theoretical background and research context that motivate the present study, highlighting current educational challenges linked to the growing prevalence of SLD diagnoses, the potential effects of labeling, the importance of teacher-student relationships, and the role of educational stance, concluding with the aims and hypotheses of the study.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref76\", \"ref33\", \"ref75\", \"ref77\", \"ref80\", \"ref78\", \"ref79\", \"ref33\", \"ref89\", \"ref101\", \"ref45\", \"ref33\", \"ref91\", \"ref93\", \"ref53\", \"ref113\", \"ref112\", \"ref54\", \"ref58\", \"ref37\", \"ref92\", \"ref112\"], \"section\": \"The protective role of teacher-student relationships\", \"text\": \"Given the challenges arising from the increasing complexity and broader acknowledgment of pupils\\u2019 varied needs, it becomes crucial to understand how the quality of the relationship between teachers and pupils can serve as a protective factor in promoting positive learning and developmental outcomes. In fact, the establishment of a supportive relationship between teacher and pupil is an important element of protection that is advantageous to the development of the child (Pianta, 2001a). The systematic review conducted by Garc\\u00eda-Rodr\\u00edguez et al. (2023) highlighted that teachers\\u2019 attachment style and their availability in responding to student needs are associated with students\\u2019 school adjustment. The Student-Teacher Relationship Scale (Pianta, 1994, 2001b; Pianta et al., 1995; Pianta and Nimetz, 1991; Pianta and Steinberg, 1992), identified as \\u201cthe most frequently used measurement instrument\\u201d (Garc\\u00eda-Rodr\\u00edguez et al., 2023, p. 24), highlights various beneficial aspects of a relationship characterized by high levels of closeness and low levels of conflict and dependency within the theoretical framework of the attachment theory (Sabol and Pianta, 2012). Research consistently highlights the association of the quality of the teacher-student relationship with multiple components of pupils\\u2019 school experience, such as students\\u2019 engagement (Thornberg et al., 2022), self-efficacy in both the learning domain and academic success (Jederlund and von Rosen, 2023), concept-of-self, emotional regulation (Garc\\u00eda-Rodr\\u00edguez et al., 2023), and cognitive abilities (Sankalaite et al., 2023; Semeraro et al., 2014). It is worth noting, however, that individual differences may also shape these relationships. From both teachers\\u2019 and pupils\\u2019 perspectives, some results in the literature on the quality of the pupil-teacher relationship suggest the presence of gender and age differences: females seem to build relationships characterized by greater closeness and less conflict (Koomen and Jellesma, 2015; Zee and Koomen, 2017; Zee et al., 2020). Therefore it is particularly important to highlight that, in the Italian context, these results could also be influenced by the prevalence of female teachers, especially in primary schools (La Grutta et al., 2023; Longobardi et al., 2016). Considering age differences, some studies showed a decrease in school attachment in the last years of primary school (Guidetti et al., 2017; Schneider et al., 2008; Zee et al., 2020). Consequently, it could be hypothesized that the overall quality of teacher-student relationships also declines during these final years, making age a relevant factor to consider among the other factors examined in this study.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref10\", \"ref46\", \"ref61\"], \"section\": \"The protective role of teacher-student relationships\", \"text\": \"Finally, previous studies suggest that higher workload and accumulated fatigue may hinder the ability to sustain consistent emotional attunement both with pupils in general and with those requiring more complex support (Bodenheimer and Shuster, 2020; Jennings and Greenberg, 2009; Maas et al., 2021). Therefore, among the factors that shape teachers\\u2019 perceptions, professional characteristics such as teaching experience and weekly working hours can be considered relevant, as they may influence the emotional resources and responsiveness available to build high-quality relationships with pupils.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref73\", \"ref112\", \"ref112\", \"ref112\", \"ref112\", \"ref2\", \"ref65\", \"ref2\", \"ref71\"], \"section\": \"SLD labeling and its impact on teachers\\u2019 perceptions\", \"text\": \"Beyond teachers\\u2019 professional characteristics, the quality of teacher-student relationships may vary significantly when pupils face learning challenges or carry diagnostic labels, such as those related to SLD, which can influence teachers\\u2019 perceptions and attitudes. Building on the literature showing the protective role of the teacher-student relationship, it is important to examine how this role applies when students face significant learning challenges, such as SLD. Pennings et al. (2016) emphasize the association between a high-quality teacher-student relationship and teachers\\u2019 ability to respond to their students\\u2019 needs. More specifically, Zee et al. (2020) studied the influence of the teacher-student relationship in a sample of primary school pupils who were diagnosed with dyslexia, an autism spectrum disorder (ASD), or attention deficit hyperactivity disorder (ADHD). They found that the impact of the relationship quality seemed to be higher in pupils with behavior-related disorders rather than in pupils with learning disorders (Zee et al., 2020). Nevertheless, the multi-informant approach also enabled the detection of some differences in the quality of the relationship as perceived by teachers, classmates, and pupils with a dyslexia diagnosis (Zee et al., 2020). Although all informants agreed on the lower level of conflict in the relationship, pupils with a dyslexia diagnosis, unlike both teachers and classmates, perceived less closeness with their teachers compared to pupils without this type of diagnosis (Zee et al., 2020). These differences in pupils\\u2019 and teachers\\u2019 perspectives suggest the relevance of collecting both teachers\\u2019 and pupils\\u2019 points of view, as done in the present study. Moreover, the perception of lower emotional availability of teachers coming from children with learning disorders in comparison with pupils without a diagnosis, detected also in previous studies (Al-Yagon and Mikulincer, 2004; Murray and Greenberg, 2000), provides a first reason to explore the quality of the teacher-student relationship in a sample of pupils an SLD diagnosis. The second reason stems from the investigation of teachers\\u2019 perspectives. In their relationship with pupils having a Specific Learning Disorder, teachers reported a lower perception of emotional closeness (Al-Yagon and Mikulincer, 2004) and a higher dependency in the relationship (Pasta et al., 2013) in comparison with the quality of the relationship with pupils without this diagnosis.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref44\"], \"section\": \"SLD labeling and its impact on teachers\\u2019 perceptions\", \"text\": \"The understanding of teachers\\u2019 attitudes toward pupils with an SLD diagnosis has been explored in the literature using both implicit and explicit measures. The use of implicit measures detected the presence of a \\u201clabeling bias\\u201d of teachers toward pupils with dyslexia (Hornstra et al., 2010), suggesting that the diagnosis itself may influence teachers\\u2019 perceptions beyond the observable learning difficulties. Therefore, distinguishing between pupils with an SLD diagnosis and low achievers without this diagnosis is crucial to disentangle the effects of SLD labeling from those of performance.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref103\", \"ref96\", \"ref97\", \"ref106\", \"ref83\", \"ref103\", \"ref15\", \"ref15\"], \"section\": \"SLD labeling and its impact on teachers\\u2019 perceptions\", \"text\": \"At a more explicit level of investigation, studies explored educators\\u2019 perceptions (Tosun et al., 2021) and misconceptions concerning the knowledge of dyslexia (Soriano-Ferrer and Echegaray-Bengoa, 2014; Soriano-Ferrer et al., 2016; Wadlington and Wadlington, 2005). The role of primary school teachers\\u2019 knowledge about SLD was found to be a predictor for the use of compensatory tools, and teachers\\u2019 beliefs and emotions about SLD can predict the use of dispensatory accomodations (Re et al., 2021), showing the connection between teachers\\u2019 representational framework and their educational practice. Moreover, Tosun et al. (2021) revealed a weak correlation between teaching experience and the negative perception of dyslexia. The explorative study conducted by Cornoldi et al. (2018) extended the focus on primary school teachers\\u2019 attitudes toward learning disabilities, including beliefs about the roles of legislation, teachers, specialized intervention, compensatory measures, integration in the classroom, the labelling effect and emotional challenges in the experience of their pupils with learning disabilities. Although some beliefs were shared across the three countries involved in the study (Italy, Spain, and the U.S.), significant differences emerged (Cornoldi et al., 2018), suggesting the relevance of investigating the multiple variables that could intervene in influencing teachers\\u2019 attitudes and beliefs that guide their educational practices.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref9001\", \"ref101\", \"ref32\", \"ref36\", \"ref22\", \"ref9001\", \"ref22\", \"ref62\", \"ref24\", \"ref87\", \"ref88\", \"ref23\", \"ref48\", \"ref81\", \"ref26\", \"ref21\", \"ref28\", \"ref23\", \"ref23\", \"ref9002\", \"ref24\", \"ref9001\"], \"section\": \"Educational stance and relationship quality in the classroom\", \"text\": \"In addition to diagnostic labels, another key factor to consider is the broader perspective of the educational stance that shapes teachers\\u2019 approaches to relationships with pupils and classroom management. To do so, it seems necessary to develop a deeper understanding of teachers\\u2019 ideas concerning an SLD diagnosis within a comprehensive and complex framework of beliefs and attitudes that contribute to defining the teacher\\u2019s educational stance (Florio, 2018). Adultcentrism and Bleak Pedagogy are two constructs that enable the exploration of specific beliefs and attitudes influencing teachers\\u2019 educational stance, as manifested in their methodological and classroom management choices. Indeed, as recognized by students, to define a \\u201cgood teacher,\\u201d it is necessary to consider the interconnection among personal, relational, methodological, and management factors (Thornberg et al., 2022). Adultcentrism (Furioso, 2000; Goode, 1986) is an implicit paradigm of thought rooted in a sociocultural framework in which adulthood is regarded as the full realization of human development, in a dichotomous comparison with the perceived incompleteness of childhood (Florio et al., 2020a). Although adults intend to act in the best interests of children, the power asymmetry and the depiction of children as incomplete, immature, and incompetent individuals that characterized the adultcentric perspective could lead to inadequate recognition, understanding, and responses to children\\u2019s needs (Florio, 2018; Florio et al., 2020a; Mackay, 2003). Florio et al. (2022a) demonstrated that, in a sample of primary school teachers, an adultcentric perspective is a consistent predictor of the values and beliefs of an old-fashioned and authoritarian educational stance, namely the construct of Schwarze P\\u00e4dagogik (Rutschky, 1977, 2015), which literally translates as Black Pedagogy (Florio et al., 2020b). It is necessary to clarify that this label does not pertain in any way to education involving black students and teachers, nor school programs including black studies (Johnson et al., 2014; Pitre et al., 2008): the label has been recently adapted, and the term currently used by this group of authors is Bleak Pedagogy (Florio et al., 2024a, 2024b; Florio et al., n.d.). Bleak Pedagogy comprises values, beliefs, and methods focused primarily on shaping the pupils according to ideal standards of educators and society, thereby undermining the pupils\\u2019 will (Florio et al., 2020b). The Bleak Pedagogy methods also include forms of both physical and psychological violence for disciplinary purposes (Florio et al., 2020b; Perticari, 2016). The sociocultural context also influences educators\\u2019 ability to recognize this detrimental educational stance, which \\u2013 especially in its psychological forms \\u2013 tends to remain implicit and, for this very reason, is often enacted in practice (Florio et al., 2022a). The first evidence (Florio, 2018) suggested that the agreement with Adultcentrism and especially Bleak Pedagogy could predict specific features of teachers\\u2019 representations concerning an SLD diagnosis, such as a critical attitude toward the role of society (e.g., medicalization, how children are raised nowadays), specialists, parents, and classmates in determining the presence and the school experience of pupils with an SLD diagnosis. These results underline the importance of investigating teachers\\u2019 attitudes and beliefs regarding their complex role and its impact on the quality of the teacher-student relationship, especially when dealing with a pupil diagnosed with an SLD or perceived as demonstrating low academic achievement.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref44\", \"ref15\", \"ref103\", \"ref112\"], \"section\": \"From theory to research framework\", \"text\": \"Despite the growing body of research on teacher-student relationships and SLD, few studies have directly compared teachers\\u2019 perceptions of pupils with a formal SLD diagnosis and low achievers without a diagnosis. This comparison is particularly relevant for disentangling the potential effects of labeling from the underlying learning difficulties. Building on the literature on labeling bias (Hornstra et al., 2010) and regarding teachers\\u2019 attitudes toward SLD (Cornoldi et al., 2018; Tosun et al., 2021), our study aims to examine how teachers perceive and describe the quality of their relationships with these two groups of pupils. We also explore whether factors such as teaching experience and weekly working hours are associated with these perceptions, as suggested by previous findings on teacher fatigue and workload (Zee et al., 2020). Specifically, we ask whether the quality of teacher-student relationship differs between pupils with and without an SLD diagnosis, and how teachers\\u2019 professional characteristics\\u2014such as educational stance\\u2014might influence their views. These considerations form the basis of our study objectives and hypotheses, outlined below. Building on these premises, the present study integrates variables such as labeling effects, teacher workload, experience, and pupils\\u2019 age, to offer a multi-layered perspective that extends existing research on teacher-student relationships and SLD.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref44\", \"ref71\", \"ref83\", \"ref112\", \"ref37\", \"ref54\", \"ref58\", \"ref103\"], \"section\": \"Aims of the study\", \"text\": \"The first objective is to investigate perceptions of the quality of the teacher-student relationship with pupils who have an SLD diagnosis (Specific Learning Disorder diagnosis\\u2014SLD pupils). The first hypothesis (A) is that the presence of the diagnostic label of a Specific Learning Disorder could influence teachers\\u2019 and pupils\\u2019 representations of their relationship, even when compared with pupils who, in the eyes of teachers, have some difficulties in dealing with learning challenges (Without a Diagnosis - WD pupils) (Hornstra et al., 2010; Pasta et al., 2013; Re et al., 2021; Zee et al., 2020). The second hypothesis (B) is that pupils\\u2019 age and gender, as well as teachers\\u2019 experience and working hours, may influence the quality of the teacher-student relationship, as highlighted by previous literature (Guidetti et al., 2017; La Grutta et al., 2023; Longobardi et al., 2016; Tosun et al., 2021).\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref83\", \"ref112\", \"ref71\"], \"section\": \"Aims of the study\", \"text\": \"The second objective is to enrich the perspective on the quality of student-teacher relationship through the investigation of teachers\\u2019 beliefs and attitudes toward SLD diagnosis and the paradigm behind their educational approach, considering also the possible influence of teachers\\u2019 demographics. The first hypothesis (C) is that adult-centered and authoritarian educational practices are associated with difficulties in the educational relationship and greater teacher fatigue in dealing with the complexity that SLD diagnoses bring to the table. The second hypothesis (D) is that teachers\\u2019 beliefs and attitudes toward learning disabilities could influence both the pupils\\u2019 school experience (Re et al., 2021; Zee et al., 2020) and how teachers perceive the relational quality with pupils (Pasta et al., 2013).\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref111\", \"ref3\", \"ref1\"], \"section\": \"Participants and procedures\", \"text\": \"Teachers and children belonged to 26 different primary schools located in the province of Bergamo, in Italy. All participants were treated in accordance with the Declaration of Helsinki (World Medical Association, 2013), as well as with the ethical guidelines for research provided by the American Psychological Association (APA, 2017) and the Italian Psychological Association (AIP, 2022). Participants were also provided with the researcher\\u2019s contact information to receive further clarifications if needed, and informed consent was obtained to proceed with the battery of questionnaires. Participants received clear information about the objectives and phases of the research and about their rights as participants, including the guarantee of anonymity and the possibility of dropping out of the study at any moment. The following subsections describe the characteristics of the two participating groups\\u2014teachers and pupils\\u2014as well as the procedures adopted for each group.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref75\", \"ref77\", \"ref80\", \"ref78\", \"ref79\", \"ref65\", \"ref102\", \"ref25\", \"ref23\", \"ref24\"], \"section\": \"Measures\", \"text\": \"According to the aim of the study, measures investigating the quality of the teacher-student relationship were used. Both teachers\\u2019 perspectives (STRS - Student-Teacher Relationship Scale; Pianta, 1994, 2001b; Pianta et al., 1995; Pianta and Nimetz, 1991; Pianta and Steinberg, 1992) and pupils\\u2019 perspectives (STRQ - Student-Teacher Relationship Questionnaire; Murray and Greenberg, 2000; Tonci et al., 2012) were taken into account, and a measure to collect teachers\\u2019 attitudes and beliefs regarding SLD diagnosis (RADSA - Representations and Attitudes towards SLD diagnoses questionnaire; Florio et al., 2022b) was also included. In order to fulfill the second specific objective, the Bleak Pedagogy Scale (Florio et al., 2020b) and the Adultcentrism Scale for Teachers (Florio et al., 2022a) were used, allowing further exploration of the complexity of the educational relationship, including the paradigm of thought, values, and methods underlying teachers\\u2019 educational approaches.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref75\", \"ref77\", \"ref80\", \"ref78\", \"ref79\", \"ref29\", \"ref77\", \"ref77\"], \"section\": \"Measures administered to teachers\", \"text\": \"Teachers were asked to indicate one pupil with an SLD diagnosis (SLD pupils), and one pupil they considered to have poor school performance but without any diagnosis (WD pupils). In particular, to investigate the quality of the teacher-pupil relationship they were asked to complete the Italian version of the Student-Teacher Relationship Scale (Pianta, 1994, 2001b; Pianta et al., 1995; Pianta and Nimetz, 1991; Pianta and Steinberg, 1992) twice. The STRS is composed of 22 items in its Italian version validated by Fraire et al. (2013), reporting good internal consistency for the subscales (\\u201cCloseness\\u201d: \\u03b1\\u202f=\\u202f0.85; \\u201cConflict\\u201d: \\u03b1\\u202f=\\u202f0.92; \\u201cDependency\\u201d: \\u03b1\\u202f=\\u202f0.69). The initial instructions of the STRS ask teachers to think of a particular pupil while answering the questions. Subsequently, items describing relationship characteristics are presented, and participants indicate how much each statement reflects their relationship with the pupil in question. Responses are set on a 5-point Likert scale of applicability (1\\u202f=\\u202fDefinitely does not apply; 2\\u202f=\\u202fDoes not really apply; 3\\u202f=\\u202fNeutral, not sure; 4\\u202f=\\u202fSomewhat applies; 5\\u202f=\\u202fDefinitely applies). The Italian version of STRS maintains the original three-dimensional structure, composed of \\u201cCloseness\\u201d (8 items), \\u201cConflict\\u201d (10 items) and \\u201cDependency\\u201d (4 items). The \\u201cCloseness\\u201d subscale provides information about the degree of warmth, affection and open communication experienced by the teacher with a particular student; the \\u201cConflict\\u201d subscale provides an evaluation of the level of negativity and conflict present in the relationship; lastly, the \\u201cDependency\\u201d subscale measures the extent to which the teacher perceives the student as overly dependent (Pianta, 2001b). A higher total score on the STRS indicates an overall positive relationship, which tends to be characterized by lower levels of conflict and dependency, and by higher levels of closeness (Pianta, 2001b, p. 12).\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref25\", \"ref25\", \"tab1\", \"ref25\"], \"section\": \"Teachers\\u2019 attitudes towards SLD diagnoses\", \"text\": \"In addressing the complexities associated with SLD diagnoses, the Representations and Attitudes towards SLD diagnoses questionnaire was chosen (RADSA; Florio et al., 2022b) as the tool that best allows teachers to examine their own representations. This makes evident possible implicit obstacles that may affect the quality of the relationship, especially with SLD pupils. The RADSA questionnaire, validated on a sample of Italian primary school teachers (Florio et al., 2022b), is composed of 62 items with a 4-point Likert scale measuring the level of agreement (1\\u202f=\\u202fFully disagree; 2\\u202f=\\u202fSlightly agree; 3\\u202f=\\u202fAgree; 4\\u202f=\\u202fFully agree). RADSA is organized in four sections: \\u201cIncrease of SLD diagnoses,\\u201d \\u201cPeer group,\\u201d \\u201cParents\\u2019 attitudes towards SLD diagnosis,\\u201d \\u201cThe moment when the teacher refers parents to an SLD specialist.\\u201d Each section is composed of three factors. Given the multidimensional structure of the instrument, Table 1 illustrates its sections and factors, along with example items. For a comprehensive overview of the RADSA and its characteristics, (see Florio et al., 2022b). As no internal consistency estimates were reported in the original validation of the RADSA, we computed Cronbach\\u2019s \\u03b1 on the full set of items. The resulting internal consistency was good (\\u03b1\\u202f=\\u202f0.84).\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref23\", \"ref24\", \"ref23\"], \"section\": \"Teachers\\u2019 educational stance\", \"text\": \"Teachers\\u2019 agreement with attitudes and methods typical of an old-fashioned and authoritarian educational stance was evaluated with the first section of the Bleak Pedagogy Scale (originally published as the Black Pedagogy Scale; Florio et al., 2020b). The first section is named \\u201cBleak Pedagogy Observation\\u201d section (BPO), which consists of 24 items, with a Cronbach\\u2019s \\u03b1 of 0.87. Teachers are requested to express their degree of agreement with each item on a 4-point Likert scale. This section of the instrument is composed of three factors: \\u201cValues of Bleak Pedagogy\\u201d (\\u03b1\\u202f=\\u202f0.87; high values: agreement with the main educational values and objectives typical of the Bleak Pedagogy\\u2019s perspective), \\u201cEducation of children over time\\u201d (\\u03b1\\u202f=\\u202f0.75; high values: nostalgic attitude towards old-fashioned educational practices as more effective and useful), and \\u201cMethods of Bleak Pedagogy\\u201d (\\u03b1\\u202f=\\u202f0.74; high values: agreement with Bleak Pedagogy disciplinary and educational methods, such as pedagogical beating, humiliation, coldness towards the child, etc.). The original instrument also includes a first item for temporal placement and a further a section assessing the perceived prevalence of Bleak Pedagogy methods in the past and today, which were not used in the present study. The Bleak Pedagogy Scale has been validated on a sample of Italian teachers (Florio et al., 2022a), following a first study (Florio et al., 2020b) conducted with a sample of Italian university students and parents.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref24\", \"ref24\", \"ref22\"], \"section\": \"Teachers\\u2019 educational stance\", \"text\": \"The second measurement tool used is the Adultcentrism Scale for Teachers (ADT; Florio et al., 2022a), composed of 9 items (Cronbach\\u2019s \\u03b1\\u202f=\\u202f0.74) with a 4-point Likert scale based on agreement. High scores indicate that teachers tend to think of their pupils as disempowered and without responsibility, as an empty receptacle in need of being filled by adults with social and cultural values. The validation of this version of the instrument on the Italian teachers\\u2019 sample was conducted by Florio et al. (2022a) after the first validation study for the original scale (Florio et al., 2020a) with a sample of Italian parents and university students.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref76\", \"ref65\", \"ref102\", \"ref102\", \"ref18\", \"ref34\", \"ref99\", \"ref70\", \"ref38\"], \"section\": \"Measure administered to pupils\", \"text\": \"Students play an active role in constructing the teacher-student relationship and determining its quality and dynamics (Pianta, 2001a). Therefore, to avoid a partial representation of this relationship, the Student-Teacher Relationship Questionnaire (Murray and Greenberg, 2000) in its Italian version (Tonci et al., 2012) was presented to children, to collect data on how they experience their relationship with their teacher. The instrument, validated by Tonci et al. (2012) on a sample of 4th and 5th grade Italian pupils, consists of 22 items with possible answers based on a 4-point Likert scale (1\\u202f=\\u202fNot at all true; 2\\u202f=\\u202fNot very true; 3\\u202f=\\u202fQuite true; 4\\u202f=\\u202fVery true), loading on the following three subscales: \\u201cCloseness with teachers\\u201d (\\u03b1\\u202f=\\u202f0.86; high scores: positive evaluation of the relationship in terms of emotions, closeness and support), \\u201cDissatisfaction with teacher\\u201d (\\u03b1\\u202f=\\u202f0.76; high scores: child\\u2019s negative opinions and feelings concerning the teacher), and \\u201cBond with school\\u201d (\\u03b1\\u202f=\\u202f0.71; high scores: positive feelings towards the school environment and activities). In our sample, Cronbach\\u2019s \\u03b1 exceeded 0.76 for the younger group and 0.74 for the older group. The coefficients, ranging from 0.742 to 0.875, are generally considered acceptable to good (DeVellis, 2016; George and Mallery, 2003; Taber, 2018). For each STRQ subscale, mean corrected item-total correlations (r.drop) were also computed: coefficients ranged from 0.481 to 0.682 across age groups. These values are considered acceptable and ideal values (Nunnally and Bernstein, 1994; Hair et al., 2019), supporting internal consistency alongside Cronbach\\u2019s \\u03b1.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref66\", \"ref17\", \"ref50\", \"ref109\", \"ref102\", \"ref18\", \"ref34\", \"ref99\", \"ref99\"], \"section\": \"Data analysis\", \"text\": \"The analyses were performed in R (Version 2025.05.0\\u202f+\\u202f496). Analyses were designed to address the specific hypotheses outlined in section 1.6 (Aims of the Study). The corresponding statistical procedures are briefly recalled below to guide the reader through the structure of the results. The absence of important violations of normality was assessed, as values of skewness and kurtosis between \\u22121 and +1 are considered acceptable (Muth\\u00e9n and Kaplan, 1985). Thresholds of sk > 2 and ku > 7, values that would indicate a severe violation of normality, were not exceeded (Costello and Osborne, 2005; Kim, 2013; West et al., 1995). Regarding STRQ, internal consistency coefficients (Cronbach\\u2019s \\u03b1) were computed separately for younger and older children, since the original validation of STRQ was conducted on 4th and 5th grade pupils only (Tonci et al., 2012). Values of \\u03b1 >\\u202f0.7 were considered acceptable, whereas \\u03b1 >\\u202f0.8 indicated good reliability (DeVellis, 2016; George and Mallery, 2003; Taber, 2018). We acknowledge that Cronbach\\u2019s \\u03b1 by itself provides only a partial index of scale quality and should not be interpreted as definitive evidence of dimensionality or construct validity; further analyses (e.g., item-total correlations or factor analysis) would be warranted to fully support the internal structure of the subscales (Taber, 2018).\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref102\", \"ref43\", \"ref85\", \"ref70\", \"ref38\"], \"section\": \"Data analysis\", \"text\": \"However, we chose not to conduct factor analysis for several reasons: first, STRQ was used in its validated structure without modification. Second, as detailed communalities (h) were not reported in the original Italian validation of STRQ (Tonci et al., 2012), it is difficult to assess whether our sample size meets the conditions for stable factor recovery: according to Hogarty et al. (2005), when h are unknown or potentially low, sample to variable ratios (N/p) below 10:1\\u2014such as in our case (N/p\\u202f=\\u202f6.5:1)\\u2014may be insufficient to ensure stability. Lastly, factor analysis was beyond the scope of the present study, which did not aim to revalidate STRQ on a younger sample. Instead, item-total correlations were computed using the \\u201csirt\\u201d package (Robitzsch, 2024). The mean corrected item-total correlation (r.drop) is reported and interpreted following Nunnally and Bernsteof (1994), who suggest accepting values of r.drop \\u2265 0.3, with values above 0.5 considered ideal (Hair et al., 2019).\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref13\"], \"section\": \"Data analysis\", \"text\": \"To fulfill the first purpose of the study and to verify the related hypotheses (A and B), analyses considering teachers\\u2019 and pupils\\u2019 perceptions of the relationship involving SLD vs. WD students were carried out. These included descriptive statistics for STRS and STRQ subscales, correlations to detect associations between STRS subscales and STRQ scores, and paired t-tests to compare STRS and STRQ subscales scores. For paired-sample t-tests, Cohen\\u2019s d was calculated to estimate the magnitude of differences, in accordance with standard guidelines (Cohen, 1988). Correlations with teachers\\u2019 demographics and independent-sample t-tests have been conducted to detect significant differences in all variables based on each student\\u2019s school grade (i.e., younger: 2nd and 3rd grade, older: 4th and 5th grade) and gender, to verify the second hypothesis related to this first objective.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref13\", \"ref9003\", \"ref9004\"], \"section\": \"Data analysis\", \"text\": \"In line with the second aim of the study (hypotheses B and C), descriptive statistics for the other measures (i.e., RADSA and Adultcentrism/Bleak Pedagogy) were calculated. To verify the related hypotheses, correlations were performed among all the subscales and teachers\\u2019 demographics. Regarding Pearson product\\u2013moment correlations, Cohen\\u2019s guidelines (Cohen, 1988) were used to interpret their strength: values of r falling within the range of 0.50 to 1 are considered \\u201clarge,\\u201d those between 0.30 and 0.49 \\u201cmedium,\\u201d and those between 0.10 and 0.29 \\u201csmall.\\u201d When significant correlations warranted further investigation, to account for the nested structure of the data (i.e., students nested within teachers and schools), Linear Mixed-effects Models (LMMs) were performed using the \\u201clme4\\u201d package (Bates et al., 2015) for LMMs and the \\u201clmerTest\\u201d package (Kuznetsova et al., 2017) to compute p-values via Satterthwaite\\u2019s approximation for degrees of freedom. When it was not possible to include both school and teacher as random intercepts due to convergence issues, only the school-level intercept was retained to account for the nested data structure. For these analyses, data were first restructured from wide to long format to allow for within-teacher comparisons, since each teacher completed STRS twice (once for a student with SLD and once for a student without a diagnosis). LMMs were also based on the relevant hypotheses: fixed effects included diagnostic status (SLD vs. WD), pupil demographics (gender, age group), teacher characteristics (experience, weekly hours), and theoretical variables such as Adultcentrism and Bleak Pedagogy scores. Random intercepts for school were included in all models, and random intercepts for teacher were added when multiple ratings per teacher were available. When teacher-level variance was negligible (singular fit), only school-level random intercepts were retained.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref19\", \"ref55\", \"ref104\", \"ref95\", \"ref60\", \"ref47\", \"ref68\", \"ref67\"], \"section\": \"Data analysis\", \"text\": \"Intraclass correlation coefficients (ICC) from null models were computed to justify the multilevel approach and are presented in the Results section. ICCs were derived from the variance components obtained via the VarCorr{lme4} function (Devine et al., 2024). Values were interpreted as follows: ICC < 0.05\\u202f=\\u202fpoor, 0.05\\u20130.75\\u202f=\\u202fsmall to moderate, 0.75\\u20130.90\\u202f=\\u202fgood, > 0.90\\u202f=\\u202fexcellent (Liljequist et al., 2019; Trevethan, 2017). However, it should also be noted that ICC values between 0.05 and 0.25 are quite common in social sciences (Snijders and Bosker, 2012). When the ICC was equal to 0, indicating no variance attributable to the grouping factor, the LMM approach was nonetheless retained if the model was statistically significant, to ensure consistency across analyses. To complement the information provided on each LMM, we computed marginal and conditional R2 values with \\u201cperformance\\u201d package (L\\u00fcdecke et al., 2021). These quantify, respectively, the proportion of variance explained by the fixed effects alone (R2\\nm) \\u2013 providing insight into the effect size attributable to the predictors \\u2013 and the total variance explained by the full model (R2\\nc), accounting for both fixed and random effects (Johnson, 2014; Nakagawa and Schielzeth, 2013; Nakagawa et al., 2017).\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab2\"], \"section\": \"Teachers\\u2019 perspective\", \"text\": \"Table 2 presents the descriptive analyses performed on teachers\\u2019 responses to the STRS measure, in its two administered versions. In line with hypotheses A and B, we explored whether the diagnostic label and teacher-related variables shaped teachers\\u2019 perceptions of the relationship and their educational stance.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab3\", \"tab3\"], \"section\": \"Teachers\\u2019 perspective\", \"text\": \"STRS Closeness (SLD pupils) was negatively correlated with weekly teaching hours (r\\u202f=\\u202f\\u22120.25, p\\u202f=\\u202f0.04), indicating that the greater the number of work hours, the less closeness was perceived in the relationship with an SLD pupil. The same dimension of Closeness with a WD pupil was positively correlated with years of teaching (r\\u202f=\\u202f0.24, p\\u202f=\\u202f0.05), showing a marginally significant association. With WD pupils, the STRS Dependency dimension was positively correlated with hours of teaching per week (r\\u202f=\\u202f0.25, p\\u202f=\\u202f0.04). These three associations were further examined through separate LMMs, each including teaching experience (years of teaching) and weekly hours spent with pupils as fixed effects and school as a random intercept. Neither the effect of teachers\\u2019 work experience on Closeness (WD pupils), nor that of weekly hours spent on Closeness (SLD pupils) yielded significant results. Thus, only the LMMs that showed significant effects are reported in detail. Specifically, for Closeness with WD pupils, the number of weekly hours spent with the pupil was a significant predictor (cf. M1, Table 3), indicating that more contact time was associated with higher perceived Closeness in the relationship with a WD pupil. Finally, Dependency scores related to pupils without a diagnosis were also significantly predicted by the weekly hours spent with the pupil (cf. M2, Table 3), suggesting that teachers perceive undiagnosed pupils as more dependent when they spend more time with them.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab3\"], \"section\": \"Teachers\\u2019 perspective\", \"text\": \"Given these results, we explored through LMMs whether teachers\\u2019 endorsement of Adultcentrism and Bleak Pedagogy moderated the associations between time spent with the WD pupil and STRS-perceived Closeness or Dependency. This analysis further contributed to the investigation of hypothesis D. The only significant interaction emerged for BP\\u2019s methods subscale (cf. M3, Table 3), suggesting that the more teachers endorsed traditional methods of discipline and authority, the less they perceived increased contact time with undiagnosed pupils as indicating Dependency. The null model for this outcome, involving again the Dependency subscale of the STRS, also returned an ICC\\u202f=\\u202f0, indicating no variance attributable to school-level clustering.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab3\"], \"section\": \"Teachers\\u2019 perspective\", \"text\": \"Regarding educational stance and its association with teacher-related variables, we found that BPO (r\\u202f=\\u202f\\u22120.28, p\\u202f=\\u202f0.034) and BP \\u201cValues of Bleak Pedagogy\\u201d factor (r\\u202f=\\u202f\\u22120.26, p\\u202f=\\u202f0.046) were negatively correlated with years of teaching. Conversely, BP \\u201cEducation of children over time\\u201d factor, which reflects a nostalgic attitude towards old-fashioned practices considered effective and useful, was positively correlated with teachers\\u2019 working hours per week (r\\u202f=\\u202f0.26, p\\u202f=\\u202f0.042). To further test these associations, LMMs were fitted including a random intercept for school: results did not confirm the association between years of teaching and scores on BPO or BP\\u2019s values, but a positive association that approached significance was found between teaching hours per week and \\u201cEducation of children over time\\u201d (cf. M4, Table 3).\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab3\", \"tab3\"], \"section\": \"Teachers\\u2019 perspective\", \"text\": \"Moreover, paired-sample t-tests highlighted two important differences. Firstly, teachers scored significantly higher (M\\u202f=\\u202f29.7, SD\\u202f=\\u202f5) on STRS Closeness when referring to an SLD pupil, than when considering a WD pupil (M\\u202f=\\u202f26.2, SD\\u202f=\\u202f5.9): t(63)\\u202f=\\u202f4.708, p\\u202f<\\u202f0.001 (95% CI\\u202f=\\u202f2.18\\u20135.39), d\\u202f=\\u202f0.59. Secondly, teachers tended to perceive higher levels of dependency in the relationship when a pupil with SLD was considered (M\\u202f=\\u202f7.6, SD\\u202f=\\u202f2.86) compared to a pupil without a diagnosis but with poor school performance (M\\u202f=\\u202f6.3, SD\\u202f=\\u202f2.23): t(64)\\u202f=\\u202f4.131, p\\u202f<\\u202f0.001 (95% CI\\u202f=\\u202f0.71\\u20132.05), d\\u202f=\\u202f0.51. To verify whether these dimensions of the pupil-teacher relationship varied depending on the student\\u2019s diagnostic status (hypothesis A), two LMMs were conducted: one for the Closeness subscale and one for Dependency. In both models, teachers and schools were treated as random factors to account for the nested structure of the data. Results showed that Closeness scores were significantly higher when teachers described their relationship with a pupil with an SLD compared to one without a diagnosis (cf. M5, Table 3). A similar pattern was found for the Dependency subscale, where higher scores emerged in the case of SLD pupils (cf. M6, Table 3).\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab4\"], \"section\": \"Pupils\\u2019 perspective\", \"text\": \"In accordance with our first objective (hypotheses A and B), we examined children\\u2019s perceptions of the teacher-pupil relationship and potential differences based on gender, age group, and diagnostic status. Table 4 presents the distributions of children\\u2019s responses on each of the three STRQ subscales, together with internal consistency coefficients (Cronbach\\u2019s \\u03b1, r.drop), which were satisfactory across all three STRQ subscales.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab5\"], \"section\": \"Pupils\\u2019 perspective\", \"text\": \"Distribution data by SLD/WD status and age group (younger/older) were not reported, as no differences were observed across these subgroups. Nonetheless, differences in scores were visible between males and females, considering the whole sample and the younger children subgroup (cf. Table 5).\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab6\", \"tab6\", \"tab6\"], \"section\": \"Pupils\\u2019 perspective\", \"text\": \"To further examine whether children\\u2019s responses to STRQ varied as a function of their gender, three LMMs were estimated including random intercepts for teacher and school. No significant main effect of gender was found on the three factors of STRQ, namely Closeness (cf. M1, Table 6), Dissatisfaction (cf. M2, Table 6), and Bond with School (cf. M3, Table 6). Although gender did not emerge as a significant predictor on any STRQ dimension, the analyses provided valuable insights into the variance structure of the data. Children\\u2019s scores on the Dissatisfaction subscale appear to be influenced by contextual factors: the high conditional R2 (96.7%) and moderate clustering at the school (ICC\\u202f=\\u202f0.69) and teacher levels (ICC\\u202f=\\u202f0.28) suggest that children\\u2019s dissatisfaction reflects the environment they experience rather than their gender. For completeness, given the pattern found for teachers regarding weekly hours, we also tested whether pupils\\u2019 STRQ scores were linked to the time the teacher reported spending with the specific child, but no significant correlations emerged, either for SLD or WD pupils. This may be due to the fact that children completed the STRQ with reference to their entire group of teachers, so this analysis might prove useful in future studies with individually matched teacher-pupil data.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab7\"], \"section\": \"Associations between teachers\\u2019 and pupils\\u2019 perspective on their relationship\", \"text\": \"STRQ Dissatisfaction (SLD pupils) was positively correlated with STRS Conflict (SLD pupils) r\\u202f=\\u202f0.4, p\\u202f=\\u202f0.012, and with STRS Dependency (SLD pupils) r\\u202f=\\u202f0.33, p\\u202f=\\u202f0.025. To account for clustering by school, a LMM was fitted including a random intercept for school. ICC and R2 values are reported collectively, as the predictors were tested within a single model. A random intercept for teacher was not included, as each teacher contributed only one data point for the pupil with SLD, making it impossible to estimate teacher-level variance. Results showed that STRS Conflict significantly predicted higher levels of STRQ Dissatisfaction, while STRQ Dependency did not (cf. Table 7). These results should be interpreted with caution, as they reflect associations between aggregated teacher and student perceptions across schools, rather than matched dyadic data. Nonetheless, the conditional R2 of 63.1% and the ICC of 0.48 suggest that a considerable portion of variance in pupils\\u2019 dissatisfaction is linked to school-level factors, highlighting the role of context even when individual predictors are not significant.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab8\"], \"section\": \"Teachers\\u2019 beliefs on SLD\", \"text\": \"Table 8 summarizes the descriptive analyses conducted on the results of the RADSA instrument, providing an initial overview of teachers\\u2019 attitudes and beliefs on SLD diagnosis. In pursuit of the second objective, aimed at deepening the understanding of teacher\\u2013pupil relationship quality by relating it \\u2013 among other factors \\u2013 to teachers\\u2019 beliefs about SLD diagnosis, we proceeded to explore the RADSA dimensions and their associations with teacher-related variables.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab9\", \"tab9\", \"tab9\", \"tab9\"], \"section\": \"Teachers\\u2019 beliefs on SLD\", \"text\": \"To test whether these correlations remained when accounting for school-level clustering, four LMMs were estimated including random intercepts for school. Results confirmed significant effects of teacher age on \\u201cAttention to the class emotions and individual needs\\u201d (cf. M1, Table 9) and \\u201cRoles and information\\u201d (cf. M2, Table 9). In addition, years of teaching predicted higher scores on \\u201cFairness of evaluation\\u201d (cf. M3, Table 9) and lower scores on \\u201cDiagnosis as alibi\\u201d (cf. M4, Table 9). These findings also reveal that the explained variance is not negligible, particularly for \\u201cRoles and information\\u201d (R2c\\u202f=\\u202f37.9%), where school-level differences (ICC\\u202f=\\u202f0.33) play a significant role. This suggests that teacher-related variables interact with school contexts in shaping these specific beliefs about SLD diagnosis.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab10\"], \"section\": \"Associations of teachers\\u2019 beliefs on SLD with the educational stance and practices\", \"text\": \"Table 10 illustrates descriptive statistics of the Adultcentrism paradigm and the Bleak Pedagogy educational stance in the present sample. To address hypothesis C, we investigated how adult-centered tendencies and authoritarian educational values and practices align with specific RADSA subscales.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab11\"], \"section\": \"Associations of teachers\\u2019 beliefs on SLD with the educational stance and practices\", \"text\": \"The correlations observed between the RADSA instrument, the ADT scale, and BPO scale, were informative (cf. Table 11). Both Adultcentrism and Bleak Pedagogy positively correlated with two subscales of RADSA section concerning the increase of SLD diagnoses (\\u201cSystem-level causes\\u201d and \\u201cHow children are raised nowadays\\u201d).\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab12\"], \"section\": \"Associations of teachers\\u2019 beliefs on SLD with the educational stance and practices\", \"text\": \"In order to test more rigorously the associations between RADSA dimensions and the constructs of Adultcentrism and Bleak Pedagogy, a series of LMMs were run (cf. Table 12): results of these analyses confirmed the pattern observed in the bivariate correlations. The ICC and R2 values further indicate that, while the associations between Adultcentrism, Bleak Pedagogy, and certain RADSA dimensions are mainly driven by individual teacher beliefs (low ICC and R2c), some dimensions appear to be more sensitive to school-level factors, as suggested by higher conditional R2 values.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab13\", \"tab13\", \"tab13\", \"tab13\", \"tab13\"], \"section\": \"Teachers\\u2019 beliefs on SLD and the complexity of the teacher-student relationship\", \"text\": \"Results showed that teachers\\u2019 agreement with \\u201cDiagnosis usefulness\\u201d dimension significantly predicted STRQ Dissatisfaction scores in SLD pupils (cf. M1, Table 13). Children\\u2019s Dissatisfaction was also positively predicted by teachers\\u2019 emphasis on the \\u201cRoles and Information\\u201d dimension (cf. M2, Table 13). Regarding Bleak Pedagogy, higher teacher endorsement of \\u201cValues of Bleak Pedagogy\\u201d predicted lower STRQ Closeness (cf. M3, Table 13) and weaker Bond with School in SLD pupils (cf. M4, Table 13). Similarly, higher Adultcentrism association with lower STRQ Bond with School scores was confirmed (cf. M5, Table 13).\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab13\", \"tab13\", \"tab13\"], \"section\": \"Teachers\\u2019 beliefs on SLD and the complexity of the teacher-student relationship\", \"text\": \"Teachers\\u2019 perception of the relationship quality with SLD pupils was also affected by their educational stance and practice: higher scores on \\u201cMethods of Bleak Pedagogy\\u201d were associated with lower perceived STRS Closeness (cf. M6, Table 13), but agreement with \\u201cValues of Bleak Pedagogy\\u201d predicted higher scores on Closeness perceived in the relationship (cf. M7, Table 13). Interestingly, while BP\\u2019s values were associated with higher perceived closeness, this pattern diverged from what was found for BP\\u2019s methods. This prompted us to focus on BP\\u2019s second factor (\\u201cEducation of children over time\\u201d), which also reflects aspects of educational practice\\u2014although framed in temporal terms. Among all variables, the only one directly linked to both this second factor and pupils themselves was the weekly hours spent with them. This observation guided our further analysis: we tested whether agreement with BP\\u2019s methods would predict higher scores on \\u201cEducation of children over time\\u201d (a nostalgic attitude towards old-fashioned methods), assuming that the way teachers manage time with pupils would reflect a consistent pattern shaped by those methods. The LMM confirmed this hypothesis of association: a higher agreement with BP\\u2019s methods predicted a more nostalgic attitude towards those same methods (cf. M8, Table 13). However, the mediation analysis testing the role of weekly hours as a mediator between BP\\u2019s methods and \\u201cEducation of children over time\\u201d did not yield significant results (ACME\\u202f=\\u202f\\u22120.005, p\\u202f=\\u202f0.86), while the direct effect of BP\\u2019s methods remained significant (ADE\\u202f=\\u202f0.421, p\\u202f=\\u202f0.001).\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"tab13\", \"tab13\", \"tab13\"], \"section\": \"Teachers\\u2019 beliefs on SLD and the complexity of the teacher-student relationship\", \"text\": \"Finally, higher Adultcentrism was confirmed as a predictor of higher STRS Dependency considering both SLD and WD pupils (cf. M9, Table 13). These results should be interpreted within the framework of partially overlapping samples, as they illustrate meaningful associations between beliefs and perceptions at the group level, without assuming matched identities between individual teachers and students. Taken together, the relatively high school-level ICCs and R2c values observed for STRQ models (up to 65.8%)\\u2014compared to the minimal clustering in STRS\\u2014may suggest that pupils\\u2019 perceptions of relational quality are more context-dependent, whereas teachers\\u2019 self-perceptions appear comparatively stable. The recurring associations with Bleak Pedagogy-related predictors (M3, M4, M6; Table 13) and the potential influence of Adultcentrism (M5, M9; Table 13) could indicate that authoritarian educational stances and adult-centered attitudes play a role in shaping both relationship quality and pupils\\u2019 bonds with the school context.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref2\", \"ref65\", \"ref112\", \"ref44\", \"ref71\", \"ref89\", \"ref71\", \"ref112\", \"ref112\"], \"section\": \"Discussion\", \"text\": \"This study investigated teachers\\u2019 perceptions of the quality of their relationships with pupils diagnosed with SLD compared to low achievers without diagnosis, by considering a set of teacher-related factors (teaching experience and weekly working hours) and pupil-related characteristics (age and gender). Furthermore, we examined how these elements interact with broader constructs, such as labeling effects and educational stance, to provide a deeper understanding of the dynamics underlying teacher-student relationships. The quality of such relationship appears to be partly influenced by the presence of an SLD diagnosis, since both the perceived levels of Closeness and Dependency were significantly higher when teachers described the relationship with a pupil with an SLD, rather than with a pupil without a diagnosis but identified by the teacher as having poor school performance was taken into consideration. Previous studies have mainly compared pupils with and without an SLD diagnosis, focusing on differences in teacher-student relationships (e.g., Al-Yagon and Mikulincer, 2004; Murray and Greenberg, 2000; Zee et al., 2020) and exploring teachers\\u2019 implicit and explicit attitudes (Hornstra et al., 2010). To our knowledge, very few studies (e.g., Pasta et al., 2013) have directly compared pupils with a formal SLD diagnosis to low achievers without diagnosis. By adopting this finer distinction, our study contributes a novel perspective that helps disentangle the effects of diagnostic labeling from those related solely to academic performance. Our results align with both the primary expectations of the present research and the findings of previous studies. In fact, the STRS Closeness dimension is considered the most representative of a positive relationship (Sabol and Pianta, 2012), and it has been found that the presence and type of diagnosis can be a factor that may influence it (Pasta et al., 2013; Zee et al., 2020). However, this comparison should be interpreted with caution, as the group of WD pupils may include heterogeneous profiles, and the definition of low achievement was based on teachers\\u2019 subjective assessments rather than standardized criteria. Therefore, it is important to note that the comparison group may have included pupils with diverse and possibly undiagnosed difficulties, which could intervene in challenging the quality of the pupil-teacher relationship, thus diminishing the perception of closeness in a similar way to what was found for pupils with difficulties that had a greater impact at the behavioral level (Zee et al., 2020).\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref63\", \"ref27\"], \"section\": \"Discussion\", \"text\": \"Instead, the significant effect of weekly hours spent with the WD pupil was confirmed as a predictor of higher levels of Dependency (STRS) and Closeness (STRS) perceived by the teacher. This pattern suggests that increased contact time may be associated with both enhanced relational involvement and a stronger perception of the pupil\\u2019s reliance on the teacher. However, the moderation effect of Bleak Pedagogy\\u2019s methods indicates that this interpretation is not uniform: the more teachers endorsed authoritarian educational styles, the less they interpreted increased time as a sign of Dependency. This moderation effect may reflect not a lack of relational engagement, but rather a specific normative view of educational responsibility, where closeness is valued and cultivated, yet interpreted through a lens of directive guidance and control \\u2013 a stance that, if accepted by the other member of the relationship (i.e., the child), may indeed foster dependency without necessarily generating Dissatisfaction, as also suggested by the absence of significant associations with this STRQ subscale. This interpretation is consistent with Maturana and D\\u00e1vila\\u2019s (2006) claim that adult control in the educational relationship generates dependence in children. And, as suggested by Florio et al. (2023), teachers who endorse Bleak Pedagogy\\u2019s methods may be genuinely convinced that increased involvement and directivity serve the child\\u2019s best interests. In this view, spending more time with a WD pupil does not necessarily imply fostering autonomy or shared agency, but rather ensuring protection, order, and \\u201cproper\\u201d development. In sum, such time investment may enhance relational closeness while simultaneously fostering a perception of the pupil as dependent \\u2013 unless such dependency is reframed, by some degree of adherence to authoritarian educational stance, as appropriate receptivity to adult guidance. These interpretations remain provisional, as the mechanisms linking contact time with the pupil, perceived dependency, and educational stance may involve bidirectional dynamics or contextual factors beyond the scope of this study. Longitudinal research would be valuable to examine how these relationships evolve over time and to disentangle the direction of these effects.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref113\"], \"section\": \"Discussion\", \"text\": \"Regarding years of experience, previous research has shown that teaching experience can influence teacher-student relationship perceptions (Zee and Koomen, 2017). In our study, however, no significant association was found between teaching experience and the quality of teacher-student relationships, suggesting that this link may depend on contextual factors or other variables not captured in the present design.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref112\", \"ref71\", \"ref71\", \"ref31\", \"ref110\"], \"section\": \"Discussion\", \"text\": \"The finding that teachers report greater Closeness and Dependency in their relationships with SLD pupils compared to peers without a diagnosis, albeit with equally poor school performance, suggests that the diagnostic label might activate a more empathic, protective, or affiliative stance. The result concerning Closeness appears consistent with the literature emphasizing the positive impact of diagnosis as a relational \\u201cframework\\u201d (Zee et al., 2020; Pasta et al., 2013), helping teachers to interpret students\\u2019 difficulties through a lens that reduces blame and facilitates supportive behaviors. Regarding Dependency, the higher scores in relation to SLD pupils might reflect teachers\\u2019 increased willingness to provide help and guidance to students officially recognized as struggling. However, this interpretation warrants caution: over-involvement and perceptions of excessive need may also emerge, especially in classroom environments where the diagnosis becomes a cue for permanent vulnerability rather than a signal of support potential (cf. Pasta et al., 2013). In fact, prior work suggests that a formal diagnosis may alter teachers\\u2019 attributions, leading them to view academic difficulties as less changeable and more reflective of the pupil\\u2019s inherent limitations which can ultimately reinforce low expectations (Frohlich et al., 2020; Woodcock and Vialle, 2011). It has to be underlined that the cross-sectional nature of our data does not allow for determining such temporal direction of the associations we found.\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref70\", \"ref38\", \"ref102\"], \"section\": \"Discussion\", \"text\": \"Some noteworthy findings emerged from the exploration of associations between pupils\\u2019 and teachers\\u2019 representations of their relationship. Before delving into these results, it is worth recalling that internal consistency analyses supported the reliability of STRQ subscales across both younger and older children. All three subscales showed acceptable to good internal consistency, as indicated by Cronbach\\u2019s \\u03b1 and by mean corrected item-total correlation (r.drop), in line with recommended thresholds (Nunnally and Bernstein, 1994; Hair et al., 2019). These findings strengthen the reliability of the instrument even among younger pupils, despite the original validation of STRQ being conducted primarily on 4th and 5th grade students (Tonci et al., 2012).\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref33\", \"ref39\", \"ref45\", \"ref101\", \"ref83\", \"ref112\", \"ref76\"], \"section\": \"Discussion\", \"text\": \"Correlational analyses between children\\u2019s and teachers\\u2019 responses revealed that when teachers express that they perceive more Conflict and Dependency in their relationship with SLD pupils, the same pupils report more Dissatisfaction in their relationship with the teachers. This relationship was also confirmed by LMM analysis, showing that especially teacher-perceived conflict seems to play a predictive role toward perceived Dissatisfaction among SLD pupils. These findings align with the extensive literature emphasizing the protective role of positive teacher-student relationships (Garc\\u00eda-Rodr\\u00edguez et al., 2023; Haldimann et al., 2023; Jederlund and von Rosen, 2023; Thornberg et al., 2022), particularly for pupils with diagnosed vulnerabilities (Re et al., 2021; Zee et al., 2020). They also support a systemic perspective, which conceptualizes both student and teacher as active constructors of the relationship quality (Pianta, 2001a).\"}, {\"pmc\": \"PMC12546017\", \"pmid\": \"\", \"reference_ids\": [\"ref112\", \"ref53\"], \"section\": \"Discussion\", \"text\": \"Differently from a previous study that explored teacher-pupil relationship quality using a multi-informant approach (Zee et al., 2020), no significant differences emerged in our sample between SLD and WD pupils in children\\u2019s STRQ scores. This result should be considered

Metadata

"{}"

PMC Articles

When learning with SLD is a hard challenge: a complex perspective on teacher-student relationships in primary school

Abstract

Full Text

Sections

Metadata