The relationship between teacher–student relationship and chemistry achievement: a chain mediation model
Li Yanli
*a and
Zhao Dongyabc
aSchool of Education, Central China Normal University, Wuhan, 430079, China. E-mail: liyanli202203@163.com
bShaanxi Normal University, Xian, 710062, China
cHenan Normal University Affiliated High School, Xinxiang, 453007, China
First published on 19th March 2026
Abstract
The teacher–student relationship is a crucial environmental factor influencing students’ academic development; however, the specific mechanisms through which it affects chemistry achievement remain unclear. While existing research affirms the importance of the teacher–student relationship, empirical evidence remains particularly scarce concerning the chain pathway through which it influences students’ interest in chemistry and their academic engagement in the subject, ultimately affecting their achievements in chemistry. This study aimed to develop and test a chain mediation model to examine this pathway. Data were collected from 678 junior and senior high school students, including measures of chemistry achievement, teacher–student relationships, interest in chemistry, and chemistry academic engagement. Structural Equation Modeling (SEM) was employed to analyze the direct and indirect effects. The results indicated that: (1) closeness in the teacher–student relationship had a significant positive direct effect on chemistry achievement. (2) Low conflict in the teacher–student relationship had a significant negative direct effect on chemistry achievement, but could indirectly promote chemistry achievement through students’ interest in chemistry. (3) For both dimensions of the teacher–student relationship (closeness and low conflict), a significant chain mediating effect was identified, following the pathway “teacher–student relationship → interest in chemistry → chemistry academic engagement → chemistry achievement”. These findings suggest that chemistry educators should prioritize fostering close relationships with students, effectively managing conflicts within teacher–student relationships, and ultimately enhancing achievement by cultivating students’ interest and engagement in chemistry.
Introduction
Student academic achievement, defined as an individual's performance in intellectual domains taught in school (Spinath, 2012), is a crucial determinant of both individual and societal success. At the individual level, it is strongly associated with socioeconomic attainment, health, and well-being (Rindermann, 2008; Rimfeld et al., 2018) and serves as a critical gateway to higher education opportunities and career paths, thereby shaping life trajectories (Hübner et al., 2022). At the societal level, the collective educational attainment of a population is a fundamental driver of economic competitiveness and the effective functioning of democratic societies (Gylfason and Zoega, 2003; Spinath, 2012). Consequently, identifying the factors that influence academic achievement and developing methods to enhance it have become central concerns for educational researchers and policymakers worldwide.The teacher–student relationship is established through the process of teaching and learning. It is the most fundamental and important interpersonal relationship in school education (Pianta and Nimetz, 1991; Roorda et al., 2011). When it comes to explaining variance in student achievement, the teacher is the most significant factor, after student characteristics (Hattie, 2009). As a key dimension of this influential teacher factor, the teacher–student relationship has been widely identified as crucial for students’ academic achievement (Longobardi et al., 2016; Roorda et al., 2017). Research indicates that closeness in the teacher–student relationship can foster a supportive learning environment through teacher care, thereby enhancing students’ psychological and behavioral engagement and ultimately positively impacting their academic achievement (Nie and Lau, 2009; Quin, 2017; Sointu et al., 2017). In contrast, conflict in the teacher–student relationship may lead students to avoid interaction, reduce their engagement in learning, and ultimately undermine their academic achievement (Jerome et al., 2009; Berti et al., 2010; Bryce et al., 2019).
The impact of the teacher–student relationship on students’ academic achievement may vary across subjects. Chemistry, as a core STEM discipline (González and Paoloni, 2015), has two notable characteristics in its teaching and learning processes. First, its knowledge system is highly abstract. Students must establish connections among macroscopic phenomena, microscopic principles, and symbolic representations, which often leads them to perceive chemistry as a challenging subject (Treagust and Chittleborough, 2001; Treagust et al., 2018). Second, as an experimental discipline, chemistry learning typically involves laboratory activities that are conducted under the guidance of teacher (Baldwin and Orgill, 2019). These characteristics result in more frequent interactions between teachers and students in chemistry classes, with students relying heavily on teachers’ guidance and feedback. In such interaction-intensive contexts, the quality of teacher–student relationships may be especially influential in chemistry learning. Supportive relationships can help students maintain interest and engagement when facing abstract concepts and experimental challenges, whereas conflictual relationships may hinder students’ sustained involvement in learning activities. Therefore, in chemistry, the influence of the teacher–student relationship on academic achievement may follow unique pathways and mechanisms that warrant special investigation. However, existing research on the relationship between teacher–student interactions and academic achievement primarily focuses on subjects such as mathematics or English (Ma et al., 2018; McKinnon and Blair, 2019; Xu and Qi, 2019; Hajovsky et al., 2020), with empirical studies in chemistry remaining relatively limited. By elucidating the connections between teacher–student relationships and students’ achievement in chemistry, this study aims to increase chemistry teachers’ awareness of the crucial role they play in the students’ chemistry learning process. The findings are intended to offer practical guidance to chemistry teachers, enabling them to better understand and manage their interactions with students in everyday classroom settings.
Furthermore, existing research has demonstrated that teacher–student relationship, interest in chemistry, academic engagement, and student achievement are positively correlated with one another (Köller et al., 2001; Ainley and Ainley, 2011; Lei et al., 2018; Asadzadeh et al., 2019; Liu, 2024). A positive teacher–student relationship can stimulate students’ intrinsic interest in learning (Asadzadeh et al., 2019), which in turn fosters their engagement and ultimately enhances academic achievement (Harackiewicz et al., 2008). However, no research has yet integrated these variables into a single model to examine the interaction pathways among them. To this end, the study proposes and tests a chain mediation model to explore these mechanisms more comprehensively. In this model, “teacher–student relationship” is the independent variable, both (a) “interest in chemistry” and (b) “chemistry academic engagement” are the mediating variables, and “chemistry achievement” is the dependent variable. This study aims to investigate the mediating and chain mediating effects of interest in chemistry and chemistry academic engagement between teacher–student relationship and chemistry achievement. Ultimately, it aims to provide theoretical support and suggest practical intervention strategies to enhance the academic achievement of junior and senior high school students in chemistry, thereby contributing to the effective improvement of the quality of chemistry education.
Review of the literature
Teacher–student relationship and chemistry achievement
Interaction and communication between teachers and students are central to the educational process and have long attracted considerable attention. When assessing the teacher–student relationship, researchers typically adopt one of two perspectives. One approach is from the teacher's point of view, relying on the teacher's perceptions and evaluations of their interactions with students to assess the nature of the teacher–student relationship. Another approach places greater emphasis on students’ own feelings, considering them as the primary focus. Through methods such as questionnaires or interviews, it seeks to understand the nature of the teacher–student relationship they have genuinely experienced, thereby shifting the focus of evaluation from teachers to students. The most commonly used tool for measuring the teacher–student relationship is the Student–Teacher Relationship Scale (STRS), developed by Pianta et al. (1995). This scale reflects the perspective of teachers. Teachers’ perceptions of the teacher–student relationship were measured across three dimensions: closeness, conflict, and overdependence. Subsequent researchers have revised and adapted the scale developed by Pianta, generally adopting the student assessment method, which better reflects the actual teacher–student relationship as perceived by students (Li, 2023; Liu, 2024; Yu et al., 2024). Some studies have directly used the Teacher–Student Relationship Scale in the PISA 2012 Student Questionnaire (Ma et al., 2018; Zhou et al., 2020). Pieratt (2011) classified the teacher–student relationship into two types: positive and negative. The characteristics of a positive teacher–student relationship include frequent interactions, an understanding of students’ abilities and interests, high expectations, care, trust, and emotional bonds, whereas negative teacher–student relationships lack these qualities. Although various studies categorize the teacher–student relationship differently, it can be broadly divided into two fundamental types: closeness in the teacher–student relationship, defined by harmonious and intimate interactions, and conflict in the teacher–student relationship, defined by inconsistency and disharmony in behavior and emotion. Based on the classification above, this study defines the teacher–student relationship between students and chemistry teachers as the connections perceived by students in terms of cognition, emotion, and behavior with chemistry teachers during their daily interactions in chemistry teaching. It is particularly noteworthy that, as an experiment-based discipline, chemistry offers greater opportunities for face-to-face interaction and communication during the process of teachers guiding students through experimental procedures. This distinctive teaching characteristic makes the interaction between teachers and students in chemistry classes more practical and contextually relevant.A review of previous literature reveals that most studies have concluded that the teacher–student relationship is significantly correlated with students’ academic achievement. Most researchers believe that a positive teacher–student relationship can significantly predict strong academic achievement, and this finding is consistent across various countries and at all stages of student development. For instance, Li et al. (2022), using empirical data from the Missouri Educator Effectiveness Evaluation System in the United States, revealed that positive teacher–student relationships enhance teaching quality. Košir and Tement (2014) conducted a one-year longitudinal study involving 816 Slovenian students across three age groups: late childhood, early adolescence, and middle adolescence. They analyzed the data using a cross-lagged panel model and found a significant bidirectional relationship between teachers’ acceptance of students and the students’ academic achievement. Teacher acceptance at the beginning of the academic year not only predicted improvements in academic achievement by the end of the year, but academic achievement at the start of the year also predicted an increase in the level of teacher acceptance by the end of the year. Sointu et al. (2017) conducted a three-year longitudinal study involving students in grades 5 to 7 from 46 schools in eastern Finland. Using a multi-informant design – which included students’ self-ratings of the teacher–student relationship, parents’ assessments of behavioral and emotional strengths, and teachers’ reports of academic achievement – and a cross-lagged panel model, they found that the teacher–student relationship was significantly correlated with students’ average credit grade point across all subjects. In other words, students who had better relationships with their teachers tended to perform better in Grade 7. Roorda et al. (2017) conducted a meta-analysis of 189 studies and found that the total effect size of a positive teacher–student relationship on students’ academic achievement was 0.14. This indicates that for every 1 standard deviation increase in the quality of the teacher–student relationship, students’ academic performance increases by an average of 0.14 standard deviations. This suggests that the relationship between teachers and students can, to some extent, influence students’ academic achievement. Most existing research on the teacher–student relationship focuses on the impact of the relationships established between head teachers and students on students’ comprehensive achievement across multiple subjects. A few studies adopting a disciplinary perspective primarily focus on subjects such as English and mathematics, exploring the relationship between subject teachers and students’ academic achievement in these subjects. In contrast, from the perspective of the discipline of chemistry, research specifically investigating how the relationship between chemistry teachers and students influences their achievement in chemistry is rather scarce.
The potential mediating effect of interest in chemistry
Hidi and Renninger (2006) define interest as a motivational variable refers to the psychological state of engaging or the predisposition to reengage with particular classes of objects, events, or ideas over time. On this basis, researchers typically classify learning interest into two types: situational interest, which refers to a temporary state of interest triggered by specific environmental conditions; and individual interest, which denotes a relatively stable internal tendency gradually developed through the accumulation of knowledge and emotional investment (Schiefele, 1991; Hidi and Renninger, 2006). Because personal interests tend to be more stable and sustainable, and existing studies have demonstrated a closer correlation between personal interest and the willingness to explore, as well as the perseverance shown by students during the learning process (Hidi and Renninger, 2006; Rotgans and Schmidt, 2017), therefore this study primarily focuses on students’ personal interest in chemistry learning.Students’ learning interest is a crucial factor influencing their learning behavior and achievement (Renninger and Hidi, 2019). Based on a large-scale questionnaire survey of 1095 students from 10 schools in Portugal, Abrantes et al. (2007) found that the quality of interaction between teachers and students not only directly enhances students’ positive emotions towards teaching methods, but also indirectly optimizes their interest in learning and achievement through multiple mediating pathways, ultimately significantly improving their perceptual learning levels. Stroet et al. (2013) systematically analyzed 71 relevant studies and categorized teachers’ demand-supportive teaching into three dimensions: autonomous support, structural support, and emotional investment. Their in-depth descriptive analysis revealed that teachers’ emotional care for students and positive teacher–student relationships were among the strongest predictors of students’ academic achievement. Teachers’ teaching behaviors can indirectly affect students’ academic achievement by influencing their learning interests (Ryan and Deci, 2020). Ahn et al. (2021) found that the autonomous support provided by teachers effectively meets students’ basic psychological needs, thereby enhancing their autonomous motivation and ultimately exerting a significant positive impact on their academic achievements in mathematics. Therefore, it can be expected that interest in chemistry plays a mediating role in the relationship between teacher–student relationship and students’ chemistry achievement.
The potential mediating effect of chemistry academic engagement
Schaufeli et al. (2002a, 2002b) believes that academic engagement is a continuous, active, and fulfilling state experienced by students during the learning process, encompassing three specific dimensions. The dimension of vigor refers to a student's abundant energy during the learning process, their willingness to exert effort, and their perseverance in overcoming difficulties encountered throughout the process. Dedication refers to the sense of value students derive from their studies, which enables them to face various learning challenges with a positive attitude. Absorption involves concentrating one's attention and devoting oneself wholeheartedly to learning, thereby experiencing the joy that learning brings. Fredricks et al. (2004) identified academic engagement as comprising behavioral, emotional, and cognitive components. Therefore, academic engagement can be understood as the overall level of student participation in learning activities. In this study, chemistry academic engagement refers to the extent to which students are willing to consistently exert effort, demonstrate strong enthusiasm for learning, and achieve a state of focused and immersive engagement during the chemistry learning process. Skinner et al. (2009) found that teachers’ actions and students’ perceptions of those actions can influence students’ academic engagement. When students feel supported by their teachers, they are more likely to increase their interest in learning and experience enjoyment at school. The research conducted by Young and Bruce (2011) indicates that teacher support, as perceived by students, can directly enhance their effort and participation in the learning process. Therefore, this study holds that a good teacher–student relationship can provide students with the essential psychological security and help them develop a positive identification with the discipline of chemistry. This sense of security and identity makes students more confident in facing learning challenges, more proactive in participating in inquiry activities, and more deeply engaged in experiencing the joy of learning chemistry, thereby effectively enhancing their level of learning engagement.The input–output model posits that students’ learning input is a prerequisite for achieving good academic achievement, with both the quality and quantity of this input determining academic achievement. Within a certain range, academic engagement, as an “input” influences the “output” of academic achievement. The more students engage, the more likely their academic achievement will improve (Roorda et al., 2017). Furthermore, studies have shown that students with higher levels of academic engagement tend to master effective learning strategies and demonstrate stronger self-control. These traits serve as important foundations for achieving excellent academic achievement (Luo et al., 2023). According to a meta-analysis of 69 studies conducted by Lei et al. (2018), there is a consistent positive correlation between student engagement and academic achievement. Among the different types of engagement, behavioral engagement had the strongest influence (r = 0.35), while emotional engagement had the weakest influence (r = 0.22). Sankar and Benjamin (2024) performed a regression analysis on data from 476 high school students. After controlling for initial ability differences, such as pre-test scores, they found that students’ academic engagement significantly and positively predicted their chemistry academic achievement. In conclusion, this study suggests that a high level of academic engagement substantially promotes improvements in students’ chemistry achievement. Based on this, the current study postulated that chemistry academic engagement would mediate the correlations of teacher–student relationship with students’ chemistry achievement.
The chain mediating effect of (a) interest in chemistry and (b) chemistry academic engagement
Regarding the intrinsic connection between (a) interest in chemistry and (b) chemistry academic engagement, this relationship can be inferred from the reliability and validity tests conducted by Mazer (2013) on the student interest scale and the student engagement scale. This study confirmed the stability and reliability of the scale factor structure through confirmatory factor analysis (CFA). Furthermore, the empirical data demonstrated a significant positive correlation between students’ learning interest and their academic engagement. Interest can stimulate students’ positive emotions and encourage active participation in learning activities (Schiefele, 1991). Therefore, this study posits a close correlation between (a) interest in chemistry and (b) chemistry academic engagement. For instance, the research by Hui et al. (2019) in the blended learning environments indicates that optimizing instructional design and guiding learning behaviors can effectively stimulate students’ situational interest. This situational interest is a key driver for enhancing academic engagement and ultimately improving academic achievement. Based on Pekrun et al. (2002) control-value theory, Ainley and Ainley (2011) conducted an empirical analysis using large-scale international data from PISA 2006. Their research demonstrates that students’ enjoyment of science subjects effectively stimulates and sustains their interest in science, thereby motivating continuous academic engagement and ultimately leading to higher academic achievement. Therefore, this study posits that (a) interest in chemistry and (b) chemistry academic engagement serve as chain mediators in the association between the teacher–student relationship and students’ chemistry achievement.Research questions and hypotheses
To investigate the relationships among the teacher–student relationship, interest in chemistry, chemistry academic engagement, and chemistry achievement, a questionnaire was used to survey these variables. Accordingly, the study addressed the following research questions:(1) In the context of chemistry learning, what is the nature of the relationship between the teacher–student relationship (including aspects of closeness and low conflict) and students’ chemistry achievement?
(2) Does interest in chemistry mediate the relationship between the teacher–student relationship (closeness or low conflict) and students’ chemistry achievement?
(3) Does chemistry academic engagement mediate the relationship between the teacher–student relationship (closeness or low conflict) and students’ chemistry achievement?
(4) Do interest in chemistry and chemistry academic engagement function as sequential mediators between the teacher–student relationship (closeness or low conflict) and students’ chemistry achievement?
In response to these research questions, this paper reviews the literature addressing key concepts such as the teacher–student relationship, interest in chemistry, chemistry academic engagement, and chemistry achievement. It also explores the relationships among these variables. Based on these theoretical foundations and empirical findings, the following hypotheses are proposed (see Fig. 1):
 | ||
| Fig. 1 Hypothesized relationships among the variables. | ||
Hypothesis 1: closeness in the teacher–student relationship is positively associated with students’ chemistry achievement.
Hypothesis 2: low conflict in the teacher–student relationship is positively associated with students’ chemistry achievement.
Hypothesis 3: interest in chemistry plays a mediating role in the relationship between the teacher–student relationship (closeness or low conflict) and students’ chemistry achievement.
Hypothesis 4: chemistry academic engagement plays a mediating role in the relationship between the teacher–student relationship (closeness or low conflict) and students’ chemistry achievement.
Hypothesis 5: (a) interest in chemistry and (b) chemistry academic engagement function as sequential mediators between the teacher–student relationship (closeness or low conflict) and students’ chemistry achievement.
Methods
Participants
This research was conducted in Henan Province, a populous and educationally significant region in China. Its education system has long been highly competitive and has been deeply involved in the promotion of national college entrance examination reforms. In September 2014, the “Opinions on Deepening the Reform of the Examination and Enrollment System’’ (General Office of the State Council, PRC) was released, marking the launch of a new phase of college entrance examination reform. Building on this, Henan Province officially announced its provincial comprehensive college entrance examination reform implementation plan in June 2022, clearly stating that the model would be implemented starting with students entering high school in the autumn of 2022 (People's Government of Henan Province). Under this model, students must complete three nationally unified examination subjects: Chinese, mathematics, and a foreign language. They then choose one subject from physics or history as their preferred subject and select two elective subjects from chemistry, biology, politics and ideology, and geography. Consequently, chemistry is a selective examination subject at the high school level, and choosing it directly influences students’ university major selection and future career paths. Existing studies have shown that during students’ critical academic decision-making and adaptation periods, teachers are key factors influencing academic achievement (Brophy, 1986; Polly et al., 2018; Peng et al., 2022). Therefore, at this stage, the supportive, guiding, and emotional roles played by teachers are particularly crucial, as they not only affect students’ short-term learning experiences and grades but also have a profound impact on their subsequent major selection and long-term development.This study selected students from two high schools and one junior high school in Henan Province as research subjects and employed a two-stage sampling method to ensure an adequate sample size and representativeness. The selection of these particular schools was based on their geographical accessibility and the feasibility of data collection. The first phase was implemented in May 2025. Four classes (a total of 12 classes) were randomly selected from each grade at High School A, consisting of students who chose chemistry as their elective subject. All students in the selected classes participated in the survey. Simultaneously, four classes were randomly selected from the ninth grade at Junior High School B as the junior high school sample. To further increase the sample size and enhance the statistical power of the analysis, a second round of sampling was conducted in early July 2025. This involved selecting four classes from the 2023 cohort at High School C (i.e., students who enrolled in 2023 and were to enter their senior year in September 2025) in which chemistry was an elective; these classes served as supplementary samples. This sampling plan utilizes random cluster sampling at the class level. This method captures heterogeneity across different grades and schools, thereby reducing selection bias. Additionally, it ensures the feasibility of field research. All participating students are at a critical stage of systematic chemistry learning. As a group, they represent secondary school students from diverse teaching environments and learning progressions. This makes the sample suitable for examining the relationship between the teacher–student relationship and chemistry achievement.
This study was conducted using a questionnaire survey. The data collection process strictly adhered to standardized protocols, as outlined in the following steps. First, the second author of this study, who served as the head of the first grade at School A, communicated with the relevant administrators at each sample school. He explained the study's purpose, content, duration, and data confidentiality measures. After obtaining permission, the second author coordinated with the schools to schedule a specific time for on-site administration. Secondly, on-site collective testing was employed. The second author of this study served as the chief examiner and entered the classroom at the scheduled time to conduct the investigation. Before the test, the examiner read standardized instructions to explain to the students the anonymity and voluntariness of the questionnaire, its purely academic purpose, and the strict confidentiality of the data, in order to minimize potential concerns and ensure the authenticity of the responses. Finally, students independently complete their answers on the paper answer sheets using black marker pens, a process that takes approximately 15 to 20 minutes. During the testing, the examiner remains present to monitor the room and provide neutral clarifications for any items with unclear meanings. After the questionnaires are completed, they are collected uniformly by the second author of this study and stored in sealed bags.
A total of 908 questionnaires were distributed in this survey. After collection, to ensure the quality of the data for subsequent analysis, we conducted a rigorous data cleaning procedure. The following exclusion criteria were applied: (1) more than one-third of the questionnaire items were left unanswered; (2) obvious contradictions in responses, such as providing completely opposite answers to positively and negatively worded items within the same scale; and (3) patterned responses, such as selecting the same option for all items or providing consistent but clearly meaningless answers to multiple consecutive items (Sjöström et al., 1999). After individual review, a total of 230 invalid questionnaires were discarded, resulting in 678 valid questionnaires. In social science research, a response rate above 70% is generally considered acceptable (Babbie, 2016). The overall effectiveness rate of this study was 74.67%, meeting the standard. Among the valid samples, there were 312 high school students who chose chemistry as an elective subject and 366 junior high school students.
Survey administration
The data collection was conducted in two phases. The first phase took place in May 2025 at High School A and Junior High School B and lasted one week. The second phase was conducted at High School C in early July 2025 and lasted four days. This study was approved by the Academic Ethics Committee of Central China Normal University and other relevant institutions. Informed consent was obtained from the parents of all student participants prior to the study. Before the investigation began, the research objectives and data usage were explained to all participants. All personal information is strictly confidential and handled in accordance with applicable national and provincial laws and regulations. Participation in this study was voluntary, and no additional rewards were provided to the students.Survey instruments
This study selected three scales: teacher–student relationship, interest in chemistry, and chemistry academic engagement. All three have been widely used in related fields (Pianta et al., 1995; Schaufeli et al., 2002a, 2002b; Ferrell and Barbera, 2015). Chinese versions of all selected scales are available. Notably, the scales for interest in chemistry and chemistry academic engagement have been applied in chemistry education contexts and demonstrate considerable contextual adaptability (Qian et al., 2023, 2024; Sun et al., 2025). To assess participants’ understanding of the scale items, we conducted cognitive interviews with several students from grades 9 to 12, asking them to explain their interpretation of each item. Based on their feedback, wording adjustments were made to some items to ensure that students’ interpretations aligned with the original intent of the questions, thereby providing procedural evidence supporting data validity. Data collection was conducted using paper questionnaires, which were scanned and converted into Excel format after retrieval. The research team then entered the valid data into SPSS software for subsequent statistical analysis.Data analysis procedures and tools
Preliminary statistical analyses were conducted using SPSS (Version 26.0). First, the mean, standard deviation, skewness, and kurtosis of each variable were calculated to assess the data distribution. Second, correlation coefficients were computed to examine the bivariate relationships among the variables. Following the recommendations of Curran et al. (1996), a distribution is considered approximately normal if the absolute values of skewness and kurtosis are less than 2 and 7, respectively.To assess the structural validity of the revised three scales, this study employed a combined approach using exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) for validation. First, the valid questionnaires (n = 678) were randomly divided into two equal parts (n1 = n2 = 339). SPSS (Version 26.0) was used to perform EFA on Part 1 (n1 = 339) to determine whether the scale's factor structure aligned with the predefined dimensions (Lee et al., 2008). Principal component analysis was utilized to extract factors, and the maximum variance method was applied. According to established criteria, the Kaiser–Meyer–Olkin (KMO) value must exceed 0.6, and the p value of Bartlett's test of sphericity must be less than 0.05 for the data to be considered suitable for factor analysis (Bartlett, 1951; Guo et al., 2022). Items with cross-loading scores less than 0.32 and factor-loading scores greater than 0.4 were retained (Wei et al., 2021).
This study employed Mplus (Version 8.3) to conduct CFA on each scale individually using Part 2 (n2 = 339) (Velayutham and Aldridge, 2013). All three scales utilize a 5-point Likert scale, and the data approximately follow a normal distribution. Therefore, maximum likelihood estimation (ML) was used for parameter estimation in the CFA. Model fit was evaluated based on the criteria proposed by Hu and Bentler (1999): χ2(df) < 5, RMSEA < 0.08, CFI > 0.90, TLI > 0.90, and SRMR < 0.08. These thresholds were adopted as indicators of good model fit (Marsh et al., 2004; Lee et al., 2008; Opperman et al., 2013).
For reliability assessment, although Cronbach's α is widely used in psychological scales, it relies on relatively strong statistical assumptions and can underestimate reliability (Peters, 2014). In contrast, McDonald's omega has less restrictive assumptions, making it more suitable for complex measurement models. Consequently, it is widely regarded as a more robust estimator (McNeish, 2018). Therefore, this study employed McDonald's omega for reliability analysis. McDonald's omega ranges from 0 to 1, with values closer to 1 indicating greater internal consistency among the items measuring the construct (McDonald, 1999). Generally, a value above 0.7 is considered to indicate acceptable reliability (Green and Yang, 2015). Analyses were conducted in SPSS (Version 26.0) using the dedicated omega.spd plug-in developed by Andrew F. Hayes (afhayes.com).
Before testing the hypothesized model, potential common method bias was assessed using the unmeasured latent method construct approach (Podsakoff et al., 2003) in Mplus (version 8.3). First, a baseline model (M1) comprising three first-order latent variables was specified. Subsequently, a second model (M2) was constructed by adding a method factor to M1. The presence of common method bias was evaluated by comparing the fit indices of M1 and M2. According to the criteria proposed by Wen et al., (2018), a significant increase in CFI and TLI (Δ > 0.10), along with a notable decrease in RMSEA and SRMR (Δ > 0.05), indicates the presence of substantial common method bias.
The data were analyzed using Mplus (Version 8.3). Structural equation modeling (SEM) was employed to test the hypothesized chain mediation pathways. The analysis was conducted in two complementary steps, each reflecting a different operationalization of the teacher–student relationship construct. First, an SEM was specified to examine the distinct effects of the two dimensions of the teacher–student relationship. Closeness (x1) and Low Conflict (x2) were modeled as correlated first-order latent variables, indicated by items g1–g4 and g5–g8, respectively. Interest in Chemistry (m1) was specified as a latent variable measured by items e1–e4. Chemistry Academic Engagement (m2) was modeled as a second-order latent construct, reflected by two first-order factors: Dedication (h1; items d1–d6) and Vigor–absorption (h2; items d7, d8, d10, d11, d13, and d17). Chemistry Achievement (y) was treated as an observed variable and represented by standardized Z-scores. Second, to examine the overall effect of teacher–student relationship quality, a second SEM was estimated. In this model, a higher-order latent factor, Teacher–Student Relationship (TSR), was specified and indicated by the two first-order latent variables Closeness (x1) and Low Conflict (x2). The model specifications for Interest in Chemistry (m1), Chemistry Academic Engagement (m2), and Chemistry Achievement (y) remained identical to those in the first model.
Model parameters were estimated using the maximum likelihood (ML) estimator. Model fit was evaluated using the same criteria as described for the CFA (Hu and Bentler, 1999). For the mediation analysis, the bias-corrected percentile bootstrap method was employed, as it is widely recognized as one of the most robust approaches (Hayes, 2015). This method does not require the assumption of normality or large samples, and it does not rely on standard error estimates to derive the mediation effect intervals (Preacher and Hayes, 2008; Taylor et al., 2008; Fang et al., 2014). Accordingly, this study applied this method with 5000 bootstrap samples at a 95% confidence level to test the significance of the mediating effects (Hayes, 2017). A mediating effect is considered statistically significant if the 95% bias-corrected bootstrap confidence interval (CI) does not include zero (Lau and Cheung, 2012; Fang et al., 2014; Hayes, 2015). When both the direct and indirect effects are significant, it indicates partial mediation; when the indirect effect is significant but the direct effect is not, it indicates complete mediation (González and Paoloni, 2015).
Suitability for measurement instruments
| KMO | Bartlett's test of sphericity | χ2/df | RMSEA | CFI | TLI | SRMR | Omega | ||
|---|---|---|---|---|---|---|---|---|---|
| χ2 | p | ||||||||
| Complete model | |||||||||
| Teacher–student relationship | 0.861 | 1099.897 | <0.001 | 1.344 | 0.032 | 0.995 | 0.992 | 0.030 | 0.854 |
| Interest in chemistry | 0.749 | 428.137 | <0.001 | 2.880 | 0.074 | 0.995 | 0.970 | 0.012 | 0.792 |
| Chemistry academic engagement | 0.935 | 2634.622 | <0.001 | 2.199 | 0.059 | 0.974 | 0.968 | 0.033 | 0.927 |
| Individual factor of teacher–student relationship | |||||||||
| Closeness | 0.807 | 544.429 | <0.001 | 2.123 | 0.058 | 0.998 | 0.990 | 0.007 | 0.864 |
| Low conflict | 0.751 | 374.934 | <0.001 | 1.897 | 0.051 | 0.998 | 0.985 | 0.009 | 0.780 |
| Individual factor of chemistry academic engagement | |||||||||
| Dedication | 1.286 | 0.029 | 0.998 | 0.996 | 0.014 | 0.877 | |||
| Vigor-absorption | 1.776 | 0.048 | 0.994 | 0.990 | 0.017 | 0.893 | |||
| Acceptable | >0.6 | — | — | <5 | <0.08 | >0.9 | >0.9 | <0.08 | >0.7 |
Results
Common method deviation test
To assess the potential impact of common method bias, the unmeasured latent method factor technique (Podsakoff et al., 2003) was employed as the primary test. After including a common method factor, the changes in the model fit indices were as follows: ΔCFI = 0.011, ΔTLI = 0.007, ΔRMSEA = 0.003, and ΔSRMR = 0.005. All changes fell below the recommended thresholds for meaningful improvement (Wen et al., 2018), indicating that adding of the method factor did not substantially improve model fit. Collectively, these results suggested that common method bias was unlikely to pose a serious threat to the validity of the findings in this study.Descriptive statistics and correlational analysis
The mean, standard deviation, skewness, kurtosis of each variable, and the correlation coefficients among the variables are presented in Table 2. The normality test results indicated that skewness values ranged from −0.719 to −0.142, and kurtosis values ranged from −0.569 to 0.319, all within the acceptable limits for a normal distribution (Curran et al., 1996). The correlation analysis revealed that both the magnitude and direction of the correlation coefficients between variables were consistent with expectations. Specifically, closeness and low conflict in the teacher–student relationship were each significantly and positively correlated with students’ interest in chemistry, chemistry academic engagement (including dedication and vigor-absorption), and chemistry achievement.| Variables | M | SD | Skewness | Kurtosis | 1 | 2 | 3 | 4 | 5 | 6 |
|---|---|---|---|---|---|---|---|---|---|---|
| Note: *p < 0.05, **p < 0.01, ***p < 0.001; M = mean; SD = standard deviation. | ||||||||||
| (1) Closeness | 3.702 | 0.749 | −0.719 | 0.319 | 1 | |||||
| (2) Low conflict | 3.994 | 0.756 | −0.489 | −0.451 | 0.553** | 1 | ||||
| (3) Interest in chemistry | 3.958 | 0.688 | −0.649 | −0.007 | 0.522** | 0.526** | 1 | |||
| (4) Dedication | 3.742 | 0.681 | −0.488 | 0.030 | 0.561** | 0.482** | 0.670** | 1 | ||
| (5) Vigor-absorption | 3.314 | 0.851 | −0.142 | −0.569 | 0.529** | 0.438** | 0.636** | 0.742** | 1 | |
| (6) Chemistry achievement | 0.043 | 0.910 | — | — | 0.431** | 0.277** | 0.443** | 0.473** | 0.413** | 1 |
Analyses of the chain mediating effects
This study utilized Mplus (version 8.3) to construct structural equation models, with closeness and conflict in the teacher–student relationship as independent variables; chemistry achievement as the dependent variable; and interest in chemistry and chemistry academic engagement as mediating variables, to examine their chain mediation effects. Bootstrap sampling was set to 5000 repetitions, and effects were considered significant if the 95% confidence intervals did not include zero (Hayes, 2015). The overall model demonstrated good fit to the data: χ2/df = 2.741, CFI = 0.948, TLI = 0.941, RMSEA = 0.051, SRMR = 0.039. The results of the analysis are presented in Fig. 2 and Table 3. | ||
| Fig. 2 Significant effects from path model: standardized beta coefficients. Notes: *p < 0.05, **p < 0.01, ***p < 0.001. The dotted line indicates that the path coefficient is not significant. | ||
| Path | Effect | SE | CI95 | p | |
|---|---|---|---|---|---|
| LL | UL | ||||
| Note: SE = standard error, CI95 = 95% confidence interval, LL = lower level, UL = upper level. Confidence intervals were calculated using 5000 bootstraps. All reported estimates are standardized. | |||||
| Direct effect | |||||
| Closeness → chemistry achievement | 0.256 | 0.061 | 0.090 | 0.344 | <0.001 |
| Low conflict → chemistry achievement | −0.175 | 0.043 | −0.241 | −0.086 | 0.001 |
| Indirect effects | |||||
| Closeness → Interest in chemistry → chemistry achievement | 0.092 | 0.024 | 0.049 | 0.147 | <0.001 |
| Closeness → chemistry academic engagement → chemistry achievement | 0.060 | 0.016 | 0.034 | 0.097 | <0.001 |
| Closeness → interest in chemistry → chemistry academic engagement → chemistry achievement | 0.053 | 0.017 | 0.033 | 0.101 | 0.001 |
| Low conflict → interest in chemistry → chemistry achievement | 0.113 | 0.041 | 0.043 | 0.212 | <0.001 |
| Low conflict → chemistry academic engagement → chemistry achievement | −0.003 | 0.018 | −0.033 | 0.035 | 0.810 |
| Low conflict → interest in chemistry → chemistry academic engagement → chemistry achievement | 0.065 | 0.021 | 0.034 | 0.134 | <0.001 |
| Total effect | |||||
| Closeness → chemistry achievement | 0.461 | 0.073 | 0.368 | 0.616 | <0.001 |
| Low conflict → chemistry achievement | 0.000 | 0.080 | −0.148 | 0.156 | 0.993 |
The direct effect analysis indicated that closeness in the teacher–student relationship has a significant positive direct impact on chemistry achievement (direct effect = 0.256, 95% CI [0.090, 0.344]). Conversely, low conflict in the teacher–student relationship has a significant negative direct effect on chemistry achievement (direct effect = −0.175, 95% CI [−0.241, −0.086]).
Indirect effect analysis reveals complex mediating mechanisms. Closeness in the teacher–student relationship has a significant positive impact on chemistry achievement through three indirect pathways: the independent mediating effect of interest in chemistry was significant (indirect effect = 0.092, 95% CI [0.049, 0.147], p < 0.001); the independent mediating effect of chemistry academic engagement was significant (indirect effect = 0.060, 95% CI [0.034, 0.097], p < 0.001); and the chain mediating effect of “Closeness in the teacher–student relationship → interest in chemistry → chemistry academic engagement → chemistry achievement” was significant (indirect effect = 0.053, 95% CI [0.033, 0.101], p = 0.001). For low conflict in the teacher–student relationship, the independent mediating effect through interest in chemistry was significant (indirect effect = 0.113, 95% CI [0.043, 0.212], p < 0.001). However, the independent mediating effect through chemistry academic engagement was not significant (indirect effect = −0.003, 95% CI [−0.033, 0.035], p = 0.810). The chain mediating effect through the pathway “Low conflict in the teacher–student relationship → interest in chemistry → chemistry academic engagement → chemistry achievement” was significant (indirect effect = 0.065, 95% CI [0.034, 0.134], p < 0.001).
Discussion
The relationship between teacher–student relationship and chemistry achievement
The results of this study revealed distinct patterns of influence for different dimensions of the teacher–student relationship on chemistry achievement. Specifically, closeness had a significant direct positive effect on achievement. In contrast, lower conflict was associated with a significant negative direct effect on achievement. This result was not entirely consistent with the original hypothesis. Therefore, to further clarify the overall effect of the teacher–student relationship, this study combined closeness and conflict into a comprehensive variable termed the “overall quality of the teacher–student relationship” and reconstructed a structural equation model for analysis. The overall model demonstrated good fit to the data: χ2/df = 2.815, CFI = 0.946, TLI = 0.939, RMSEA = 0.052, SRMR = 0.040. The results indicated that a higher overall quality of the teacher–student relationship, characterized by greater closeness and lower conflict, predicted better chemistry achievement. Further mediation analysis revealed that this positive impact was primarily achieved through three pathways: first, the mediating effect of interest in chemistry (effect value = 0.155, 95% CI [0.125, 0.235], p < 0.001); second, the mediating effect of chemistry academic engagement (effect value = 0.064, 95% CI [0.049, 0.111], p < 0.001); and third, the chain mediating pathway of “Teacher–student relationship → Interest in chemistry → Chemistry academic engagement → Chemistry achievement” (effect value = 0.109, 95% CI [0.082, 0.170], p < 0.001). Detailed results are presented in the Appendix (Table 4 and Fig. 3).Firstly, the present study found that closeness in the teacher–student relationship positively impacted students’ chemistry achievement. This can be explained by social support theory, which posits that support from significant others, such as teachers, can encourage individuals to transform external trust into internal motivation, thereby enhancing academic performance (Duckworth et al., 2014). In chemistry, where learning relies on grappling with abstract concepts, a close teacher–student relationship can be crucial. It likely promotes achievement by reducing anxiety and creating a psychologically safe environment that encourages active inquiry and persistence. This finding aligns with numerous empirical studies indicating a positive correlation between teacher–student closeness and academic achievement. Research by Kashy-Rosenbaum et al. (2018) demonstrates that both a positive class-level emotional atmosphere and support from head teachers significantly predict students’ academic achievement. Furthermore, numerous recent studies have consistently shown that closeness in the teacher–student relationship is positively correlated with students’ academic achievement (Sointu et al., 2017; Xu and Qi, 2019; Sherub and Nima, 2022). However, the findings of this study revealed some discrepancies compared to a minority of existing research. For example, Rucinski et al. (2018), in a study of elementary school students, did not find a significant predictive effect of teacher–student relationships on mathematics achievement. Similarly, Bryce et al., 2019 reported no significant association between closeness in the teacher–student relationship and student academic achievement. These results suggest that the relationship may vary across samples and contexts. These discrepancies can be understood by considering several factors. First, our participants were middle and high school students who had chosen to study chemistry. This elective status suggested a higher baseline interest, which a close teacher–student relationship could further nurture and sustain. Second, the cultural context of our study, which emphasized respect for teachers, might have amplified the influence of this relational bond on learning behaviors. Third, methodological differences, such as the source of relationship reporting (teacher vs. student), can lead to divergent results (Hughes et al., 2012).
Secondly, this study found that low conflict teacher–student relationship had a direct negative impact on chemistry achievement. Previous research examining the relationship between teacher–student conflict and student achievement generally reveals two patterns. One line of research has found that higher levels of teacher–student conflict are weakly negatively associated with student achievement. For instance, Hajovsky et al. (2020) specifically found that teacher–student conflict had both direct and indirect effects on students’ mathematics achievement. Similarly, a meta–analysis conducted by Göktaş and Kaya (2023) indicated that negative teacher–student relationships were significantly, though weakly, negatively associated with students’ academic achievement. Another line of research has suggested that the negative association between teacher–student conflict and academic achievement is not significant. For example, Mason et al. (2017) analyzing longitudinal data, found that while teacher–student conflict had a statistically significant impact on subsequent math achievement, the effect was statistically significant but small in magnitude. Similarly, McKinnon and Blair (2019) in a longitudinal study of young children found that conflictual relationships significantly predicted executive function and reading achievement, but were not related to math achievement. In the present study, when teacher–student closeness was controlled, low teacher–student conflict was significantly negatively associated with students’ chemistry achievement. At first glance, this finding appears inconsistent with previous research. However, this result should be interpreted in light of model specification and the relationships among variables. In the structural equation model, teacher–student closeness and low teacher–student conflict were highly correlated (r = 0.678), indicating that the two constructs share substantial information regarding relationship quality in real educational contexts. When both constructs were included simultaneously, the model estimated their unique effects independent of each other, rather than the overall effect of relationship quality. In this process, the positive components of low conflict that overlapped with closeness were statistically partialled out. As a result, the remaining component of low conflict may carry different psychological and instructional meanings. Methodological research indicates that after controlling for shared variance among latent variables, model estimates may diverge from bivariate associations, and path directions or significance can change (Cheung and Lau, 2008). In this study, the unique component of low conflict may not represent high-quality teacher–student interactions. It may instead indicate relationships with limited interaction, avoidance of disagreement, or shallow academic communication. Previous studies have suggested that classroom interactions that overly emphasize surface harmony while lacking constructive discussion may weaken students’ cognitive engagement and deep learning (Nystrand et al., 2003; Gutentag et al., 2022; Morek et al., 2023). For subjects like chemistry, which require cognitive conflict, inquiry, and conceptual understanding, relationships lacking academic challenge and cognitive engagement may hinder learning and achievement.
Several additional factors may help explain the unexpected negative association. First, cultural context and classroom structure are important. The participants were students in Henan Province, China, where the educational environment is highly competitive. In such contexts, low conflict with teachers does not necessarily indicate supportive or attentive interactions. Research on class size shows that in larger classes, students, particularly those with lower academic performance, receive less individualized attention and participate less actively in classroom interactions, while smaller classes promote more teacher–student interaction and higher engagement (Blatchford et al., 2011). Consequently, students with low conflict relationships in large, competitive classes may not benefit from sufficient guidance or encouragement, potentially explaining the observed negative association between low conflict and chemistry achievement. Second, the findings may be influenced by the student-only perspective. Teacher–student relationship data were collected solely through students’ self-reports, and students’ perceptions of low conflict may differ from teachers’ views (Hughes et al., 2012). Therefore, the negative association may reflect students’ subjective experience rather than the objective quality of interactions, highlighting a potential measurement limitation. Together, these factors underscore the complexity of how different dimensions of teacher–student relationships affect student achievement in specific cultural and educational settings. The negative association observed in the model does not imply that teacher–student conflict is beneficial. Instead, it highlights that the mechanisms through which teacher–student relationship dimensions influence academic performance are complex and may vary when multiple dimensions operate simultaneously.
In conclusion, although different dimensions of teacher–student relationships in the structural equation model exhibit varying direct effects, the overall findings of this study indicate that higher quality teacher–student relationships are associated with better student performance in chemistry. This conclusion aligns with existing research and further supports the view that teacher–student relationships are a crucial environmental factor influencing students’ chemistry achievement.
The mediating roles of (a) interest in chemistry and (b) chemistry academic engagement
This study further examined the mediating roles of (a) interest in chemistry and (b) chemistry academic engagement in the relationship between teacher–student relationships and chemistry achievement. For teacher–student closeness, the results demonstrated a significant positive direct effect on chemistry achievement. Additionally, teacher–student closeness exerted significant indirect effects through two pathways: (a) interest in chemistry alone and (b) chemistry academic engagement alone. These findings indicate that both (a) interest in chemistry and (b) chemistry academic engagement partially mediate the relationship between teacher–student closeness and chemistry achievement. For low teacher–student conflict, a different pattern emerged. The direct effect of low conflict on chemistry achievement was significantly negative. Regarding indirect effects, low conflict was positively associated with chemistry achievement through interest in chemistry alone, indicating that interest served as a significant mediator. However, the mediating role of chemistry academic engagement alone was not significant, as the path from low conflict to chemistry academic engagement was not statistically significant. Taken together, these findings suggest that the relationship between low teacher–student conflict and chemistry achievement is complex.The mediating role of interest in chemistry was supported. First, both closeness and low conflict in the teacher–student relationship positively predicted students’ interest in chemistry. This finding aligns with self–determination theory, which posits that intrinsic motivation arises from satisfying the need for relatedness, among other factors (Deci and Ryan, 2000). Empirically, positive teacher–student interactions have been shown to enhance learning interest (Asadzadeh et al., 2019; Yang et al., 2025). Conversely, conflict can cause students to feel disrespected, potentially leading to resistance toward the teacher and the subject, thereby reducing interest. Using latent profile analysis, Burns et al. (2022) identified four types of teacher–student relationships: positive, complex, distant, and negative. Among these, students with distant relationships exhibited lower learning interest, while those in the negative relationship group demonstrated the lowest learning interest. Second, interest positively predicted chemistry achievement. This positive relationship between learning interest and academic achievement is well established in the literature (Köller et al., 2001; Heinze et al., 2005; Ryan and Deci, 2020). This connection can be further understood through the lens of attachment theory, which posits that teacher–student relationships form an affective bond characterized by closeness, warmth, and care. A supportive relationship provides a secure emotional base from which students can confidently explore their school environment, including challenging subjects, and promotes academic achievement by enhancing students’ emotional and cognitive regulation (Pianta and Stuhlman, 2004). Therefore, by integrating these two perspectives, the current study confirms that interest in chemistry serves as a significant mediator. This research offers empirical support for attachment theory and self–determination theory within the context of chemistry education, thereby establishing a theoretical foundation for fostering positive teacher–student relationships. The findings also emphasize the importance of teachers cultivating supportive relationships with students, enhancing their interest in chemistry through diverse approaches, and providing more opportunities for self-affirmation.
The mediating role of chemistry academic engagement requires nuanced interpretation, as it operates differently across the two dimensions of teacher–student relationships. For teacher–student closeness, we found that chemistry academic engagement significantly mediated the relationship between teacher–student closeness and chemistry achievement. This finding aligns with prior research demonstrating that positive teacher–student relationships foster students' academic engagement. For example, the systematic review by Quin (2017), which included 46 studies, revealed that high-quality teacher–student relationships are significantly and positively associated with multiple dimensions of learning engagement among middle school students, including psychological engagement, academic performance, and attendance rates. Furthermore, academic engagement has been consistently linked to improved academic achievement. Analyzing data from the PISA 2000 assessment in the United States, Lee (2014) found a significant positive relationship between student engagement and academic achievement. Taken together, these studies suggest that engagement serves as a key mechanism through which positive teacher–student relationships translate into academic success. This interpretation is further supported by the meta-analysis of Roorda et al. (2011), which reported that while both closeness and conflict showed relatively weak direct correlations with achievement (r = 0.16 and −0.18, respectively), their associations with engagement were considerably stronger (r = 0.34 and −0.31). Consistent with this body of evidence, our results provide robust empirical support for chemistry academic engagement as a mediator in the link between teacher–student closeness and chemistry achievement.
For teacher–student conflict, however, the role of academic engagement was not supported as a mediator. The direct path from low conflict to chemistry academic engagement was not statistically significant and showed a negative trend. This indicates that merely having a low conflict relationship with teachers does not necessarily lead students to actively engage in chemistry learning. From the perspective of social conflict theory (Coser, 1957), this finding can be explained by distinguishing between different types of conflict. The “low conflict” measured in this study reflects the absence of non-realistic conflict, specifically a reduction in interpersonal tension, disrespect, or unfair treatment. While minimizing such emotionally charged conflict fosters a psychologically safe environment, it does not inherently provide the intellectual stimulation or task-focused challenges characteristic of realistic conflict. Realistic conflict arises from disagreements over tasks and goals and can serve constructive purposes in learning. Several empirical findings further support this interpretation. First, low conflict primarily reduces interpersonal stress but does not provide positive guidance (Praetorius et al., 2018), encouragement, or motivational support that directly stimulates learning behavior. Students may experience minimal conflict without feeling challenged, supported, or inspired to participate actively in classroom activities. In contrast, closeness reflects positive emotional and instructional support that fosters engagement. Second, in the competitive educational context of China's Henan Province, low conflict may be perceived as neutral rather than supportive; teachers may maintain low conflict by limiting interaction or attention to each student (Fortes and Tchantchane, 2010), especially in large classes, which may fail to motivate active learning engagement. Therefore, low conflict relationships may not significantly influence students’ behavioral or cognitive engagement, which explains the non-significant mediating role of academic engagement in the relationship between low conflict and chemistry achievement. In contrast, for teacher–student closeness, chemistry academic engagement emerged as a significant mediator, highlighting its role as a key mechanism through which positive, supportive relationships lead to improved chemistry achievement.
The chain mediating effect of (a) interest in chemistry and (b) chemistry academic engagement
The third finding of this study is the significant chain mediating effect of the sequence “teacher–student relationship (closeness/low conflict) → interest in chemistry → chemistry academic engagement → chemistry achievement.” For teacher–student closeness, the chain mediation pathway was significant: closeness positively predicted interest in chemistry, which in turn positively predicted academic engagement, ultimately leading to higher achievement in chemistry. This sequential pattern indicates that close teacher–student relationships foster chemistry achievement not only through the independent effects of interest and engagement but also by first stimulating interest, which then activates engagement. For low teacher–student conflict, the chain mediation pathway was also statistically significant. However, this finding should be interpreted cautiously in light of the direct effects. As shown in Table 3 and Fig. 2, low conflict had a significant negative direct association with chemistry achievement, while its direct path to chemistry academic engagement was not significant. Therefore, the chain mediation does not suggest that low conflict directly promotes students’ academic engagement. Instead, the results indicate that when low conflict is associated with greater interest in chemistry, this increased interest may subsequently enhance students’ academic engagement, which in turn relates to higher achievement. In this way, the positive sequential indirect effect partially offsets the negative direct relationship between low conflict and chemistry achievement, making the overall association less negative. These findings align with prior research. The connection between teacher–student relationships and student interest is well established in motivational theories, such as self-determination theory. Moreover, interest has been identified as a key precursor to engagement (Mazer, 2013; Eseryel et al., 2014; Kahu et al., 2017). By empirically validating this sequential pathway, our study clarifies that (a) interest in chemistry and (b) chemistry academic engagement serve as sequential mediators in the relationship between teacher–student relationships and chemistry achievement.Implication for practice
The results of this study indicate that closeness in the teacher–student relationship is significantly and positively correlated with students’ chemistry achievement. In addition, lower levels of conflict in the teacher–student relationship enhance students’ chemistry achievement through the mediating effect of increased interest in chemistry learning. Based on these findings, the following recommendations for educational practice are proposed.First, chemistry teachers should intentionally cultivate close and trusting teacher–student relationships by fostering supportive classroom interactions grounded in respect, care, and professional guidance. According to self-determination theory (Deci and Ryan, 2000; Verschueren, 2015), close teacher–student relationships help satisfy students’ needs for autonomy, competence, and relatedness, thereby enhancing intrinsic motivation and promoting positive learning behaviors. In chemistry classrooms, such relationships can be fostered through dialogic communication and the provision of continuous academic and emotional support. When teachers demonstrate respect, understanding, and care during instruction, they help create positive emotional connections that promote students’ learning and development (Sherub and Nima, 2022). In addition, the quality of teacher–student relationships in chemistry may be influenced by teachers’ affective engagement with the subject during classroom instruction. Research grounded in social–cognitive theories of emotion suggests that teachers’ positive emotions in the classroom can be conveyed to students through expressed enthusiasm, thereby shaping students’ learning experiences (Frenzel et al., 2009). When chemistry teachers use vivid explanations, real-life examples, and emphasize the relevance and value of chemistry, they can transmit their passion for the subject, which fosters students’ interest and, in turn, promotes higher levels of academic engagement in chemistry.
Second, chemistry teachers may benefit from adopting a more cautious and balanced approach when addressing teacher–student conflicts in the classroom and considering how different types of conflict can influence students’ learning. According to the conflict theory proposed by Coser (1957), conflicts in social interactions can have both destructive and constructive functions. Emotionally driven interpersonal conflicts may undermine trust and support in teacher–student relationships, thereby reducing students’ motivation to learn, whereas constructive conflicts focused on ideas or learning tasks may promote deeper thinking and engagement (Asterhan and Schwarz, 2007). The findings of this study provide some support for this perspective. On one hand, lower levels of teacher–student conflict were found to indirectly promote students’ chemistry achievement through the chain mediation pathway of interest in chemistry → chemistry academic engagement. On the other hand, the results also showed that lower conflict was negatively associated with chemistry achievement. This seemingly contradictory finding suggests that the relationship between teacher–student conflict and chemistry achievement may be complex. It indicates that different types of conflict may exist in the classroom, and their educational functions should not be interpreted simplistically. In authentic chemistry classroom settings, certain task-related conflicts stemming from learning norms and academic standards – such as correcting the writing of chemical symbols, balancing chemical equations, or enforcing laboratory procedures – can help students develop productive learning habits. Therefore, in chemistry teaching practice, teachers may need to tailor their responses to different types of conflict based on the specific context. While maintaining clear academic expectations and laboratory standards, teachers should also minimize emotionally driven interpersonal conflicts by fostering respectful and supportive communication. For example, when students struggle to understand concepts or perform laboratory procedures, teachers can use non-judgmental communication and emphasize problem-solving rather than emotional confrontation (Penner et al., 2024). Simultaneously, teachers can use purposeful questioning to stimulate constructive, task-related discussions or cognitive conflicts during learning activities (Dohrn and Dohn, 2018). Such questions encourage students to explain their reasoning and engage more deeply with chemistry concepts.
Third, the design of chemistry instruction should focus on stimulating interest and encouraging engagement. Research indicates that both (a) interest in chemistry and (b) chemistry academic engagement serve as chain mediators between the two types of teacher–student relationships and chemistry achievement. This requires that chemistry classrooms should shift from the explanation of knowledge to the stimulation of learning motivation. Chemistry teachers should design learning activities that stimulate students’ intrinsic motivation and interest in the subject (Qian et al., 2023). Examples include inquiry-based experiments, collaborative group tasks, and chemistry projects related to current social issues. These activities enable students to actively engage with and appreciate the appeal of the chemistry discipline. At the same time, by designing progressively challenging learning tasks and providing timely feedback, it helps students establish a positive cycle in which their efforts are rewarded, thereby continuously enhancing their behavioral and cognitive engagement.
Limitations and further directions
Although this study revealed, through a chain mediation model, the intrinsic mechanism by which the teacher–student relationship influences chemistry achievement via interest and engagement in chemistry learning and reached valuable conclusions, several limitations remain. Future research can build upon these findings to explore the topic in greater depth.Firstly, this study employed a cross-sectional design, with all data collected at a single time point, which limits our ability to draw causal inferences between variables. Although the structural equation model utilized latent variable modeling to distinguish the unique effects of closeness and low conflict teacher–student relationships, to make their influences as independent as possible, these results still do not directly reflect causal relationships or dynamic effects over time. Therefore, the findings regarding the effects of the low conflict and closeness dimensions of teacher–student relationships on chemistry achievement should be interpreted with caution. Future research could adopt a longitudinal design, measuring teacher–student relationships, learning interest, academic engagement, and achievement at multiple time points, combined with latent variable modeling, to verify the stability and causal direction of these pathways.
Secondly, this study has limitations related to sample representativeness and data structure. Although the rigorous data cleaning process improved measurement quality, it also reduced the sample size. While the final valid response rate of 74.67% meets the acceptable standard in social science research (Babbie, 2016), the exclusion of a substantial number of questionnaires may still limit the generalizability of the findings. Specifically, it cannot be ruled out that the excluded responses systematically differed from the retained sample in terms of learning motivation, academic engagement, or achievement, potentially introducing selection bias. Additionally, the sample was drawn from a relatively concentrated source, comprising one middle school and two high schools, with classes rather than individual students selected as the sampling units. It is important to note that the sample included both Grade 9 students, for whom chemistry is a compulsory subject, and senior secondary students who continue studying chemistry under the subject selection system. Therefore, participants from both school levels were receiving systematic chemistry instruction at the time of data collection. Consequently, the findings primarily reflect the characteristics of students actively engaged in formal chemistry learning and may not be fully generalizable to those who discontinue chemistry study after fulfilling basic academic proficiency requirements in senior secondary education. Together, these factors limit the extent to which the findings can be generalized to broader student populations following different instructional pathways. It is also important to note that the data have a nested structure, with students nested within classes and schools. Due to the limited number of classes, multilevel modeling was not employed to systematically control for class-level or school-level effects in the formal analyses. Under these circumstances, midterm examination scores were standardized using Z-scores to enhance comparability across classes and grade levels. Consequently, the achievement variable primarily reflects students’ relative standing within their respective instructional contexts, rather than absolute differences in academic performance across schools. Future research could broaden the range of sampled schools and regions to enhance representativeness, employ multilevel modeling techniques to explicitly account for class-level and school-level influences, and incorporate multiple assessments or standardized achievement measures to strengthen the robustness and external validity of the model estimates.
Thirdly, the data collection primarily relies on student self-reporting, which may introduce common methodological biases. This is particularly true for subjective variables, such as perceptions of the teacher–student relationship, where data from a single source can result in measurement errors. Future research should incorporate multi-source data, including teacher evaluations and classroom behavior observations, and consider using objective academic indicators (such as the average of multiple test scores). These approaches would enhance the objectivity and accuracy of the measurements and provide a more comprehensive validation of the study's findings.
Fourth, this study primarily focuses on the emotional dimension (closeness/low conflict) of the teacher–student relationship and its impact on students’ chemistry achievement, adopting a relatively narrow perspective. In future research, additional variables could be introduced to more thoroughly elucidate the underlying mechanisms. For example, moderating variables such as students’ gender, grade level, and school climate could be examined. Additionally, factors like cognitive interactions between teachers and students and students’ metacognitive strategies could be incorporated to more comprehensively and clearly reveal the complex processes through which the teacher–student relationship influences students’ chemistry achievement. Furthermore, future research could more precisely differentiate between various types of classroom conflict in chemistry learning contexts, such as interpersonal conflicts and task-related conflicts, and investigate how these distinct forms of conflict differentially affect students’ learning motivation, academic engagement, and chemistry achievement.
Conclusions
This study aimed to investigate the mechanisms through which the teacher–student relationship influences chemistry achievement among junior and senior high school students. The results indicate that closeness in the teacher–student relationship is positively associated with chemistry achievement, and that this association operates primarily through three mediating pathways: interest in chemistry, chemistry academic engagement, and the sequential pathway from interest in chemistry to chemistry academic engagement. In other words, closeness in the teacher–student relationship not only directly enhances chemistry achievement but also indirectly promotes it by fostering students’ interest in chemistry and chemistry academic engagement. In contrast, low conflict in the teacher–student relationship shows a significant negative direct association with chemistry achievement. Specifically, although low conflict in the teacher–student relationship is directly associated with lower chemistry achievement, it can indirectly enhance chemistry achievement by increasing students’ interest in chemistry. Moreover, low conflict in the teacher–student relationship also promotes chemistry achievement through a chain mediation pathway, in which it first enhances interest in chemistry, which in turn increases chemistry academic engagement, ultimately leading to higher chemistry achievement. Overall, this study highlights the critical role of the teacher–student relationship in students’ chemistry achievement. While closeness in the teacher–student relationship can enhance chemistry achievement by stimulating students’ interest in chemistry and chemistry academic engagement, the nuanced effects of low conflict in the teacher–student relationship suggest that teachers should focus on maintaining high-quality classroom interactions and learning motivation, balancing students’ interest in chemistry and chemistry academic engagement to achieve optimal chemistry learning outcomes.Conflicts of interest
There are no conflicts to declare.Data availability
The raw data supporting the findings of this study are not publicly available due to ethical restrictions and the sensitive nature of the information, which could compromise research participant privacy. However, de-identified data generated for the main analyses are available from the corresponding author, Yanli Li, upon reasonable request. Requestors will be required to complete a data use agreement, ensuring the data are used solely for agreed-upon purposes and that participant confidentiality is preserved. Inquiries and access requests should be directed to liyanli202203@163.com.Supplementary information (SI) is available. See DOI: https://doi.org/10.1039/d5rp00404g.
Appendix
Table 4 and Fig. 3.| Path | Effect | SE | CI95 | p | |
|---|---|---|---|---|---|
| LL | UL | ||||
| Note: SE = standard error, CI95 = 95% confidence interval, LL = lower level, UL = upper level. Confidence intervals were calculated using 5000 bootstraps. All reported estimates are standardized. Path abbreviations: x = teacher–student relationship; m1 = interest in chemistry; m2 = chemistry academic engagement; y = chemistry achievement. | |||||
| Direct effect | |||||
| x → y | 0.124 | 0.046 | 0.024 | 0.204 | 0.007 |
| Indirect effects | |||||
| x → m1 → y | 0.155 | 0.030 | 0.125 | 0.235 | <0.001 |
| x → m2 → y | 0.064 | 0.018 | 0.049 | 0.111 | <0.001 |
| x → m1 → m2 → y | 0.109 | 0.022 | 0.082 | 0.170 | <0.001 |
| Total effect | |||||
| x → y | 0.452 | 0.047 | 0.430 | 0.627 | <0.001 |
 | ||
| Fig. 3 Structural equationting roles of interest in chemistry and chemistry academic engagement between the overall teacher–student relationship and chemistry achievement. | ||
Acknowledgements
We thank the two reviewers for their careful reading of our manuscript and for offering constructive comments for revisions. This research was supported by the General Project of Education under the 14th Five-Year Plan of the National Social Science Foundation of China in 2022 (BHA220113). We express our sincere thanks to the teachers and students who participated in this study. I would very much like to thank Associate Professor Jack Barbera and Editor Professor Scott Lewis for their guidance and support during the handling of this manuscript, and I also extend my sincere gratitude to the entire editorial team of Chemistry Education Research and Practice for their hard work throughout the review and publication process.References
- Abrantes J. L., Seabra C. and Lages L. F., (2007), Pedagogical affect, student interest, and learning performance, J. Bus. Res., 60, 960–964.
- Ahn I., Chiu M. M. and Patrick H., (2021), Connecting teacher and student motivation: Student-perceived teacher need-supportive practices and student need satisfaction, Contemp. Educ. Psychol., 64, 101950.
- Ainley M. and Ainley J., (2011), Student engagement with science in early adolescence: The contribution of enjoyment to students’ continuing interest in learning about science, Contemp. Educ. Psychol., 36, 4–12.
- Asadzadeh N., Zoghi M. and Assadi N., (2019), Investigating the Impacts of Teacher–Student Interactions on Improving Learners' Academic Achievements: The Role of Academic Interest, J. Lang. Transl., 9, 105–119.
- Asterhan C. S. C. and Schwarz B. B., (2007), The effects of monological and dialogical argumentation on concept learning in evolutionary theory, J. Educ. Psychol., 99, 626–639.
- Babbie E., (2016), The Practice of Social Research, Boston, MA: Cengage Learning, 14th edn, ch. Survey Research, p. 265.
- Baldwin N. and Orgill M., (2019), Relationship between teaching assistants' perceptions of student learning challenges and their use of external representations when teaching acid-base titrations in introductory chemistry laboratory courses, Chem. Educ. Res. Pract., 20, 821–836.
- Bartlett M. S., (1951), The effect of standardization on a χ2 approximation in factor analysis, Biometrika, 38, 337–344.
- Berti C., Molinari L. and Speltini G., (2010), Classroom justice and psychological engagement: students' and teachers' representations, Soc. Psychol. Educ., 13, 541–556.
- Blatchford P., Bassett P. and Brown P., (2011), Examining the effect of class size on classroom engagement and teacher–pupil interaction: Differences in relation to pupil prior attainment and primary vs. secondary schools, Learn. Instruct., 21, 715–730.
- Brophy J., (1986), Teacher influences on student achievement, Am. Psychol., 41, 1069.
- Bryce C. I., Bradley R. H., Abry T., Swanson J. and Thompson M. S., (2019), Parents' and Teachers' Academic Influences, Behavioral Engagement, and First- and Fifth-Grade Achievement, Sch. Psychol., 34, 492–502.
- Burns E. C., Van Bergen P., Leonard A. and Amin Y., (2022), Positive, complicated, distant, and negative: How different teacher–student relationship profiles relate to students' science motivation, J. Adolesc., 94, 1150–1162.
- Cheung G. W. and Lau R. S., (2008), Testing mediation and suppression effects of latent variables – Bootstrapping with structural equation models, Org. Res. Methods, 11, 296–325.
- Cole D. A., (1987), Utility of confirmatory factor analysis in test validation research, J. Consult. Clin. Psychol., 55, 584–594.
- Coser L. A., (1957), Social Conflict and the Theory of Social Change, Br. J. Soc., 8, 197–207.
- Curran P. J., West S. G. and Finch J. F., (1996), The robustness of test statistics to nonnormality and specification error in confirmatory factor analysis, Psychol. Methods, 1, 16–29.
- Deci E. L. and Ryan R. M., (2000), The “what” and “why” of goal pursuits: Human needs and the self-determination of behavior, Psychol. Inquiry, 11, 227–268.
- Dohrn S. W. and Dohn N. B., (2018), The role of teacher questions in the chemistry classroom, Chem. Educ. Res. Pract., 19, 352–363.
- Duckworth A. L., Gendler T. S. and Gross J. J., (2014), Self-control in school-age children, Educ. Psychol., 49, 199–217.
- Eseryel D., Law V., Ifenthaler D., Ge X. and Miller R., (2014), An Investigation of the Interrelationships between Motivation, Engagement, and Complex Problem Solving in Game-based Learning, Educ. Technol. Soc., 17, 42–53.
- Fang J., Wen Z., Zhang M. and Sun P., (2014), The Analyses of Multiple Mediation Effects Based on Structural Equation Modeling, J. Psychol. Sci., 37, 735–741.
- Ferrell B. and Barbera J., (2015), Analysis of students' self-efficacy, interest, and effort beliefs in general chemistry, Chem. Educ. Res. Pract., 16, 318–337.
- Fortes P. C. and Tchantchane A., (2010), Dealing with Large Classes: A Real Challenge, Proc. – Soc. Behav. Sci., 8, 272–280.
- Fredricks J. A., Blumenfeld P. C. and Paris A. H., (2004), School engagement: Potential of the concept, state of the evidence, Rev. Educ. Res., 74, 59–109.
- Frenzel A. C., Goetz T., Lüdtke O., Pekrun R. and Sutton R. E., (2009), Emotional Transmission in the Classroom: Exploring the Relationship Between Teacher and Student Enjoyment, J. Educ. Psychol., 101, 705–716.
- Göktaş E. and Kaya M., (2023), The Effects of Teacher Relationships on Student Academic Achievement: A Second Order Meta-Analysis, Participat. Educ. Res., 10, 275–289.
- González A. and Paoloni P. V., (2015), Perceived autonomy-support, expectancy, value, metacognitive strategies and performance in chemistry: a structural equation model in undergraduates, Chem. Educ. Res. Pract., 16, 640–653.
- Green S. B. and Yang Y. Y., (2015), Evaluation of Dimensionality in the Assessment of Internal Consistency Reliability: Coefficient Alpha and Omega Coefficients, Educ. Meas.-Issues Pract., 34, 14–20.
- Guo X., Deng W., Hu K., Lei W., Xiang S. and Hu W., (2022), The effect of metacognition on students' chemistry identity: the chain mediating role of chemistry learning burnout and chemistry learning flow, Chem. Educ. Res. Pract., 23, 408–421.
- Gutentag T., Orner A. and Asterhan C. S. C., (2022), Classroom discussion practices in online remote secondary school settings during COVID-19, Comput. Hum. Behavi., 132, 107250.
- Gylfason T. and Zoega G., (2003), Education, Social Equality and Economic Growth: A View of the Landscape, CESifo Econ. Stud., 49, 557–579.
- Hajovsky D. B., Oyen K. A., Chesnut S. R. and Curtin S. J., (2020), Teacher–student relationship quality and math achievement: the mediating role of teacher self-efficacy, Psychol. Sch., 57, 111–134.
- Harackiewicz J. M., Durik A. M., Barron K. E., Linnenbrink-Garcia L. and Tauer J. M., (2008), The role of achievement goals in the development of interest: reciprocal relations between achievement goals, interest, and performance, J. Educ. Psychol., 100, 105.
- Hattie J., (2009), A synthesis of over 800 meta-analyses relating to achievement, New York: Routledge.
- Hayes A. F., (2015), An index and test of linear moderated mediation, Multivar. Behav. Res., 50, 1–22.
- Hayes A. F., (2017), Introduction to mediation, moderation, and conditional process analysis: A regression-based approach, New York: Guilford Publications.
- Heinze A., Reiss K. and Franziska R., (2005), Mathematics achievement and interest in mathematics from a differential perspective, ZDM, 37, 212–220.
- Hidi S. and Renninger K. A., (2006), The four-phase model of interest development, Educ. Psychol., 41, 111–127.
- Hu L. T. and Bentler P. M., (1999), Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives, Struct. Eq. Model.: Multidiscip. J., 6, 1–55.
- Hübner N., Spengler M., Nagengast B., Borghans L., Schils T. and Trautwein U., (2022), When Academic Achievement (Also) Reflects Personality: Using the Personality-Achievement Saturation Hypothesis (PASH) to Explain Differential Associations Between Achievement Measures and Personality Traits, J. Educ. Psychol., 114, 326–345.
- Hughes J. N., Wu J.-Y., Kwok O.-M., Villarreal V. and Johnson A. Y., (2012), Indirect effects of child reports of teacher–student relationship on achievement, J. Educ. Psychol., 104, 350.
- Hui Y. K., Li C., Qian S. and Kwok L. F., (2019), Learning engagement via promoting situational interest in a blended learning environment, J. Comput. High. Educ., 31, 408–425.
- Jerome E. M., Hamre B. K. and Pianta R. C., (2009), Teacher–Child Relationships from Kindergarten to Sixth Grade: Early Childhood Predictors of Teacher-perceived Conflict and Closeness, Soc. Dev., 18, 915–945.
- Kahu E., Nelson K. and Picton C., (2017), Student interest as a key driver of engagement for first year students, Stud. Success, 8, 55–66.
- Kashy-Rosenbaum G., Kaplan O. and Israel-Cohen Y., (2018), Predicting academic achievement by class-level emotions and perceived homeroom teachers' emotional support, Psychol. Sch., 55, 770–782.
- Köller O., Baumert J. and Schnabel K., (2001), Does interest matter? The relationship between academic interest and achievement in mathematics, J. Res. Math. Educ., 32, 448–470.
- Košir K. and Tement S., (2014), Teacher–student relationship and academic achievement: a cross-lagged longitudinal study on three different age groups, Eur. J. Psychol. Educ., 29, 409–428.
- Lau R. S. and Cheung G. W., (2012), Estimating and Comparing Specific Mediation Effects in Complex Latent Variable Models, Org. Res. Methods, 15, 3–16.
- Lee J.-S., (2014), The relationship between student engagement and academic performance: Is it a myth or reality?, J. Educ. Res., 107, 177–185.
- Lee M. H., Johanson R. E. and Tsai C. C., (2008), Exploring Taiwanese high school students' conceptions of and approaches to learning science through a structural equation modeling analysis, Sci. Educ., 92, 191–220.
- Lei H., Cui Y. and Zhou W., (2018), Relationships between student engagement and academic achievement: a meta-analysis, Soc. Behav. Pers.: Int. J., 46, 517–528.
- Li H., (2023), Perceived teacher–student relationship and growth mindset as predictors of student engagement in foreign student engagement in foreign language learning: the mediating role of foreign language enjoyment, Front. Psychol., 14, 1177223.
- Li X., Bergin C. and Olsen A. A., (2022), Positive teacher—student relationships may lead to better teaching, Learn. Instruct., 80, 101581.
- Liu X., (2024), Effect of teacher–student relationship on academic engagement: the mediating roles of perceived social support and academic pressure, Front. Psychol., 15, 1331667.
- Longobardi C., Prino L. E., Marengo D. and Settanni M., (2016), Student–teacher relationships as a protective factor for school adjustment during the transition from middle to high school, Front. Psychol., 7, 1988.
- Luo Q., Chen L., Yu D. and Zhang K., (2023), The mediating role of learning engagement between self-efficacy and academic achievement among Chinese college students, Psychol. Res. Behav. Manage., 1533–1543.
- Ma L., Du X., Hau K.-T. and Liu J., (2018), The association between teacher–student relationship and academic achievement in Chinese EFL context: a serial multiple mediation model, Educ. Psychol., 38, 687–707.
- Marsh H. W., Hau K.-T. and Wen Z., (2004), In search of golden rules: Comment on hypothesis-testing approaches to setting cutoff values for fit indexes and dangers in overgeneralizing Hu and Bentler's (1999) findings, Struct. Eq. Model., 11, 320–341.
- Mason B. A., Hajovsky D. B., McCune L. A. and Turek J. J., (2017), Conflict, closeness, and academic skills: A longitudinal examination of the teacher–student relationship, Sch. Psychol. Rev., 46, 177–189.
- Mazer J. P., (2013), Validity of the student interest and engagement scales: Associations with student learning outcomes, Commun. Stud., 64, 125–140.
- McDonald R. P., (1999), Test theory: A unified treatment, Psychology Press.
- McKinnon R. D. and Blair C., (2019), Bidirectional relations among executive function, teacher–child relationships, and early reading and math achievement: a cross-lagged panel analysis, Early Childhood Res. Q., 46, 152–165.
- McNeish D., (2018), Thanks coefficient alpha, we’ll take it from here, Psychol. Methods, 23, 412.
- Morek M., Heller V. and Kinalzik N., (2023), Engaging ‘silent’ students in classroom discussions: a micro-analytic view on teachers’ embodied enactments of cold-calling practices, Lang. Educ., 37, 731–749.
- Nie Y. Y. and Lau S., (2009), Complementary roles of care and behavioral control in classroom management: The self-determination theory perspective, Contemp. Educ. Psychol., 34, 185–194.
- Nystrand M., Wu L. L., Gamoran A., Zeiser S. and Long D. A., (2003), Questions in Time: Investigating the Structure and Dynamics of Unfolding Classroom Discourse, Disc. Processes, 35, 135–198.
- Opperman E. A., Benson L. E. and Milhausen R. R., (2013), Confirmatory factor analysis of the female sexual function index, J. Sex Res., 50, 29–36.
- Pekrun R., Goetz T., Titz W. and Perry R. P., (2002), Academic emotions in students' self-regulated learning and achievement: A program of qualitative and quantitative research, Educ. Psychol., 37, 91–105.
- Peng X., Sun X. and He Z., (2022), Influence Mechanism of Teacher Support and Parent Support on the Academic Achievement of Secondary Vocational Students, Front. Psychol., 13, 863740.
- Penner J. C., Alemán M. J., Anampa-Guzmán A., Berkowitz A. L. and Nematollahi S., (2024), “A safe, non-judgmental space where I can really challenge myself:” learner experiences in a virtual, case-based diagnostic reasoning conference for students, Med. Educ. Online, 29, 2414559.
- Peters G.-J. Y., (2014), The alpha and the omega of scale reliability and validity: Why and how to abandon Cronbach's alpha and the route towards more comprehensive assessment of scale quality, Eur. Health Psychol., 16, 56–69.
- Pianta R. C. and Nimetz S. L., (1991), Relationships between children and teachers: Associations with classroom and home behavior, J. Appl. Dev. Psychol., 12, 379–393.
- Pianta R. C. and Stuhlman M. W., (2004), Teacher–Child Relationships and Children's Success in the First Years of School, Sch. Psychol. Rev., 33, 444–458.
- Pianta R. C., Steinberg M. S. and Rollins K. B., (1995), The first two years of school: Teacher–child relationships and deflections in children's classroom adjustment, Dev. Psychopathol., 7, 295–312.
- Pieratt J. R., (2011), Teacher-student relationships in project based learning: A case study of high tech middle north county, The Claremont Graduate University.
- Podsakoff P. M., MacKenzie S. B., Lee J.-Y. and Podsakoff N. P., (2003), Common method biases in behavioral research: a critical review of the literature and recommended remedies, J. Appl. Psychol., 88, 879–903.
- Polly D., Wang C., Martin C., Lambert R., Pugalee D. and Middleton C., (2018), The Influence of Mathematics Professional Development, School-Level, and Teacher-Level Variables on Primary Students' Mathematics Achievement, Early Childhood Educ. J., 46, 31–45.
- Praetorius A. K., Klieme E., Herbert B. and Pinger P., (2018), Generic dimensions of teaching quality: the German framework of Three Basic Dimensions, Zdm-Math. Educ., 50, 407–426.
- Preacher K. J. and Hayes A. F., (2008), Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models, Behav. Res. Methods, 40, 879–891.
- Qian H. F., Wei N. N., Zhang R. L., Tang Y. F. and Luo G. X., (2023), Social support and continuing motivation in chemistry: the mediating roles of interest in chemistry and chemistry self-efficacy, Chem. Educ. Res. Pract., 24, 478–493.
- Qian H. F., Huang W. L., He Q. M., Luo S. S. and Chen Q. M., (2024), The relationship between self-handicapping in chemistry and chemistry academic engagement: a moderated mediation model investigation, Chem. Educ. Res. Pract., 25, 920–933.
- Quin D., (2017), Longitudinal and contextual associations between teacher–student relationships and student engagement: A systematic review, Rev. Educ. Res., 87, 345–387.
- Renninger K. A. and Hidi S. E., (2019), The Cambridge handbook of motivation and learning, New York, NY, US: Cambridge University Press, pp. 265–290.
- Rimfeld K., Malanchini M., Krapohl E., Hannigan L. J., Dale P. S. and Plomin R., (2018), The stability of educational achievement across school years is largely explained by genetic factors, npj Sci. Learn., 3, 16.
- Rindermann H., (2008), Relevance of education and intelligence for the political development of nations: Democracy, rule of law and political liberty, Intelligence, 36, 306–322.
- Roorda D. L., Koomen H. M., Spilt J. L. and Oort F. J., (2011), The influence of affective teacher–student relationships on students’ school engagement and achievement: a meta-analytic approach, Rev. Educ. Res., 81, 493–529.
- Roorda D. L., Jak S., Zee M., Oort F. J. and Koomen H. M., (2017), Affective teacher–student relationships and students' engagement and achievement: a meta-analytic update and test of the mediating role of engagement, Sch. Psychol. Rev., 46, 239–261.
- Rotgans J. I. and Schmidt H. G., (2017), Interest development: arousing situational interest affects the growth trajectory of individual interest, Contemp. Educ. Psychol., 49, 175–184.
- Rucinski C. L., Brown J. L. and Downer J. T., (2018), Teacher–Child Relationships, Classroom Climate, and Children's Social-Emotional and Academic Development, J. Educ. Psychol., 110, 992–1004.
- Ryan R. M. and Deci E. L., (2020), Intrinsic and extrinsic motivation from a self-determination theory perspective: definitions, theory, practices, and future directions, Contemp. Educ. Psychol., 61, 101860.
- Sankar E. and Benjamin A. E. W., (2024), Studying the student's perceptions of engagement and problem-solving skills for academic achievement in chemistry at the higher secondary level, Educ. Inf. Technol., 29, 8347–8368.
- Schaufeli W. B., Martinez I. M., Pinto A. M., Salanova M. and Bakker A. B., (2002a), Burnout and engagement in university students: a cross-national study, J. Cross-Cultural Psychol., 33, 464–481.
- Schaufeli W. B., Salanova M., González-Romá V. and Bakker A. B., (2002b), The measurement of engagement and burnout: a two sample confirmatory factor analytic approach, J. Happiness Studies, 3, 71–92.
- Scherrer V., Breit M. and Preckel F., (2025), The stability of students’ academic achievement in school: A meta-analysis of longitudinal studies, Educ. Res. Rev., 48, 100687.
- Schiefele U., (1991), Interest, learning, and motivation, Educ. Psychol., 26, 299–323.
- Sherub G. and Nima G., (2022), The impact of supportive teacher–student relationships on academic performance, Asian J. Adv. Res. Rep., 16, 15–34.
- Sjöström O., Holst D. and Lind S. O., (1999), Validity of a questionnaire survey: the role of non-response and incorrect answers, Acta Odontol. Scand., 57, 242–246.
- Skinner E. A., Kindermann T. A. and Furrer C. J., (2009), A motivational perspective on engagement and disaffection: Conceptualization and assessment of children's behavioral and emotional participation in academic activities in the classroom, Educ. Psychol. Measur., 69, 493–525.
- Sointu E. T., Savolainen H., Lappalainen K. and Lambert M. C., (2017), Longitudinal associations of student–teacher relationships and behavioural and emotional strengths on academic achievement, Educ. Psychol., 37, 457–467.
- Spinath B., (2012), Encyclopedia of Human Behavior (Second Edition), ed. Ramachandran V. S., San Diego: Academic Press, pp. 1–8.
- Stroet K., Opdenakker M.-C. and Minnaert A., (2013), Effects of need supportive teaching on early adolescents’ motivation and engagement: a review of the literature, Educ. Res. Rev., 9, 65–87.
- Sun H. R., Sun W. J., Liu X. Y., Niu M. T. and Liu Y. R., (2025), The relationship between chemistry growth mindset and chemistry academic engagement: a multiple mediation model, Chem. Educ. Res. Pract., 26, 996–1011.
- Taylor A. B., MacKinnon D. P. and Tein J. Y., (2008), Tests of the three-path mediated effect, Org. Res. Methods, 11, 241–269.
- Treagust D. F. and Chittleborough G., (2001), Subject-specific instructional methods and activities, Emerald Group Publishing Limited, pp. 239–267.
- Treagust D., Duit R. and Nieswandt M., (2018), Sources of students’ difficulties in learning Chemistry, Educ. Quim., 11, 228.
- Velayutham S. and Aldridge J. M., (2013), Influence of Psychosocial Classroom Environment on Students' Motivation and Self-Regulation in Science Learning: A Structural Equation Modeling Approach, Res. Sci. Educ., 43, 507–527.
- Verschueren K., (2015), Middle childhood teacher–child relationships: insights from an attachment perspective and remaining challenges, New Directions Child Adolescent Dev., 2015, 77–91.
- Wei J. Y., Treagust D. F., Mocerino M., Vishnumolakala V. R., Zadnik M. G., Lucey A. D. and Lindsay E. D., (2021), Design and Validation of an Instrument to Measure Students' Interactions and Satisfaction in Undergraduate Chemistry Laboratory Classes, Res. Sci. Educ., 51, 1039–1053.
- Wen Z., Huang B. and Tang D., (2018), Preliminary Work for Modeling Questionnaire Data, J. Psychol. Sci., 41, 204–210.
- Xu Z. and Qi C., (2019), The Relationship between Teacher—student Relationship and Academic Achievement: The Mediating Role of Self-efficacy, EURASIA J. Mathe., Sci. Technol. Educ., 15, em1758.
- Yang L., Cui J. and Zhang Y., (2025), The relationship between medical students’ interest in learning and their ability to solve mathematical problems: the chain-mediating role of teacher–student relationship and self-efficacy, Front. Psychol., 16, 1531262.
- Young S. and Bruce M. A., (2011), Classroom community and student engagement in online courses, J. Online Learn. Teach., 7, 219–230.
- Yu X. Q., Dong Y., Sun C. H., Wang H. F., Yang Y. C. and Yu G. L., (2024), How the teacher–student relationship influences adolescents' english academic performance: a multiple mediator analysis based on control-value theory, Psychol. Schools, 61, 2800–2819.
- Zhou D., Du X., Hau K.-T., Luo H., Feng P. and Liu J., (2020), Teacher–student relationship and mathematical problem-solving ability: mediating roles of self-efficacy and mathematical anxiety, Educ. Psychol., 40, 473–489.
| This journal is © The Royal Society of Chemistry 2026 |