Abstract
Curriculum ideological and political education (CIP) is a major reform in Chinese higher education that seeks to embed civic, moral, and ideological aims into disciplinary courses. While existing CIP research has largely focused on conceptual clarification and curriculum design, relatively little is known about how students experience CIP in everyday interactions with teachers and how these experiences relate to perceived teaching effectiveness. Drawing on international literature on teacher–student interaction, student engagement, and value-oriented education, this study examines the relationships among embedded teacher–student interaction, students’ interaction experiences, and perceived teaching effectiveness in CIP-related courses. Survey data were collected from 451 undergraduate students in three 4-year universities in Hefei, Anhui Province, China. Embedded interaction was operationalised across four learning spaces (classroom teaching, tutoring and questioning, communication and discussion, and practical guidance), interaction experience was captured through perceived accessibility, friendliness, timeliness, and helpfulness, and teaching effectiveness was measured by perceived knowledge gains, value-related concept enhancement, professional identity, and overall satisfaction. Structural equation modelling was used to test a mediational model. The results indicate that embedded teacher–student interaction is positively associated with perceived teaching effectiveness and that students’ interaction experiences are strongly and positively related to teaching effectiveness. Moreover, interaction experience partially mediates the relationship between embedded interaction and teaching effectiveness. These findings suggest that the success of value-oriented curricula such as CIP depends not only on whether value-related content is embedded in disciplinary courses, but also on whether teachers interact with students in ways that are experienced as accessible, respectful, timely, and helpful.
Plain Language Summary
This study explores how teacher-student interactions (both in and out of class) affect the effectiveness of "ideological and political education" in university courses. Through surveys, we found: 1. Interactive learning works. Different types of teacher-student interactions (such as discussions, feedback, or informal conversations) all significantly improve learning outcomes. 2. Experience matters. Students who have positive interactions with teachers are more likely to accept and benefit from the course content. 3. Teachers should focus on engagement. Simply having good teaching materials isn’t enough; teachers need to create meaningful interactions that enhance students' psychological experience.
Keywords
Introduction
In recent decades, universities have been expected not only to provide advanced disciplinary knowledge, but also to cultivate graduates’ civic responsibility, moral judgement, and professional character. Many higher education systems have therefore moved towards integrating value-oriented goals into mainstream curricula instead of relying solely on stand-alone civics or ethics courses (Allen, 2017; Amerstorfer & Freiin von Münster-Kistner, 2021; Anderson et al., 2025). In China, this development is reflected in the policy agenda of “moral education and cultivating people” as a core mission of schooling and higher education, and in the nationwide reform of Curriculum Ideological and Political Education (CIP; Baker, 2021; Bosio, 2023; Breetzke et al., 2024). CIP requires all university courses, including specialised disciplinary courses, to embed ideological, civic, and moral objectives, so that value education is woven into everyday teaching and learning rather than confined to a limited number of ideological theory classes (Cantoni et al., 2017; Chen & Khoso, 2025).
Against this policy background, a rapidly growing body of Chinese-language scholarship has examined CIP from multiple angles. Existing studies have clarified the conceptual connotations of CIP, interpreted policy documents, and explored how ideological and political elements can be identified and integrated into different disciplinary curricula (Cheng, 2022; Cheng et al., 2022; Chino et al., 2014). Several contributions discuss course design, teaching strategies, and evaluation frameworks for CIP implementation in specific subjects or institutions (Daigneault, 2024). At the same time, international research on higher education has produced a rich literature on teacher–student interaction (Du & Chen, 2022), student engagement (Dwyer, 2017), and value-oriented or civic education in universities, drawing on theories such as Tinto’s student integration model (Gao et al., 2022), Astin’s involvement theory (Kahu & Nelson, 2018), self-determination theory (Liu & He, 2022), and work on caring teacher–student relationships and formative feedback. These strands of research, however, have rarely been systematically connected in the context of CIP.
Taken together, the existing literature reveals several gaps. First, CIP research has focused predominantly on the design and content of value integration, while paying comparatively little attention to how students experience CIP in their day-to-day interactions with teachers and how these experiences relate to perceived teaching effectiveness at the course level. Second, although teacher–student interaction is widely recognised as central to student learning and development, few empirical studies in the CIP context distinguish between embedded interaction across multiple learning spaces and students’ interaction experiences in terms of accessibility, responsiveness, and perceived support. Third, there is a lack of quantitative, theory-driven analyses that test mediating mechanisms linking embedded interaction, interaction experience, and perceived teaching effectiveness in value-oriented curricula. Finally, CIP is seldom framed in ways that make its mechanisms intelligible and relevant to international debates on value-oriented curriculum reform.
This article addresses these gaps by examining how embedded teacher–student interaction and students’ interaction experiences are associated with the perceived teaching effectiveness of CIP-related courses in Chinese universities. Conceptually, we distinguish three constructs: embedded teacher–student interaction (“interactive approach”), defined as interaction distributed across classroom teaching, tutoring and questioning, communication and discussion about academic or value-related issues, and guidance in internships and social practice; interaction experience, defined as students’ perceptions of how easy it is to contact teachers and how friendly, timely, and helpful their responses are; and teaching effect, defined as students’ perceived knowledge gains, enhancement of value-related concepts, professional identity, and overall satisfaction with CIP-related teaching. Empirically, we use survey data from 451 undergraduate students in three 4-year universities in Hefei, Anhui Province, and apply structural equation modelling to test three hypotheses: that embedded interaction is positively associated with perceived teaching effectiveness, that interaction experience is positively associated with perceived teaching effectiveness, and that interaction experience mediates the relationship between embedded interaction and perceived teaching effectiveness.
The study makes three main contributions. Theoretically, it connects the CIP reform to internationally recognised frameworks on teacher–student interaction, student engagement, and value-oriented education, and proposes a conceptual model that distinguishes structural opportunities for embedded interaction from students’ experiential perceptions, highlighting a mediating psychological mechanism. Empirically, it provides quantitative evidence from a large sample of Chinese undergraduates, validates a three-factor measurement model of embedded interaction, interaction experience, and teaching effect, and estimates their interrelationships using structural equation modelling. Practically, the findings underline that the success of value-oriented curricula such as CIP depends not only on how value-related content is integrated into disciplinary courses, but also on whether teachers are able to interact with students in ways that are experienced as accessible, respectful, timely, and helpful—insights that are pertinent for institutions seeking to strengthen the civic and moral dimensions of higher education.
Theoretical Analysis and Research Hypothesis
Embedded Teacher–Student Interaction in Value-Oriented Curricula
A central premise of this study is that curriculum ideological and political education (CIP) can be understood, in comparative terms, as a form of value-oriented or civics-related curriculum that embeds moral and civic goals into disciplinary teaching rather than delivering them only through stand-alone ideology or ethics courses (Chino et al., 2014). In such curricula, teacher–student interaction is a key mechanism through which value-related content is interpreted, negotiated, and connected to students’ academic and life worlds.
Research in higher education has long demonstrated that frequent, meaningful teacher–student interaction is positively associated with students’ academic integration, engagement, and persistence (e.g. Tinto’s theory of student integration; Astin’s theory of student involvement; Meeuwisse et al., 2010; Minott, 2022). Studies on classroom discourse, formative feedback, and mentoring likewise highlight that interaction within and beyond the classroom contributes to learning outcomes, satisfaction, and professional identity formation (Moynihan et al., 2015). Building on this literature, we conceptualise “embedded” teacher–student interaction as interaction that is not confined to formal classroom episodes but is distributed across multiple learning spaces, including lectures and seminars, office hours and tutoring, informal communication and discussion, and supervision of internships and social practice (Niemiec & Ryan, 2009; Oeschger et al., 2022).
In the context of CIP, such embedded interaction aligns with the notion of “implicit” education: value-related messages are woven into disciplinary content, examples, and guidance rather than being delivered as separate ideological statements (Cantoni et al., 2017; Chen & Khoso, 2025). When interaction is embedded across curricular and co-curricular settings, students have more opportunities to encounter, question, and internalise value-related content as part of their ordinary learning processes. Accordingly, we expect that higher levels of embedded teacher–student interaction will be associated with more favourable perceptions of CIP-related teaching effectiveness.
On this basis, we propose the following hypothesis:
Interaction Experience and Perceived Teaching Effectiveness
While the frequency and breadth of interaction are important, prior research consistently shows that the quality of interaction, as experienced by students, is at least as critical. Theories of self-determination emphasise the importance of relatedness, autonomy support, and competence support for motivation and learning (Pike et al., 2012). Studies on caring teacher–student relationships and academic support suggest that when students perceive teachers as accessible, respectful, responsive, and genuinely helpful, they are more likely to engage, to trust the instructor, and to evaluate the learning experience positively (Simmonds & Dicks, 2018; Stentiford & Koutsouris, 2022).
In CIP-related courses, where value-laden content is embedded in disciplinary teaching, students’ subjective experiences of interaction are likely to be particularly salient (Tan et al., 2024). If interactions around course content, assignments, and practice-based activities are experienced as easy to initiate, timely in response, and helpful for understanding both disciplinary knowledge and its value implications, students may be more open to considering and appropriating the value messages embedded in the curriculum. Conversely, experiences characterised by perceived distance, perfunctory attitudes, or difficulties in contacting teachers may reduce students’ willingness to engage with value-related content and thus weaken perceived teaching effectiveness (Tian et al., 2025).
To shed additional light on the experiential dimension of interaction, we draw a cautious analogy from public administration research on “administrative burden” (Tormey, 2021; Wahlström & Sundberg, 2018). In citizen–state interactions, administrative burden refers to the learning, compliance, and psychological costs that citizens perceive when accessing public services; high perceived burden is associated with lower satisfaction and reduced uptake of service (Mahmudulhassan et al., 2024). We do not equate citizen–state relations with teacher–student relations. Rather, we use this literature heuristically to suggest that when students face high “interaction costs” (e.g. difficulty reaching teachers, delayed responses, or unhelpful feedback), similar psychological mechanisms—such as frustration, anxiety, or withdrawal—may be triggered and may undermine their overall evaluation of the educational experience. This analogy is used only to illuminate the role of experiential costs and benefits; the two relational contexts remain fundamentally distinct.
In sum, existing work on motivation, engagement, and supportive relationships leads us to expect that students’ positive interaction experiences—defined here in terms of accessibility, friendliness, timeliness, and perceived helpfulness—will be strongly related to their perceptions of CIP teaching effectiveness. We therefore formulate the second hypothesis:
The Mediating Role of Interaction Experience
The third element of our framework concerns the process through which embedded interaction may influence teaching effectiveness. Prior studies in higher education and educational psychology increasingly view student experience and engagement as mediating mechanisms linking instructional practices to learning outcomes (Minott, 2020; Yuan et al., 2024). Structural and behavioural features of teaching (e.g. opportunities for contact, interactive formats, availability of tutoring) shape how students experience the learning environment, these subjective experiences, in turn, influence motivation, sense of belonging, and evaluations of teaching (You et al., 2022).
In the specific context of CIP, this mediational perspective is particularly relevant. University students are generally capable of critical and independent thinking, and they may respond differently to value-related content depending on whether they perceive teachers as approachable, respectful, and trustworthy (Zhang & Wang, 2025). The reform of CIP explicitly aims to move beyond traditional, overt forms of ideological instruction by embedding value-related content into the everyday teaching of disciplinary courses. For this implicit approach to be effective, the interaction through which such content is conveyed needs to be experienced as supportive rather than coercive. Positive interaction experiences can strengthen students’ recognition of the teacher, the discipline, and the institution, making them more willing to engage with the value dimensions of the curriculum. Negative experiences, by contrast, may lead to resistance or disengagement, even when the formal structure of embedded interaction is in place.
Against this backdrop, we conceptualise embedded teacher–student interaction (interactive approach) as creating opportunities for contact across multiple settings, and interaction experience as capturing the psychological quality of how these opportunities are lived by students. Teaching effectiveness in CIP-related courses is then expected to depend not only on how often and in how many contexts interaction occurs, but also on whether those interactions are experienced as easy, friendly, timely, and helpful. It follows that interaction experience should mediate, at least in part, the relationship between embedded interaction and perceived teaching effectiveness.
Accordingly, we formulate our third hypothesis:
Research Design and Methodology
Variable Selection and Model Construction
This study focuses on three latent constructs: (a) embedded teacher–student interaction (hereinafter “interactive approach”), (b) students’ experiences of embedded teacher–student interaction (hereinafter “interaction experience”), and (c) the teaching effectiveness of Curriculum Ideological and Political Education (hereinafter “teaching effect”). Consistent with the theoretical framework, interactive approach is specified as the independent variable X, interaction experience as the mediating variable M, and teaching effect as the dependent variable Y.
On the basis of a review of the literature, consultation with experts in higher education and moral education, and exploratory interviews with undergraduate students, we translated the three constructs into measurable indicators while also considering questionnaire length and ease of completion. Each construct was operationalised as a latent factor measured by four observed items, resulting in a 12-item scale.
In line with the argument that the implicit character of curriculum ideological and political education (CIP) makes students’ subjective experience central to how embedded interactions influence outcomes, we posit that interaction experience mediates the association between interactive approach and teaching effect. The structural equation model (SEM) thus specifies paths from interactive approach to interaction experience and teaching effect, as well as from interaction experience to teaching effect. The hypothesised model is presented in Figure 1.
Structural equation model of interactive approach, interactive experience and teaching effect.
The independent variable X (interactive approach) measures the frequency and quality of interaction between students and teachers in a variety of scenarios inside and outside the classroom, including X1 (classroom teaching), X2 (tutoring and questioning), X3 (communicating and discussing), and X4 (practical guidance). The mediating variable M, interactive experience, measures students’ experience of teacher–student interaction in terms of how easy, friendly, timely and useful the interaction is. It specifically includes four items: M1 (easy to contact), M2 (friendly attitude), M3 (timely response), and M4 (helpful). The dependent variable Y, teaching effect, measures the students’ sense of acquisition and satisfaction in the teaching of “curriculum ideological and political education,” including Y1 (knowledge growth), Y2 (concept enhancement), Y3 (professional identity), and Y4 (overall satisfaction).
Measures
All items were rated on a 5-point Likert-type scale ranging from 1 (“strongly disagree”) to 5 (“strongly agree”), with higher scores indicating higher levels of the corresponding construct.
Interactive Approach (X)
Embedded teacher–student interaction was measured with four items tapping interaction in different learning spaces. Illustrative items include statements such as “My teachers actively interact with students during classroom teaching” (X1, classroom teaching), “My teachers are willing to provide tutoring and answer questions when I encounter difficulties in my studies” (X2, tutoring and questioning), “My teachers often communicate and discuss academic or value-related issues with students” (X3, communicating and discussing), and “My teachers provide guidance when I participate in internships, social practice, or other practical activities” (X4, practical guidance). Together these items reflect the extent to which teacher–student interaction is embedded across classroom and out-of-class settings.
Interaction Experience (M)
Students’ subjective experiences of embedded interaction were assessed with four items capturing perceived accessibility and quality of interaction. Example items include “It is easy for me to get in touch with my teachers when I need to” (M1, easy to contact), “My teachers’ attitudes are friendly and respectful when they interact with me” (M2, friendly attitude), “My teachers respond to my questions or requests in a timely manner” (M3, timely response), and “The advice and feedback I receive from my teachers are helpful to my study and development” (M4, helpful). These items were designed to reflect core experiential dimensions highlighted in the literature, such as availability, emotional support, and perceived effectiveness of feedback.
Teaching Effect (Y)
Perceived teaching effectiveness of CIP was measured with four items reflecting cognitive, attitudinal, and evaluative outcomes. Illustrative items are “Through these courses I have gained new knowledge and understanding” (Y1, knowledge growth), “These courses have strengthened my value-related concepts and civic or moral awareness” (Y2, concept enhancement), “These courses have enhanced my sense of identity with my major and future profession” (Y3, professional identity), and “Overall, I am satisfied with the teaching of Curriculum Ideological and Political Education in my programme” (Y4, overall satisfaction).
The items were originally developed in Chinese. The English formulations provided here are approximate translations intended to convey the meaning of the indicators to an international readership; the analyses are based on students’ responses to the original Chinese items. Interaction measures relied on students’ retrospective self-reports, a common approach in higher-education research but one that may be affected by recall bias or subjective interpretation.
Research Context, Sample, and Data Collection
The empirical study was conducted in three 4-year undergraduate universities located in Hefei, Anhui Province, China, all of which had implemented reforms to embed CIP elements into disciplinary courses. These three universities were selected because they had formally implemented institution-wide CIP reforms and granted access for data collection, making them typical cases of CIP integration in Chinese higher education rather than extreme or atypical cases. The participating institutions offer a wide range of majors in humanities and social sciences, science and engineering, and arts and sports, and thus provide a suitable context for examining embedded teacher–student interaction in value-oriented curricula.
The target population comprised full-time undergraduate students from first to fourth year who had taken at least one course in which CIP content was integrated into disciplinary teaching. Using intact class groups as the sampling frame, we distributed 507 questionnaires to students from freshmen to seniors across the three universities and obtained 499 returned questionnaires. After excluding questionnaires with substantial missing data or straight-lining (identical responses to all items), 451 valid responses remained, resulting in an effective response rate of 88.95%. Although the questionnaires were administered during scheduled class sessions for logistical convenience, participation remained voluntary; students were explicitly informed that non-participation would not be reported to instructors. Nonetheless, the possibility of perceived pressure in classroom-based surveys is acknowledged in the limitations.
A short introduction on the first page of the questionnaire informed students that the survey focused on their learning experiences in CIP-related courses, that participation was voluntary and anonymous, and that their responses would be used solely for academic research and would not affect course grades or evaluations. These procedural steps were designed to minimise potential pressure associated with the political sensitivity of CIP and to reduce social desirability concerns. The research aims to clarify the mediational effects of interaction experiences in CIP teaching, which can provide practical insights for optimising CIP instruction and promoting high-quality curriculum development, and these benefits outweigh the minimal risks of social desirability bias. The relatively high valid response rate is largely attributable to administering the survey in organised teaching settings, which ensured that most students present at the time of administration could complete and return the questionnaire immediately. At the same time, the possibility that such in-class administration may have increased social desirability bias is acknowledged in the limitations section.
In addition to demographic information (gender, grade, subject, and political background), the questionnaire asked respondents to provide self-assessments of their academic learning performance and practical performance (e.g. participation in internships, social practice, and extracurricular activities). These self-evaluations served two purposes. First, they allowed us to describe the performance profile of the sample. Second, they provided a rough check against the potential concern that students with particularly outstanding performance might be more willing to participate in the survey, which could bias evaluations of interaction experience and teaching effect.
Descriptive statistics for individual characteristics are presented in Table 1. The distributions of gender, subject area, and political background are broadly consistent with the actual student composition of the three universities. Although first-year students account for a somewhat larger share of the sample than senior students, all four cohorts are represented, and the overall grade distribution is relatively balanced. Regarding self-evaluated performance, respondents who rated their learning or practical performance as “excellent” constitute a minority; most students rated themselves as “good” or “average.” Moreover, self-ratings of learning performance are, on average, higher than self-ratings of practical performance, which aligns with the reality that many programmes place relatively more emphasis on classroom learning than on practice-based activities. Taken together, these patterns suggest that the survey sample has reasonable overall representativeness for the student populations of the participating institutions.
Descriptive Statistics of Individual Characteristics.
Reliability, Validity, and Common Method Bias
To assess the psychometric properties of the measurement model and reduce the risk of error in subsequent analyses, we examined the reliability and validity of the 12-item scale. Cronbach’s alpha for the overall scale is .914, which exceeds the commonly accepted threshold of 0.70 and indicates high internal consistency. At the construct level, composite reliabilities for interactive approach, interaction experience, and teaching effect are .850, .862, and .937 respectively, all above .60, suggesting that the observed indicators consistently reflect their intended latent variables.
The suitability of the data for factor analysis was examined using Bartlett’s test of sphericity and the Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy. Bartlett’s test was statistically significant, and the KMO value was 0.896, exceeding the recommended minimum of 0.80. These results indicate that the correlation matrix is appropriate for factor analysis.
Confirmatory factor analysis (CFA) was conducted with AMOS 24.0 to evaluate the convergent validity of the three latent constructs. As reported in Table 2, the standardised factor loadings of all 12 items range from 0.575 to 0.927, all above the commonly used threshold of 0.50. The composite reliability (CR) values for the three dimensions are higher than .60 (.850 for X, .862 for M, and .937 for Y), and the average variance extracted (AVE) values are above 0.50 (0.588, 0.616, and 0.789, respectively). These indices indicate satisfactory convergent validity and suggest that the latent constructs explain a substantial proportion of variance in their indicators.
Reliability and Validity Analysis.
Because all variables were measured using a single self-report questionnaire, we also examined the potential impact of common method bias. Procedurally, anonymity and confidentiality were emphasised, and the items were arranged to reduce respondents’ ability to infer the hypothesised relationships. Statistically, we performed Harman’s single-factor test by loading all 12 items into an exploratory factor analysis without rotation. The first unrotated factor accounted for 23.56% of the total variance, which is below the 40% benchmark commonly used to indicate serious common method variance. These results suggest that common method bias is unlikely to threaten the validity of the findings in a substantial way.
The subsequent structural equation modelling of the relationships among interactive approach, interaction experience, and teaching effect focuses on the three focal constructs and does not include additional demographic covariates (e.g. gender, grade, subject, political affiliation). The possible implications of not modelling these background factors explicitly are revisited in the limitations section.
Empirical Analysis
Structural Equation Model Test
In order to ensure the fit of the revised model to the data, it is necessary to check the condition of model fitting before the mediational effects are estimated. Based on the theoretical model, a framework of structural equation model is constructed in AMOS 24.0 and imported into the data for calculation. The Chi-square probability value of the overall model fit is .094, which is greater than .05 and cannot reject the original hypothesis (the covariance matrix of the model is equal to the covariance matrix of the sample). This indicates that the covariance matrix of the model and the covariance matrix of the sample can be fitted. The structural equation model of the “interactive approach” and “teaching effect” can be fitted with the actual sample data. The RMSEA value is 0.068, which is significantly less than 0.8. The value of GFI is greater than 0.9 and the AGFI is greater than 0.9, indicating that the model was acceptable. The rest of the test indices meet the evaluation criteria, indicating that the data fit the model well and the model can be used for path estimation (Zhou, 2022). The details of the fitted indices are shown in Table 3.
Fitting Test Results of Structural Equation Model.
Common Method Bias Test
This study used questionnaires to collect data, all of which were obtained by subjects completing their own questionnaires. Therefore, there may be common method biases in the survey. While anonymity, confidentiality and the restriction of the survey to academic research were emphasised in the data collection process to allow for procedural control of the survey, statistical control of the study was also required. To confirm the magnitude of the common method bias in this study, the Harman single-factor test was conducted by using SPSS software to place all 12 items in an exploratory factor analysis and to test the results of the unrotated factor analysis. It was found that the variance interpretation rate of the first common factor was only 23.56% but not more than 40%. Thus, common method bias has little effect on the results of this study.
Hypothesis Test and Mediational Effect Test
The data were processed by using AMOS 24.0 to test the hypothesis. The results of the hypothesis testing of the model were obtained, as shown in Table 4.
Hypothesis Test of Structural Model.
Note. *** represent significance at the 1% level.
Table 4 shows that hypotheses H1, H2, and H3 are valid. There is a strong correlation between “interactive approach” and “interactive experience.”“interactive approach” can not only directly promote “teaching effect,” but also indirectly promote “teaching effect” through the intermediate variable of “interactive experience.”
The results of the mediational effect were obtained by applying the bootstrap method, setting the number of repetitions of random sampling at 5,000 and the 95% confidence interval of the effect. None of the 95% confidence intervals for the effects contain 0, indicating that the mediational effect exists and is partially mediated (Tan et al., 2024). As shown in Table 5.
Mediational Effect Test of Structural Model.
Result Analysis
From Figure 2, The standardised factor loadings of the four indicators (X1–X4) on the latent construct “interactive approach” range from 0.679 to 0.844 and are all statistically significant, indicating that each dimension meaningfully contributes to the overall construct of embedded teacher–student interaction. These results empirically support the conceptualisation of embedded interaction as a multidimensional construct that spans both in-class and out-of-class learning contexts. The standardised path coefficient from interaction experience (M) to teaching effect (Y) is 0.69, which is substantially larger than the direct effect of interactive approach (X) on teaching effect (0.29). This indicates that students’ subjective experiences of accessibility, responsiveness, and perceived helpfulness in interacting with teachers are strongly associated with their perceived teaching effectiveness in CIP-related courses. These results provide empirical support for Hypothesis H2. The interactive approach (X) has a significant positive effect on interaction experience (M) (β = .52), which in turn predicts teaching effect (Y) (β = .69). Together with the direct effect of X on Y (β = .29), these results indicate a partial mediating effect of interaction experience. The indirect effect was estimated at 0.15, accounting for 34.09% of the total effect, and the 95% bootstrap confidence intervals did not include zero. These findings support Hypothesis H3 regarding the mediating role of interaction experience.
Path diagrams of the interaction between variables.
Discussion and Conclusion
Summary of Key Findings
This study set out to examine how embedded teacher–student interaction and students’ interaction experiences relate to the perceived teaching effectiveness of CIP-related courses in Chinese universities. Using structural equation modelling with data from 451 undergraduates, we tested a model in which embedded interaction (interactive approach) predicts teaching effectiveness both directly and indirectly through interaction experience.
Three main findings emerge. First, embedded teacher–student interaction is positively associated with perceived teaching effectiveness, supporting Hypothesis 1. Students who report more frequent and diverse interaction with teachers across classroom teaching, tutoring and questioning, communication and discussion, and practical guidance also tend to report higher levels of perceived knowledge growth, value-related concept enhancement, professional identity, and overall satisfaction with CIP teaching.
Second, interaction experience exhibits a strong positive association with teaching effectiveness, consistent with Hypothesis 2. The standardised path coefficient from interaction experience to teaching effectiveness is substantially larger than the direct effect of embedded interaction, indicating that students’ perceptions of accessibility, friendliness, timeliness, and helpfulness in their interactions with teachers play a central role in shaping their evaluations of CIP-related teaching.
Third, interaction experience partially mediates the relationship between embedded interaction and teaching effectiveness, providing support for Hypothesis 3. Embedded interaction exerts a significant direct effect on teaching effectiveness and an additional indirect effect through interaction experience. The estimated indirect effect accounts for a sizeable proportion of the total effect, and bootstrap confidence intervals confirm the statistical significance of this mediating pathway. These results suggest that embedded interaction contributes to teaching effectiveness not only by increasing opportunities for contact but also, and perhaps more importantly, by fostering positive interaction experiences.
Theoretical Implications
The findings have several implications for theory. First, they extend established work on teacher–student interaction, such as Tinto’s and Astin’s frameworks, by focusing on a value-oriented curriculum context and by distinguishing between structural opportunities for interaction (embedded interaction across multiple settings) and experiential dimensions (students’ interaction experiences). The results indicate that both dimensions matter, but that the experiential dimension is particularly strongly linked to perceived teaching effectiveness.
Second, the study adds to research on self-determination theory and caring teacher–student relationships by operationalising interaction experience in terms of students’ perceived ease of contact, friendliness, timeliness, and helpfulness. The strong predictive power of these perceptions suggests that relatedness and perceived support are crucial not only for academic outcomes but also for how students evaluate value-laden teaching. In value-oriented curricula such as CIP, where students may hold diverse views about civic and moral issues, the relational quality of interaction may influence whether they perceive the curriculum as supportive and meaningful or as distant and unengaging.
Third, the mediating role of interaction experience highlights the importance of psychological mechanisms in the implementation of value-oriented policies at the classroom level. Embedded interaction creates the conditions under which value-related messages can be communicated in a relatively implicit, integrated manner. However, the extent to which these conditions translate into positive evaluations of teaching depends on how interactions are actually experienced by students. This mediational perspective helps bridge structural features of teaching (interaction opportunities) and subjective outcomes (perceived effectiveness), offering a more nuanced understanding of how value-oriented curricula function in practice.
Implications for Value-Oriented Curricula and CIP Practice
Beyond theoretical considerations, the findings carry practical implications for CIP implementation and for value-oriented curricula more broadly. At the level of instructional design, the results suggest that simply embedding value-related content into disciplinary courses is insufficient. To enhance perceived teaching effectiveness, instructors need to engage with students through diverse interaction channels—classroom discussion, individual tutoring, informal communication, and practice-based guidance—and to ensure that these interactions are experienced as accessible, respectful, timely, and helpful.
The four dimensions of embedded interaction examined in this study reflect different ways in which teachers can accompany students through their learning trajectories. Classroom interactions provide opportunities to connect disciplinary concepts with civic or moral issues; tutoring and questioning allow for individualised clarification and support; communication and discussion beyond formal class time enable deeper exploration of value-laden topics; and guidance in internships and social practice settings helps students relate values to real-world contexts. Institutions seeking to realise the aims of CIP or similar reforms may therefore wish to support instructors in creating and sustaining these multiple interaction spaces.
At the institutional and policy levels, the findings underscore the importance of organisational conditions that facilitate high-quality interaction. Factors such as class size, staff–student ratios, teaching workloads, availability of office hours, and support for practice-oriented learning can all influence teachers’ capacity to interact with students in a sustained and caring manner. If large classes, heavy workloads, or limited institutional recognition of interaction-related work constrain teachers’ ability to respond promptly and helpfully to students, embedded curricula may fall short of their intended effects, even when value-related content is formally integrated into course syllabi.
For policy makers and institutional leaders, this implies that monitoring and supporting CIP or other value-oriented initiatives should go beyond checking whether curricula contain explicit references to value-related aims. It is also necessary to attend to the everyday interactional processes through which these aims are communicated and negotiated, and to create conditions under which teachers can invest time and emotional energy in building constructive relationships with students.
Limitations and Directions for Future Research
Several limitations of this study should be acknowledged. First, the research design is cross-sectional, which constrains our ability to make strong causal claims. Although the structural equation model is theoretically grounded and the directionality of paths is justified by prior research, longitudinal or experimental designs would be needed to establish causal relationships between embedded interaction, interaction experience, and teaching effectiveness.
Second, all variables were measured through self-report questionnaires completed by students. While we employed procedural remedies (anonymity, confidentiality, careful item ordering) and statistical tests (Harman’s single-factor test) to reduce and assess common method bias, the reliance on a single data source and on subjective perceptions remains a limitation. Future studies could incorporate complementary data such as classroom observations, digital traces of communication (e.g. emails or online messaging), and interviews with both students and teachers to triangulate findings and capture the richness of interaction processes.
Third, the sample was drawn from three 4-year universities in one city in China that have actively implemented CIP reforms. Although the sample is reasonably diverse in terms of disciplines and student characteristics, the findings may not be generalisable to all Chinese institutions, let alone to other countries. Comparative studies across regions, institution types, and national contexts would help clarify the extent to which the mechanisms identified here are specific to CIP or applicable to other value-oriented curricula.
Fourth, the structural model focuses on student-level constructs and does not explicitly incorporate contextual variables such as class size, typical teaching methods, or faculty workload. As the reviewer rightly noted, these factors are likely to shape both interaction opportunities and experiences. Future research could adopt multi-level modelling approaches that nest students within classes, programmes, or institutions and include contextual indicators to examine how organisational conditions moderate the relationships identified in this study.
Despite these limitations, the present study contributes to international discussions on value-oriented curriculum reform by using CIP in Chinese universities as a case to explore how embedded teacher–student interaction and interaction experience relate to perceived teaching effectiveness. The findings suggest that in value-oriented higher education, it is not only what content is embedded in the curriculum that matters, but also how teachers and students interact around that content, and how those interactions are experienced. Future work building on these insights can further illuminate the complex interplay between policy, pedagogy, and student experience in the pursuit of civic and moral aims in higher education.
Footnotes
Ethical Considerations
In accordance with our institution’s regulations, this type of anonymous questionnaire survey requires no ethical approval as it poses no foreseeable distress or harm to participants, maintains strict confidentiality, and aligns with APA Ethical Principles Section 8.05.
Consent to Participate
The short introduction on the first page of the questionnaire clearly informed participants that their participation was voluntary, that they could withdraw at any time, and that clicking the submit button would signify their informed consent to take part in the study.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Department of Education of Anhui Province(Project No. 2024jyxm0336), the Project of Hainan Provincial Philosophy and Social Sciences Planning Project (Project No. HNSK (ZC) 24-130), Hainan Higher Education Institution Teaching Reform Research Foundation (Project No. Hnjg2025-66) and Hefei Normal University (Project No. 2023jyxm12, No. 2024jyxm06, No. 2024jyxm15).
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
Data are available upon request.