Anxiety and Teacher-Student Relationships in Secondary School: A Systematic Literature Review

Anxiety is the most prevalent mental health disorder experienced by adolescents, with approximately 7% of young people aged between 12 to 17 years old experiencing clinically significant levels of anxiety [1]. In addition to being highly prevalent, anxiety disorders are associated with significant impairments to adolescents’ self-esteem, social competence, peer relationships, and academic performance [2], Kashani and Orvaschel [3, 4], and can increase an adolescent’s risk of developing other psychological disorders in adulthood [5]. Taken together, the high prevalence of anxiety disorders in adolescents and the associated negative consequences emphasises the need for a better understanding of the factors in adolescents’ environment which may cause and maintain anxiety during this developmental stage. To date, research on anxiety in children and adolescents has predominantly focused on the identification of risk factors in individuals (e.g., genetics, temperament, attention bias) and in the family environment (e.g., overprotective parenting and insecure attachment, [6]. Yet, students spend a significant amount of time within their school environment and thus teachers may also play an important role in the development and maintenance of anxiety beyond that of parents through positive and negative interactions with students that might increase, decrease or maintain anxiety. For example research indicates teachers use of fear appeals to motivate students to improve academic outcomes can be associated with increased anxiety Belcher et al. [7].

In addition to influencing academic outcomes, research on teacher-student relationships (TSRs) has found these relationships are a significant contributor to student mental health outcomes overall [8, 9]. In particular, TSRs that are perceived as being supportive have been associated with improved student wellbeing [10], resilience [11], self-esteem, and social skills [12, 13], and negatively associated with distress, suicidal behaviours, and depression [14], Udheim 2005, [15, 16]. Moreover, research has found that TSRs tend to become less positive as students grow older [17], which is concerning as adolescents have also been found to be more sensitive to the degree of support they receive from their teachers [18].

While the conceptual framework for TSRs has varied across the literature and has roots in education and psychology, research on the role of quality TSRs with regards to student mental health outcomes has largely been motivated by an extended attachment framework [19, 20]. According to attachment theory, early relationships between children and their parents are central to a child’s development and can provide a secure base for psychological and social development [19, 21]. Applying this theory to the school environment, teachers are believed to buffer students from developmental risks by providing them with a sense of security, and helping them relate to peers, regulate emotions, and use positive coping strategies [22]. Thus, Pianta [23] defined relationships between teachers and students as characterised by degrees of dependence, closeness, and conflict, while Murray [24] described them as characterised by warmth, trust, expectations, and involvement.

For adolescents, several conceptual models have been used to describe TSRs including attachment, developmental systems, social-motivational, and interpersonal theory [9, 25, 26]. Central to each of these theories is the importance of a sense of relatedness and involvement which includes teacher’s expression of empathy, warmth, and emotional support for the adolescent [20]. In addition, teachers are proposed to influence student motivation by increasing student’s sense of competence and autonomy [9, 25]. Thus, the term ‘positive teacher-student relationship’ is often generally used to describe when adolescents perceive that these teacher qualities are expressed towards them McHugh et al. [27]. On the other hand, negative TSRs are characterised by higher levels of conflict between teachers and students and a lack of emotional security [26].

Regarding TSRs and student anxiety specifically, although research examining this relationship in adolescents is scarce, studies based on primary-school aged children have linked positive TSRs with reduced student anxiety e.g., [28, 29], and greater teacher-student conflict with increased anxiety [30, 31]. In line with predictions from the attachment framework, Semeraro et al. [32] found that teachers who expressed more warmth, responsiveness, trust, and accessibility were better able to support students in regulating their anxiety. Further, Murray and Murray [33] found that close TSRs were associated with lower anxiety symptoms, while dependent relationships were associated with increased anxiety. Of the limited research focused on adolescents, similar findings were reported on the association between dependent teacher relationships and higher test anxiety [34]. However, findings regarding the effects of positive TSRs on anxiety are inconsistent. Keçici [35] found that positive TSRs (perceived teacher social support) were associated with reduced anxiety in female adolescents only, while Fredrick et al. [36] found a similar effect for perceived teacher support but in males. On the contrary, results from other studies found no significant effect for overall positive TSR but did for perceptions as teachers as positive motivators [34].

As these studies are predominantly cross-sectional, one of the main limitations to this research is the inability to demonstrate causation or the direction of these effects. In other words, it is equally possible that more anxious children elicit more dependent or conflictual relationships, or that students who are low in anxiety may promote more positive relationships with their teachers. In fact, theoretical models of anxiety in children and adolescents propose a bi-directional relationship between the child’s anxiety and interactions with significant others [6]. These reciprocal effects have been demonstrated in studies exploring parent–child relationships whereby anxious children are likely to elicit overprotection from their parents and, in turn, overprotective parenting is likely to lead to further anxiety [20]. As adolescence is characterised by less reliance on parents, more interactions with teachers in the secondary school environment, and an increased desire for independence and autonomy [37, 38], it is expected that greater autonomy support, rather than dependence, would lead to reduced anxiety. Teacher autonomy support is characterised by behaviours such as promoting self-directed learning and providing more opportunity for choice [39]. In a longitudinal study by Yu et al. [40], teacher autonomy support was found to reduce anxiety and depression in adolescents over time. Other studies have also reported similar findings in adolescents (e.g., [14, 41, 42], suggesting that there is considerable empirical support for the protective effect of teacher autonomy support on adolescent anxiety.

See Fig. 1 for the PRISMA flowchart that depicts the results for identifying, screening, and including studies. The database search returned a total of 4963 articles, 1657 duplicates were removed, leaving 3306 articles for eligibility checking at title and abstract stage. The full text of 146 articles were reviewed, of which a total of 26 articles were eligible for the current systematic literature review.

This systematic literature review aimed to examine the relationship between TSRs and anxiety in adolescents attending secondary school. Based on previous literature, theories of TSRs, and models of anxiety, it was hypothesised that the relationship would be reciprocal, that is that factors in the TSR would influence adolescent anxiety in secondary school, and in turn, adolescent anxiety would influence factors in the TSR. Most studies either assessed the quality of the TSR using broad measures that assessed a range of relationship domains and characteristics or focused on one specific characteristic, and whilst there was great variability in the characterises of the TSR measured in general, the research suggested that TSRs that were perceived as positive and supportive were associated with reduced anxiety in adolescents attending secondary school. This is consistent with prior research based on primary-school aged children [28, 29]. Across the studies, in general lower anxiety was associated with a broad range of positive and supportive TSR characteristics such as sensitivity, care, affection, and warmth, and when students perceived their teachers to be helping, friendly and understanding. Conversely, higher anxiety levels were reported in students who perceived higher teacher harassment and conflict, and when teachers were perceived as unfair and strict. These findings are consistent with attachment theory and prior research [71]. Emotionally warm and supportive relationships between teachers and students are thought to provide the student with a sense of relatedness and security within the school environment, thus reducing symptoms of anxiety. In contrast, the psychological need for relatedness is unlikely to be fulfilled in TSRs that are characterised by a high level of conflict, increasing their risk for anxiety. Similarly, this finding also fits with research on characteristics of the parent–child relationship that have been found to be associated with reduced child anxiety, namely, warmth and low aversiveness [72].

It is important to note that 23 of the 26 included studies used a cross-sectional design, and therefore while there is evidence that the nature of TSR is associated with student anxiety, interpretations about causation cannot be made. Findings from three longitudinal studies provide some insights as to possible causal explanations. All three studies found some evidence that the nature of the TSR was associated with subsequent anxiety levels in adolescents attending secondary school, but due to differences in methodology it is unclear what aspect of the TSR might be particularly important, and also whether the relationship was bidirectional. Whilst the study by Giota and Gustafsson [50] was of lower quality, the studies by Weymouth and Buehler [51] and Aldrup et al. [49] were higher quality, with the later being conducted in a very large sample, suggesting that the overall finding of higher quality TSR leading to lower student anxiety over time is likely a robust finding. More research is needed to understand the causal role of TSRs, and if there are specific TSR characteristics that most strongly influence anxiety symptoms in students over time, as this could have impacts on how teachers manage relationships with students over time and reduce future student anxiety.

An interesting finding from this review was that teacher support might have a divergent association with student anxiety, depending on whether the support promotes greater student dependence or autonomy. Compared to ‘teacher support’ and ‘teacher autonomy support’ which were reported to be associated with lower anxiety in all studies that examined these factors, the need for greater ‘teacher motivational support’ was associated with increased test anxiety and feelings of helplessness [47, 65]. These findings may suggest that when teachers are perceived as motivators, this can induce feelings of expectations and pressure to perform (i.e., increased test anxiety) while also increasing dependency on the TSR for support due to decreased self-efficacy. Hence, support from teachers may not only be characterised by a personal relationship but also by expectations of educational results. Some recent research in senior students aligns with this finding. In a large sample, student anxiety was greater in students who reported lower self-efficacy and more frequent use of fear appeals by their teachers in which their teachers would emphasise the negative consequences of poor examination performance in an attempt to motivate students to study [7]. Thus, consistent with theoretical models of anxiety which propose that overprotective parenting leads to further anxiety [20], TSRs that are characterised with greater involvement and dependence may increase adolescent anxiety in secondary school settings, whereas teacher autonomy support appears to have a protective effect. However, as noted above, these findings are based on cross-sectional studies and therefore conclusions regarding causal directions cannot be made. Future research should examine the impact of teacher motivational support on student anxiety over time for further understand this association.

Studies included in this review also revealed some effects of gender on anxiety levels. For example, nine of the studies reported that anxiety levels were found to be higher in girls than boys replicating other studies [73]. These results are likely reflective of prevalence rates given that girls are often recognised as being more likely to develop symptoms of anxiety than boys [6]. Furthermore, although it has been proposed in some studies that girls report closer relationships with their teachers than boys and are more reliant on them for social support [74, 75], overall results from the current review did not indicate sex differences in terms of TSRs suggesting that boys and girls perceive similar levels of support from teachers. Similarly, with the exception of two papers [35, 56], there were no studies within the review that found gender differences when examining the relationship between anxiety and TSRs. While it is unknown why two of the studies found a significant association between TSRs and anxiety in girls only, possible explanations may relate to the higher proportion of females than male participants within both studies.

Another interesting finding stems from the two papers that included samples from more than one country. In both studies, the perceived quality of TSRs was linked with anxiety levels in the German samples only, with non-significant results found in the American and Canadian samples, respectively [46, 66]. Notably, results from one study suggested that anxiety levels were lower in German students with higher teacher care [66], while the other suggested that test anxiety levels were higher in German students who reported higher quality TSRs [46]. One possible consideration for these different results is that the study by Hoferichter et al. [46] measured test anxiety rather than trait or state anxiety. As previously described, teacher motivational support has been linked with increased test anxiety (e.g., [65]. Thus, these findings could reflect that for some students, increased support from their teacher may induce anxiety more specific to doing well in that subject (i.e., increased test anxiety). Further, the differing results across the German, Canadian and American samples was surprising given they are western cultures that are interested in promoting autonomy. However, there are also notable differences in the schooling system between these countries. For example, unlike in America or Canada, German students are often taught by the same teachers for multiple years, providing a greater opportunity for stronger relationships to develop. As the association between TSRs and anxiety within other American and Canadian samples included in the study was reported to be significant e.g., [51, 61, 63], further research investigating how teacher relationships may differ across cultures is needed to better understand this relationship across different school systems.

This review has several limitations. First, in order to balance the differences in secondary school models between countries and to be inclusive of middle school students, some children aged 11 years of age (pre-adolescent) were also included. Secondly, a meta-analysis was not conducted and so results are based on a narrative synthesis of the literature only. Finally, the systematic search was limited to articles published in English limiting generalisability to non-English speaking countries or regions. These limitations reduce the generalisation of the results. Perhaps more significantly, it is important to consider the challenges of consolidating research in this field due to the wide range of definitions and measurement of TSRs, as reviewed by Phillippo et al. [76], and the overreliance of student-reported perceptions of teacher behaviours. Further the variation in definitions and measures limits our understanding of what particular factors in the TSR are key in influencing student anxiety, and which types of anxiety are influenced by these. Thus, it is important to develop alternative approaches for assessing TSRs as having multiple informants and methods is consistent with best practice in assessment, and could help to reduce the inherent biases related to relying on one informant Murray and Zvoch [77]. Despite these challenges, the evidence across these measures was generally consistent in finding that features of TSR were associated with increased and decreased levels of anxiety in students. Although most of the studies in this review used cross-sectional designs limiting interpretations regarding causality, the three longitudinal studies found similar results and suggest that TSRs may be able to influence student anxiety levels over time, despite differences in study methods and measures. With consideration to the above limitations and challenges in the field, future research may benefit from addressing these by: (1) collecting information from multiple informants (e.g., including additional teacher ratings) to minimise bias and also capture perceptions of both student factors which can influence TSR quality, (2) utilising validated TSR measures which contain items relating to both teacher and student characteristics, (3) conducting longitudinal studies to evaluate the effect of adolescent anxiety and student behaviours (e.g., help-seeking) on TSRs; and (4) controlling for covariates such as gender and cultural background given the reported differences in anxiety levels.

The current systematic review explored the literature on teacher-student relationships and anxiety in adolescents attending secondary school. In general, the findings indicated that TSR relationships that were higher in support, lower in motivational support and lower in conflict and dependence were associated with less student anxiety. Existing research on teacher support has thus far suggested that teachers could improve students’ sense of autonomy by providing more opportunities for choice, promoting independent decision making, minimising use of shame-inducing teaching strategies, and by acknowledging the students’ feelings e.g., [40, 78, 79]. Further research exploring the effects of these strategies on adolescent anxiety in secondary school settings could yield important information to refine recommendations for schools and assist teachers in establishing high quality relationships with their students. As most studies used a cross-sectional design, conclusions regarding causation and whether the relationships between TSR and student anxiety is reciprocal could not be made. However, consistent findings from longitudinal studies indicated that more positive TSR was associated with reduced student anxiety over time, offering promising targets for future interventions. This review therefore highlights a clear need for future studies to incorporate multiple methods and informants to collect data, as well as a greater focus on how adolescent anxiety may impact the TSR bidirectionally. Furthermore, more research on the specific teacher behaviours and approaches that are beneficial in promoting student mental health will better inform recommendations for teachers in secondary schools on how to develop quality relationships with their students.