How to Improve the Quality of Teacher–Child Interaction? A Teacher-Level Randomized Controlled Trial in Shanghai

Abstract

Purpose

This research aimed to determine the impact of the Kindergarten-Based Instructional Research Model (KIRM) to promote the quality of teacher–child interaction (KIRM-TCI), a professional development training intervention focused on improving the quality of teacher–student interaction.

Design/Approach/Methods

The study used a teacher-level randomized controlled trial. Fifty-one pre-K teachers across four kindergartens were randomly assigned to either the KIRM-TCI condition—integrating CLASS (Classroom Assessment Scoring System) competencies with structured peer dialogue and practitioner inquiry—or a control group. Blinded observers conducted CLASS assessments before and after the intervention.

Findings

Within the scope of the research, it was indicated that the project had a positive effect on the quality of emotional and instructional support in teacher–child interactions (TCIs). The findings substantiate KIRM-TCI's efficacy in improving interactions, attributable to its synergistic integration of evidence-based practice with reflective professional development.

Originality/Value

The experimental approach empowers theoretical contributions, offering concrete evidence to inform both the refinement of the KIRM framework and broader discussions about transformative approaches to teacher professional learning.

Keywords

Kindergarten-Based Instructional Research Model professional development randomized controlled trial Shanghai teacher–child interaction

Introduction

Existing studies have demonstrated that the quality of teacher–child interaction (TCI) is closely related to early childhood development, which can predict children's early academic competence, social behavior, and self-management to a certain extent (Hamre et al., 2014; Hatfield et al., 2021; Pianta et al., 2017). Notably, investigations of TCI quality using the Classroom Assessment Scoring System (CLASS) framework in China have highlighted a critical need for Chinese teachers to improve the quality of the instructional support (INS) they provide in order to promote optimal development in children (Han, 2015; Hu et al., 2016; Jiang et al., 2019; Luo et al., 2024).

To date, numerous empirical studies have proposed various training models for early childhood teachers aimed at enhancing the quality of TCI (e.g., Hamre et al., 2012; Pianta et al., 2014). Among the existing teacher training models in China, the Kindergarten-Based Instructional Research Model (KIRM) is a widely-used approach for improving teachers’ professional skills in kindergartens and advancing their professional development. The current study leverages the advantages of the KIRM framework in teacher-led research and collaborative learning to design the KIRM-TCI (teacher–child interaction) project, which aims to enhance the quality of TCI. Building upon KIRM's dual strengths in teacher-led inquiry and collaborative learning, we systematically address the critical challenge of enhancing interaction quality in early childhood education—a well-documented predictor of developmental outcomes. Uniquely, our intervention bridges theoretical frameworks with classroom practice through the targeted application of the CLASS, a widely validated observational tool for assessing the quality of TCI (Pianta, 2012).

The research design advances current scholarship across two key dimensions. First, it establishes a novel methodological approach by integrating KIRM's reflective cycles with CLASS's empirically validated domains (emotional support (ES), classroom organization (CO), and INS). Second, it responds to persistent calls for experimental evidence in professional development research, employing rigorous intervention methods to examine both the efficacy of KIRM-TCI and its underlying mechanisms of change. Third, our work addresses a critical gap in the KIRM literature in China, which has previously relied primarily on theoretical discussions rather than systematic implementation data.

Through its dual focus on reflective practice and the development of practical skills, the KIRM-TCI intervention demonstrates how collaborative learning models can be effectively translated into measurable improvements in classroom interactions. The study's experimental approach lends particular strength to its theoretical contributions, offering concrete evidence to inform both the refinement of the KIRM framework and broader discussions about transformative approaches to teacher professional learning.

Literature Review

TCI Quality

TCIs are essential to early childhood development. Defined as the “daily back-and-forth exchanges that teachers and children have with one another throughout each day, including those that are social and instructional in nature” (Hamre et al., 2012), these interactions occur in diverse settings, such as structured group activities, daily routines, and play corners. The mutual effect or influence can take place in organized group teaching activities as well as the daily activities of all children and teachers, such as those related to daily living, corner activities, and so on. TCI is the primary element affecting children's development in terms of their study and life in kindergarten (La Paro et al., 2004).

The quality of these interactions is influenced by various factors, including teachers’ expertise, educational backgrounds, and classroom dynamics, such as the ratio of teachers to students. However, research shows that these factors do not always predict interaction quality effectively, illustrating the complexity of the issue (Baron et al., 2023; Early et al., 2006; Justice et al., 2008).

In response to this complexity, researchers have increasingly focused on how teachers’ beliefs and their professional knowledge and skills impact the quality of TCI. Studies indicate that teachers who believe strongly in the importance of interaction and who engage in ongoing professional development tend to create high-quality interactions that benefit children's development (Downer et al., 2011; Hamre et al., 2012; Šašić et al., 2018). As noted by Chinese scholar Ye (2001), teachers’ professional growth is rooted in the accumulation of practical knowledge, which includes recognizing and responding to children's emotional, social, and perceptual cues in specific educational contexts.

Accordingly, any improvement of TCI requires teachers to accumulate practical knowledge about it, including precise recognition of clues about perception, society, and children's emotions within specific educational contexts. They should also be able to respond to these cues appropriately, using all kinds of cognitive and ES and restrictions.

Therefore, to improve the quality of TCI, it is essential to focus on enhancing teachers’ beliefs and skills related to these interactions.

CLASS offers a structured framework for evaluating and improving TCIs by emphasizing the teaching and learning process (Pianta et al., 2008a). It consists of three domains and ten dimensions: ES, which includes positive climate, negative climate, teacher sensitivity, and regard for student perspectives; CO, which encompasses behavior management, productivity, and instructional learning formats; and INS, which focuses on concept development, quality of feedback, and language modeling (Pianta et al., 2008a). Each dimension is scored on a 7-point Likert scale, categorized into low (1, 2), middle (3, 4, 5), and high (6, 7) ranges. The cross-cultural validity of CLASS has been demonstrated in various countries (Hamre et al., 2014; La Paro et al., 2004), including a validation study conducted in China (Hu et al., 2016).

Research has shown that training programs based on the CLASS framework lead to substantial improvements in both TCIs and child developmental outcomes (Downer et al., 2011; Hamre et al., 2012; Pianta & Hofkens, 2023; Pianta et al., 2008b). This underscores the significant value of structured approaches to enhancing interaction quality in early childhood education.

Existing Models of Teacher Professional Development for Improving the Quality of TCI

Currently, the two main approaches to professional development for teachers are training workshops and individualized tutorials. In training workshops, teachers gain essential knowledge for their professional growth through collaborative learning and hands-on teaching practice. Research has shown that specialized training courses can significantly improve the quality of TCI and support teachers’ professional development (e.g., Lee & Sung, 2023; Scott-Little et al., 2011). One notable study by Hamre et al. (2012) implemented a 14-week course focused on effective TCIs with 220 participating teachers. The results indicate that those who completed the course provided better emotional and INS to young children compared to a control group. However, not all training programs are effective; for example, Neuman and Cunningham (2009) found that a three-credit course aimed at enhancing literacy and language instruction did not result in any meaningful changes in teachers’ practices during later courses. This has led to a trend of moving away from traditional workshop models in favor of more sophisticated approaches to professional development (Varol et al., 2012).

Individualized tutorials have also become a popular method of supporting teachers’ professional growth (Schachter, 2015). Research has demonstrated that these tutorials can positively impact both teachers and students (e.g., Powell et al., 2010). As individual tutoring requires considerable time and resources, researchers have been working to improve this approach over the past decade by incorporating online tutoring technologies. For instance, the My Teaching Partner project has provided one-on-one tutoring focused on TCIs, enhancing both teachers’ skills and interaction quality (e.g., Downer et al., 2011; Powell et al., 2010). In a study comparing two groups of preschool teachers—one receiving online tutorials through the MTP model and the other simply watching exemplary cases online—those who engaged in the tutorials showed significantly increased improvements in the quality of the TCI (Pianta et al., 2008b). Moreover, there is a growing trend toward using video-based coaching within individualized tutorials to further develop teachers’ interactional skills (Hu et al., 2023).

Despite the effectiveness of individualized tutorials in supporting professional development, researchers continue to seek ways to optimize these models. For example, Pianta et al. (2014) found that merging training courses with a tutorial system led by expert mentors yielded more significant improvements in the quality of TCI. This underscores the importance of exploring innovative formats for professional development in education (Schachter, 2015).

As the field evolves, there is a clear demand for innovative models that effectively bridge this divide and equip teachers with the resources they need to succeed.

Teacher Professional Development in China and the KIRM

The KIRM has become a prominent approach to teacher professional development in China. Rooted in the cultural tradition that values education and respects teachers, KIRM positions teachers as researchers of their own practice. Through regular teaching and research activities, teachers collaborate to tackle practical challenges related to teaching and learning, fostering a culture of continuous improvement (Li, 2010). Since its introduction in 2006, KIRM has been widely implemented across China, with over 700 kindergartens participating in its initial rollout (Li, 2010). The Chinese government further emphasized the significance of strengthening kindergarten-based and regional instructional research in 2018, cementing KIRM's role in early childhood education (Liu, 2019).

The KIRM model is based on the belief that teachers’ professional development is most effective when closely connected to their daily practice (Li, 2010). By engaging in collaborative research and reflection, teachers can gain a deeper understanding of their teaching methods and their effects on children's learning. This model aligns with China's broader educational philosophy, which prioritizes blending theory and practice in teacher training.

A key innovation of the KIRM model is its focus on teacher-led research. Unlike traditional professional development models that depend on external experts, KIRM empowers teachers to take the initiative in identifying and addressing issues within their own classrooms. This not only enhances their professional skills but also cultivates a sense of ownership and accountability for their growth.

However, despite its widespread adoption, more empirical studies are needed to assess the effectiveness of the KIRM model. While anecdotal evidence and case studies suggest positive outcomes, rigorous research is essential to validate these findings and highlight areas for improvement. Future studies should concentrate on measuring KIRM's impact on both the quality of TCI and children's developmental outcomes.

Overall, the quality of TCI plays a crucial role in early childhood development, and improving this quality demands a multifaceted approach to teacher professional development. While existing models, such as training workshops and individualized tutorials, show promise, they possess limitations that underscore the need for innovative strategies. The KIRM model, with its emphasis on teacher-led research and collaborative learning, represents a promising alternative that aligns with the cultural and educational context of China. Nevertheless, further research is essential to fully comprehend its effects and refine its implementation. By merging theory with practice and empowering teachers in the course of their professional development, the KIRM model has the potential to significantly enhance the quality of TCI and ultimately improve outcomes for young children.

Methods

Participants

A cluster randomized controlled trial was conducted with four publicly funded kindergartens in Shanghai, China, selected through stratified random sampling from districts serving middle-income communities. These institutions maintained affordable tuition rates (approximately $25/month per child), representative of typical early childhood education settings in metropolitan China.

The participating kindergartens were randomly allocated to either the training group (n = 2) or waitlist control group (n = 2), with cluster randomization ensuring baseline equivalence in institutional characteristics. From the initial pool of 52 certified early childhood educators, 51 completed the 1-year study period (attrition rate = 1.92%). Written informed consent was obtained from all participating educators prior to their enrollment in the study. The final sample comprised:

Training group: 30 teachers (age range = 22–37 years old; M = 27.47, SD = 4.28) participating in a professional development program focused on enhancing the quality of TCI through collaborative instructional research activities.

Control group: 21 teachers (age range = 24–36 years old; M = 28.90, SD = 3.69) continuing standard teaching practices.

It has been well-documented in the literature that academic qualifications and duration of teaching experience are primary predictors of TCI quality (Denny et al., 2012). In the current investigation, these parameters were systematically compared across experimental groups during baseline assessments. No statistically significant intergroup disparities were observed in either pedagogical tenure (p = .063) or educational credentials (p = .094). In terms of the highest education qualifications of the teachers in the training group, 33.3% had a junior college degree and 66.7% had academic credentials at or above a bachelor's degree, while all the teachers in the control group had academic credentials at or above a bachelor's degree. Concerning the duration of teaching experience, 53.3% of the teachers in the training group had no more than five years’ experience and 46.7% had more than five years; for the control group, these figures were 28.6% and 71.4%, respectively.

The KIRM-TCI Project

The KIRM-TCI project implemented kindergarten-based instructional research activities tailored to the specific practices of individual kindergartens, guided by high-level TCI behaviors assessed using the CLASS framework. Specifically, researchers and educators established a community of practice within these kindergarten-based instructional research activities. This collaborative environment focused on analyzing the interactive behaviors of teachers and children in real-time contexts and implemented an intervention structured around three cyclical modules: review and application, learning and discussion, and practice and reflection.

In the review and application module and the learning and discussion module, the project provided professional support to facilitate teachers’ active inquiry and peer dialogue. The practice and reflection module emphasized teachers’ proactive exploration of their interactions with children.

The primary goal of the project was to enhance teacher behaviors and improve the quality of TCI through these methodologies (see Figure 1).

Figure 1.

Theoretical Framework for Improving the Quality of Teacher–Child Interaction.

The KIRM-TCI project developed training courses based on the CLASS behavioral framework for high-quality TCIs, specifically focusing on Pre-K settings. The training covers three domains and ten dimensions of CLASS, along with the theoretical foundations and essential skills for high-quality TCIs. The three domains are ES, CO, and INS.

ES includes four dimensions: Positive Climate (PC), Negative Climate (NC), Teacher Sensitivity (TS), and Regard for Student Perspectives (RSP).

CO encompasses three dimensions: Behavior Management (BM), Productivity (PD), and Instructional Learning Formats (ILF).

INS consists of three dimensions: Concept Development (CD), Quality of Feedback (QF), and Language Modeling (LM).

The training program consists of nine sessions in total (see Table 1).

Table 1.

Summary Table of the KIRM-TCI Training Sessions Based on CLASS Framework Dimensions.

Subject	Contents
Session 1: Introduction of the KIRM-TCI Project	(1) The introduction of the KIRM-TCI project (2) The establishment of the community of practice
Session 2: Teacher–Child Interaction	The significance of TCI to children's development
Session 3: Emotional Support Domain	(1) Positive Climate: The emotional connection, mutual respect, and enjoyment in the communication between the teacher and the children and among the children by verbal and nonverbal interaction (2) Negative Climate: The overall level of observable negativity in the classroom (3) Teacher Sensitivity: Teacher's awareness of and responsivity to children's academic and emotional needs (4) Regard for Students’ Perspectives: The degree to which the teacher's interaction with children and classroom activities place emphasis on children's interests, motivations, and points of view
Sessions 4–5: Classroom Organization Domain	(1) Behavior Management: The teacher's ability to monitor, prevent, and redirect behaviors (2) Productivity: The teacher's ability to manage instructional time and routines and to arrange activities for children (3) Instructional Learning Format: The ways in which the teacher maximizes children's interest, engagement, and ability to learn from lessons and activities.
Sessions 6–7: Instructional Support Domain	(1) Concept Development: The teacher's use of instructional activities to promote children's high-order thinking skills and cognitive levels and the teacher's focus on understanding instead of rigid instruction (2) Quality of Feedback: The degree to which the teacher gives feedback that promotes learning and understanding and encourages continued participation (3) Language Modeling: The quality and frequency of the teacher's use of language-stimulation and language-facilitation techniques
Sessions 8–9: Practical Assessment by the CLASS	Practical assessment of the quality of the TCI by the CLASS

Each session is structured around three modules: review and application, learning and discussion, and practice and reflection.

The primary aim of the review and application module is to reinforce knowledge related to TCI. This module is divided into three parts:

Researchers review the core concepts of CLASS.

Participants analyze practical video recordings: Teachers in training provide key points for discussion, record their TCIs on video, and reflect on these recordings while engaging in discussions informed by CLASS.

In the concluding section, teachers analyze and reflect on specific teaching cases of their interactions with children, documenting their observations and insights in relation to the key points reviewed.

The learning and discussion module aims to introduce new knowledge to teachers and is divided into two parts:

Researchers present detailed explanations of the CLASS framework.

High-quality video examples of TCIs are provided, enabling learners to identify specific behaviors indicative of high-level TCIs.

In the practice and reflection module, teachers in training complete writing tasks and video recordings of their TCI cases, allowing them to compare their recorded behaviors with real-time interactions and engage in reflective analysis.

This structured approach aims to deepen teachers’ understanding and enhance the overall quality of TCI, ultimately benefiting their professional development.

Research Procedures and Fidelity

During the experimental phase of the KIRM-TCI project, high-quality TCI training was provided to educators in the training group. This training consisted of nine sessions for each kindergarten, conducted every two to three weeks. Following each session, teachers were required to complete practice and reflection tasks. Each training session lasted three hours, divided into two modules: 1.5 hr dedicated to reviewing and applying concepts through learning and discussion, followed by 1.5 hr focused on practice and reflection, for a total of 27 hr of training. To maintain consistency in the intervention's implementation across all participating kindergartens, each training session was facilitated by the same trainer.

Of the 30 teachers in the training group, 21 (70%) attended all nine sessions. Additionally, five teachers (16.7%) attended eight sessions, and four teachers (13.3%) attended seven sessions. A satisfaction survey indicated that 22.9% of the teachers perceived the intervention as highly beneficial in enhancing their TCI skills, while 58.3% found it helpful, and 18.8% considered it somewhat beneficial.

The teachers in the control group also participated in KIRM training, focusing on observing children's behavior and other related objectives, which was conducted every two to three weeks.

These structured procedures, along with an emphasis on fidelity in training delivery, were designed to ensure the intervention's effectiveness, thereby fostering improved TCIs within the participating kindergartens.

Measures and Reliability Control of Raters

The Classroom Assessment Scoring System (CLASS) was utilized to evaluate the quality of TCI through standardized observational protocols. Five postgraduate students specializing in early childhood education underwent an intensive training program supervised by a certified CLASS observer, achieving mastery criteria (κ > .85) on practice coding prior to data collection.

Longitudinal video recordings of TCIs were obtained from 51 participating teachers’ classrooms during typical instructional periods (three-hour morning sessions) at two time points: baseline (pre-intervention) and post-KIRM-TCI implementation. Each observation consisted of four 20-min cycles systematically capturing diverse instructional contexts: whole-group instruction, free-choice activities, routine transitions, and individualized interactions, in alignment with CLASS assessment guidelines.

A double-blind coding procedure was implemented to ensure reliability. Trained assistants formed three coding teams, each responsible for 30–40 classrooms. All four observation cycles per classroom were independently scored by two raters using the CLASS 7-point metric (1 = low to 7 = high). Inter-rater reliability coefficients reached satisfactory levels (mean agreement = 88.41%, range = 84.2%–95%, SD = 4.01), exceeding conventional benchmarks for observational research. To mitigate expectancy bias, pre-test and post-test videos from both experimental and control groups were randomized and anonymized prior to coding.

Demographic variables, including teacher age, years of teaching experience, educational attainment, and professional development history, were documented through structured questionnaires administered during on-site visits.

Data Analysis

We wanted to establish whether the KIRM-TCI project had any effect on the quality of TCI as measured by CLASS. Therefore, analysis of variance (ANOVA), analysis of covariance (ANCOVA), and regression analysis were used in the analysis.

According to ANCOVA, the CLASS scores each teacher obtained in the three domains and 10 dimensions before the intervention were used as covariates to control and adjust the differences between the groups before the intervention (data shown in Table 2). In the pre-analysis of the covariance, we found that the interaction between the independent variable “grouping” (training group and control group) and the original score of the covariates was not significant (as shown in Table 2, the P-values are .09, .074, and .198, respectively); this is in accordance with the requirement of the same slope for each covariate group, such that a covariance analysis could be performed. The results of the covariance analysis are shown in Table 2. The intervention effects in the three domains of CLASS are as follows:

“ES pretest” has no significant effect on “ES post-test” (p = .46 > .05), which indicates that it is unnecessary to exclude the interference from the original score when comparing the differences between the two groups in the ES post-test. Therefore, a one-way ANOVA of the ES post-test was carried out.

“CO pretest” has a significant effect on “CO post-test” (p = .028 < .05), whereas the “grouping” variable has a non-significant effect on “CO post-test” (p = .084 > .05). Therefore, the covariate was retained and pretest interference was excluded.

“IS pretest” has a significant effect on “IS post-test” (p = .000), and the “grouping” variable also has a significant effect on the “IS post-test” (p = .028 < .05). Therefore, the covariate was retained and pretest interference was excluded.

Table 2.

Summary Table of Teachers’ CLASS Scores in Two Groups Before and After the Project Intervention.

	Pretest		Posttest		FCA	EPI		Test of Intervention Effect
Domain/Dimension	TGN1 ＝ 30M (SD)	CGN2 ＝ 21M (SD)	TGN1 ＝ 30M (SD)	CGN2 ＝ 21M (SD)	RSS	GEP	P	F	Sig.	Partial Eta Square^a
ES	5.12 (.29)	5.12 (.33)	5.50 (.05)	5.23 (.06)	.09	.002	.46	10.44^b	.002	.18
PC	4.89 (.47)	4.83 (.50)	5.26 (.51)	4.89 (.44)	.43	.009	.026	7.37	.009	.13
NC	6.84 (.25)	6.96 (.12)	6.94 (.14)	7.00 (.00)	.20	.21	.002	1.61	.21	.032
TS^c	4.42 (.54)	4.28 (.52)	4.94 (.40)	4.48 (.52)	.023	.001	.85
RSP	4.32 (.54)	4.74 (.54)	4.85 (.47)	4.52 (.52)	.44	.013	.032	6.59	.013	.12
CO	4.61 (.45)	4.58 (.44)	4.99 (.50)	4.76 (.42)	.074	.084	.028	3.12	.084	.061
BM	4.74 (.54)	4.80 (.43)	5.12 (.40)	4.84 (.57)	.13	.038	.26	4.27^b	.044	.087
PD	4.76 (.61)	4.73 (.56)	5.20 (.75)	4.88 (.63)	.79	.088	.00	3.03	.088	.059
ILF	4.32 (.46)	4.22 (.57)	4.66 (.57)	4.56 (.35)	.86	.51	.40	0.56^b	.46	.009
IS	3.58 (.57)	3.33 (.64)	3.87 (.54)	3.44 (.57)	.198	.028	.00	5.17	.028	.097
CD	3.23 (.74)	3.18 (.76)	3.71 (.73)	3.48 (.67)	.097	.27	.004	1.24	.27	.025
QF^c	3.99 (.67)	3.61 (.79)	4.18 (.57)	3.60 (.72)	.038	.015	.00
LM	3.51 (.54)	3.21 (.55)	3.73 (.51)	3.24 (.56)	.34	.014	.007	6.55	.014	.12

Note. FCA = forecast analysis of covariance; EPI = whether or not need to exclude pretest interference; RSS = whether the rates of slope are the same; GEP = group effect on post-test P; P = P value of post-test effect on pretest; TG = training group; CG = control group.

“Partial Eta Square” can reflect the magnitude of the effect caused by the experimental treatment, or how much of the data variation is caused by the experimental treatment. The criterion of judgment: 0.01 means small research effect, 0.06 means medium research effect, 0.14 means large research effect.

F of the “ES” domain is an ANOVA value and all the other Fs are ANCOVA values.

The assumption of homogeneity of regression slopes—a prerequisite for analysis of covariance (ANCOVA)—was violated in both the TS (p = .023) and QF (p = .038) dimensions. Consequently, ANCOVA was abandoned and regression analysis was adopted as the analytical approach.

For further exploration of which dimensions of CLASS the intervention project played a role in, we compared differences in the quality of TCI before and after the project across the different dimensions of CLASS. We collected data for all dimensions to determine whether they met the hypothetical requirement of covariance. The results show that the other eight dimensions met the requirements of the same slope in the covariance analysis for each group, except for the teacher sensitivity and feedback quality dimensions. We carried out covariance analysis on teachers’ post-test scores across these eight dimensions, and stepwise regression analysis on the other two dimensions (the sample size of 51 participants was in line with the reasonable sample size for regression) (Van Voorhis & Morgan, 2007).

Results

The KIRM-TCI Project's Intervention Effects on the Three Domains of CLASS

First of all, the project had a significant effect on the improvement of ES quality in TCI through the intervention. The results shown in Table 2 indicate that the “grouping” variable had a significant effect on teachers’ scores in ES after the intervention (F = 10.44, p = .002 < .01), and reached an extremely high level. The partial Eta squared is 0.18, which indicates that the research effect was undoubtedly large. The scores of the training group in ES were evidently higher than those of the control group and 18% of the variations between the two groups were caused by the intervention project.

Secondly, the project had a nonsignificant effect on the improvement of class organization quality in TCI. The results show that there was no significant difference in class organization post-test between the two groups (p = .084 > .05). The partial Eta squared is 0.061. Accordingly, the intervention had a medium-level effect. There was no significant difference in teachers’ scores of the two groups in class organization after the intervention; only 6.1% of the variations between the two groups were caused by experimental factors.

Lastly, the project had a significant effect on the improvement of the quality of INS in TCIs. The results show a significant difference in IS post-tests between the two groups (p = .028 < .05); the partial Eta squared is 0.097, so the research effect is also medium. This indicates that the training group scored significantly higher than the control group did in INS, as 9.7% of the variations between the two groups were caused by experimental factors. Therefore, the intervention project played a relatively significant role in the improvement of INS quality in TCI.

Effects of the KIRM-TCI Project Intervention

For further exploration into which dimension of CLASS the project played a role in, we compared the quality of TCI before and after the KIRM project across the different dimensions of CLASS. In the covariance analysis, after controlling for the pretest, a significant difference was found between the training group and the control group in the four dimensions of the CLASS post-test scores, including positive climate, regard for student perspectives, behavior management, and language modeling (see Table 2 for specific data). The effect, according to variance analysis, was 0.13, 0.12, 0.087, and 0.12, respectively, i.e., medium- or upper-level. This shows that 8.7%–13.3% of the variations in the above four dimensions were caused by the intervention. Therefore, the intervention project had a significant effect on positive climate, regard for student perspectives, behavior management, and language modeling in the quality of TCI.

The researchers also found that after controlling for pretest scores, there was no significant difference in CLASS post-test scores between the training group and the control group in the four dimensions of negative climate, productivity, instructional learning format, and concept development (see Table 2 for concrete data). This indicates that the intervention project had a non-significant effect on negative climate, productivity, instructional learning format, and concept development in the quality of TCI.

In the stepwise regression analysis, we used grouping, teachers’ background information (including their highest education qualification, full-time or non-full-time status, academic background related to the profession, professional title, and teaching experience), and CLASS pretest scores as independent variables to investigate the factors influencing teacher sensitivity (TS) and feedback quality (QF).

As illustrated in Table 3, only grouping variables entered the regression equation in the regression analysis; this affected post-test scores in the teacher sensitivity dimension (TS post-test). The researchers tried to determine whether there was a collinearity problem with the independent variables, but found no such problem. The results of the regression analysis show that the grouping variables had a significant effect on scores in the teacher sensitivity dimension. This indicates that the intervention project played a significant role in the improvement of teacher sensitivity in TCI. In the regression analysis of the post-test scores (QF post-test), which affected the quality of feedback dimension (QF post-test), the grouping variables and the pre-test scores of feedback quality (QF pre-test) entered the regression equation. The results further prove that there was no collinearity problem with the independent variables. Therefore, the results of the regression analysis show that the grouping variables and QF pretest had a synergistic effect on QF post-test scores. By combining the two methods, 45.3% of the variation in QF post-test scores can be predicted. Therefore, the intervention project played a significant role in improving feedback quality in TCI.

Table 3.

Regression Analysis Summary of Teacher Sensitivity and Feedback Quality After Intervention.

	Intervention Variable	B	SE	β	R	R ²	Adjustment R²	F
TS	Grouping	0.45	0.13	0.45	0.448	.201	0.18	12.303**
QF	Pretest	0.51	0.10	0.54	0.673	0.453	0.430	19.843**
QF	Grouping	0.39	0.15	0.28	0.673	0.453	0.430	19.843**

Note. **: p<0.01.

Discussion

Potential of the KIRM-TCI Project to Improve the Quality of TCI

The most significant effect observed in this research was the improvement in the quality of ES, followed by the enhancement in INS. However, CO did not show any significant improvement. Specifically, the KIRM-TCI project resulted in improvements in both emotional and INS within the classroom. After the intervention, teachers in the training group demonstrated varying levels of improvement in the quality of their TCI across the three domains when compared to the control group. Notably, ES showed the most substantial enhancement, while the effect on CO lacked significance.

One key factor contributing to these results was the relatively low baseline quality of emotional and INS observed during the pretest of the KIRM-KCT. Specifically, the original scores for INS primarily fell within the 3- to 4-point range, indicating considerable room for improvement.

These findings are largely consistent with those of a study by Hamre et al. (2012), suggesting a systematic extension of research conducted in China.

Interestingly, this study also produced results that differ from those of some related studies. In examining specific subscales within the three domains, increases were noted in ES (PC—Positive Climate, TS—Teacher Sensitivity, RSP—Regard for Student Perspectives), CO (BM—Behavior Management), and INS (QF—Quality of Feedback, LM—Language Modeling). While another study found that teachers who completed a training course provided more effective emotional and instructional interactions than their control group counterparts, their increases were noted in different areas: ES (RSP), CO (ILF—Instructional Learning Formats), and INS (CD—Concept Development, QF, LM) (Hamre et al., 2012).

Culturally, it is important to consider that within the context of Chinese education, teachers typically exhibit less emotional expressiveness and prioritize instructional guidance (Pu, 2009). This cultural factor results in lower scores in ES while yielding relatively higher scores in INS. In contrast, Hamre and her colleagues noted that American teachers often score lower in INS but relatively higher in ES. The differential improvement patterns reflect deep-seated cultural pedagogies in Chinese early education. Confucian traditions emphasizing strict teacher authority and reverence for teacher knowledge naturally predispose teachers toward instructional guidance rather than emotional expressiveness. Consequently, the KIRM-TCI project resulted in a significant increase in scores for ES but not in INS (CD).

Additionally, the projects varied in their approach to CO. The KIRM-TCI project improved scores related to behavior management (BM), whereas the research by Hamre et al. (2012) focused on enhancing instructional learning formats (ILFs). This distinction arises from the emphasis on discipline in Chinese education, wherein behavior management practices that clarify behavioral expectations and redirect misbehavior resonate more strongly with teachers and are easier to implement. However, with regard to ILFs, since Chinese preschool education emphasizes effective facilitation and a diverse array of modalities and materials, the KIRM-TCI project did not increase ILF scores because these were already at a high level.

The KIRM-TCI Project as a Promising Teacher Development Approach

This study is grounded in the tradition of Chinese preschool education and fully leverages the unique characteristics of the KIRM model, which specifically focuses on enhancing teachers’ practical knowledge and literacy through kindergarten practices.

The KIRM-TCI project unfolded in six steps: learning, imitating, practicing, reflecting, sharing, and associating. In the project's learning and discussion module, teachers first viewed videos demonstrating high-quality TCI. They then reflected on these cases and engaged in discussions with their peers. While the high-quality interaction videos served as learning tools, all other cases were contributed by the teachers from the two participating kindergartens. This case-based discussion not only fostered connections among teachers but also encouraged targeted reflections, facilitating active inquiry. Such a process enabled teachers to learn from exemplary interactions. During the practice and reflection module, teachers imitated these high-quality cases before applying the interaction skills in real settings, followed by reflective practice on their imitations and implementations. In the review and application module, teachers shared their own interaction videos and reflections within their professional community. This collaborative process helped transform unconscious patterns of interaction into conscious cognitive schemas.

To encourage ongoing reflection, the KIRM-TCI project prioritized peer dialogue. According to Simons et al. (2003), collaborative meaning-making is crucial within professional communities where teachers learn from one another. Peer dialogue fosters a supportive learning environment, aligning with Lave's (1988) view of knowledge as distributed among individuals. In the KRIM model, cooperative relationships among teachers are reinforced, making it easier for them to collaborate effectively. In the context of the KIRM-TCI project, teacher-to-teacher cooperation primarily occurred through peer dialogue, where teachers engaged in discussions that deepened their understanding of TCIs. This shared inquiry led to varying degrees of improvement in their interactions.

The KIRM-TCI project was designed in response to teachers’ ongoing needs for professional development and training, emphasizing supportive activities aimed at enhancing the quality of TCI. Such activities require a sustained commitment (Early et al., 2006; Justice et al., 2008; Schachter, 2015). Teacher professional development has often faced challenges of disconnection and ineffectiveness (Ball & Cohen, 1999), largely due to insufficient focus on systematically linking suitable experiences to practical application (Pianta et al., 2008b).

In the KIRM-TCI project, a community of practice was formed by inviting early childhood teachers from two kindergartens, with systematic and progressive support tailored to their needs. In the learning and discussion module, teachers deepened their understanding of high-quality TCIs through active inquiry and peer dialogue. In the practice and reflection module, teachers applied what they learned in each session, recording their interactions for further discussion in subsequent training sessions. The review and application module emphasized peer dialogue, enabling teachers to analyze each other's videos, receive individualized feedback, and reflect on their strengths and limitations. This iterative process resulted in improved quality of TCI across three cycles: learning and discussion, practice and reflection, and review and application.

In conclusion, the KIRM-TCI project effectively integrated the advantages of training workshops with individualized tutoring, facilitating knowledge acquisition, case sharing, and expert guidance. Research by Pianta et al. (2014) found that integrating curriculum training with expert instruction yields more significant improvements in the quality of TCI. Maintaining continuity is vital for enabling teachers to learn from their experiences, and the KIRM-TCI project, spanning two semesters, fulfilled this requirement. Liu (2012) noted that sporadic and disjointed training often leads to diminished effects, underscoring the importance of ongoing research and training for effective practice improvement.

In teacher training, it is essential to stimulate reflective intelligence and situational responsiveness. The KIRM-TCI project, grounded in kindergarten practices and focusing on teachers’ acquisition of practical knowledge, represents a valuable option for inspiring wisdom among teachers.

Limitations and Implications

In addition to the limitations noted earlier, one major limitation of the current work is that the effect of the KIRM-TCI project can only be reflected by changes in the quality of TCI. As is well known, teachers’ professional development is ultimately reflected in children's development. However, it was a pity that the current study did not collect data on children's developmental outcomes. Future work should include assessments of children's competencies.

Besides, the current study involved a great deal of assessment work on the quality of TCI. The project lasted for one year and some participants quit, so only 51 teachers could ultimately attend. To enhance the persuasiveness of the research, future studies should expand the participant pool to a sufficient level, as a sample of four kindergartens is hardly representative of preschool education in Shanghai as a whole.

Furthermore, while the current work used the improvement of TCI quality to demonstrate the efficacy of the project, self-reporting by teachers is also an important way to achieve this. Thus, future work should add self-reporting by teachers to prove the effectiveness of the project.

There is one last limitation worth noting: Taking the project as an intervention, the current study focused on the extent of intervention and the mechanisms for change rather than the entire teaching and classroom context. Some potentially relevant variables, such as class sizes and teaching environments in different kindergartens, have not been adequately considered. Future work should pay more attention to the entire teaching and classroom context, in which additional factors may also improve teachers’ development.

Conclusion

This study investigated the KIRM-TCI, a professional development program grounded in Chinese preschool education traditions. The results reveal significant improvements in the quality of TCI, with ES demonstrating the strongest gains, followed by measurable progress in INS. The KIRM-TCI framework innovatively integrates structured training workshops with individualized coaching, implementing its phased design through evidence-based strategies targeting interaction quality. The program's phased structure addressed practitioners’ developmental needs while maintaining cultural alignment with Chinese educational contexts. These findings advance theoretical understanding of professional learning mechanisms by demonstrating how iterative practice-reflection cycles reshape interaction dynamics. The experimental evidence provides crucial guidance for refining culturally responsive professional development models and informs broader discussions of transformative teacher learning methodologies.

Footnotes

ORCID iDs

Chunhong Han

Jingying Wang

Si Chen

Linhui Li

Li Zhang

Ethical Considerations

This study received ethical approval from the Ethics Review Committee of East China Normal University (Approval no. HR751-2022). Written informed consent was obtained from all participating teachers and legal guardians of the children involved. Data collection procedures, including classroom observations and teacher information acquisition, were conducted strictly in accordance with the approved ethical protocols. Participants were fully informed about the study's purpose, procedures, and their right to withdraw at any stage without prejudice. Confidentiality of all collected data was maintained through anonymization and secure storage practices in compliance with institutional ethical guidelines.

Author Contributions

Chunhong Han contributed to project administration, conceptualization, methodology, formal analysis, and original draft preparation. Jing Zhou contributed to conceptualization, methodology, and supervision. Jingying Wang contributed to validation, resources, and visualization. Si Chen contributed to methodology and statistical analysis. Li Zhang contributed to investigation, data curation, software, and review and editing of the manuscript. Linhui Li contributed to investigation and data curation.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Education Science Planning Key Project of the Ministry of Education, China (Project No.: [DHA230403]) under the research program “Support Strategies for Young Children's Social-Emotional Learning (SEL)”.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

Ball

D. L.

Cohen

D. K.

(1999). Developing practice, developing practitioners: Toward a practice-based theory of professional education. In Darling-Hammond

Sykes

(Eds.), Teaching as the learning profession: Handbook of policy and practice (pp. 3–32). Jossey-Bass.

Baron

Linberg

Lehrl

(2023). Interaction quality in German early childcare settings: Investigating the domains of CLASS toddler and the associations with structural characteristics. Early Child Development and Care, 193(13–14), 1485–1502. https://doi.org/10.1080/03004430.2023.2256997

Denny

J. H.

Hallam

Homer

(2012). A multi-instrument examination of preschool classroom quality and the relationship between program, classroom, and teacher characteristics. Early Education & Development, 23(5), 678–696. https://doi.org/10.1080/10409289.2011.588041

Downer

J. T.

Pianta

R. C.

Fan

Hamre

B. K.

Mashburn

Justice

(2011). Effects of web-mediated teacher professional development on the language and literacy skills of children enrolled in prekindergarten programs. NHSA Dialog, 14(4), 189–212. https://doi.org/10.1080/15240754.2011.613129

Early

D. M.

Bryant

D. M.

Pianta

R. C.

Clifford

R. M.

Burchinal

M. R.

Ritchie

Howes

Barbarin

(2006). Are teachers’ education, major, and credentials related to classroom quality and children’s academic gains in pre-kindergarten? Early Childhood Research Quarterly, 21(2), 174–195. https://doi.org/10.1016/j.ecresq.2006.04.004

Hamre

Hatfield

Pianta

Jamil

(2014). Evidence for general and domain-specific elements of teacher–child interactions: Associations with preschool children’s development. Child Development, 85(3), 1257–1274. https://doi.org/10.1111/cdev.12184

Hamre

B. K.

Pianta

R. C.

Burchinal

LoCasale-Crouch

Downer

J. T.

Howes

LaParo

Scott-Little

(2012). A course on effective teacher–child interactions: Effects on teacher beliefs, knowledge, and observed practice. American Educational Research Journal, 49(1), 88–123. https://doi.org/10.3102/0002831211434596

Han

. (2015). A study on the quality of teacher-child interaction in Shanghai early education settings [in Chinese] [Doctoral dissertation, East China Normal University].

Hatfield

B. E.

Finders

J. K.

Zandbergen

D. L.

Lewis

(2021). Associations between consistent and high-quality teacher–child interactions and preschool children’s self-regulation and activity in the stress response system. Early Education and Development, 33(7), 1222–1236. https://doi.org/10.1080/10409289.2021.1961198

10.

Fan

Yang

(2016). Applicability of the classroom assessment scoring system in Chinese preschools based on psychometric evidence. Early Education and Development, 27(5), 714–734. https://doi.org/10.1080/10409289.2016.1113069

11.

Guan

LoCasale-Crouch

Song

Dou

Chen

Huang

Meng

Wang

Zhang

(2023). Effects of using video-based coaching to promote preservice teachers’ interactional skills in Chinese preschool classrooms. Early Childhood Research Quarterly, 65, 284–294. https://doi.org/10.1016/j.ecresq.2023.07.002

12.

Jiang

Guo

Lyu

(2019). Meta-analysis of the teacher–child interaction study from the perspective of CLASS—Based on the results of the teacher–child interaction quality assessment of 892 teachers in 14 provinces and cities of China [in Chinese]. Studies in Preschool Education, 4, 32–44. https://doi.org/10.13861/j.cnki.sece.2019.04.003

13.

Justice

L. M.

Mashburn

A. J.

Hamre

B. K.

Pianta

R. C.

(2008). Quality of language and literacy instruction in preschool classrooms serving at-risk pupils. Early Childhood Research Quarterly, 23(1), 51–68. https://doi.org/10.1016/j.ecresq.2007.09.004

14.

La Paro

K. M.

Pianta

R. C.

Stuhlman

(2004). The classroom assessment scoring system: Findings from the prekindergarten year. The Elementary School Journal, 104(5), 409–426. https://doi.org/10.1086/499760

15.

Lave

(1988). Cognition in practice: Mind, mathematics and culture in everyday life. Cambridge University Press.

16.

Lee

J. Y.

Sung

(2023). Effects of in-service programs on childcare teachers’ interaction quality: A meta-analysis. Teaching and Teacher Education, 124, 104017. https://doi.org/10.1016/j.tate.2023.104017

17.

(2010). Current situation and challenges of kindergarten-based instructional research [in Chinese]. Early Childhood Education, (2), 8–11.

18.

Liu

(2012). National research on the kindergarten quality-by comparison with the situation in 1992 [in Chinese]. Studies in Preschool Education, (2), 3–10. https://doi.org/10.13861/j.cnki.sece.2012.02.005

19.

Liu

(2019, January 21). Early childhood education instructional research is facing transformation and upgrading [in Chinese]. China Education News, 2.

20.

Luo

Qiu

Lyu

Liu

(2024). Variation in the quality of teacher–child interactions in Chinese toddler classrooms. Early Education and Development, 35(8), 1847–1861. https://doi.org/10.1080/10409289.2024.2350876

21.

Neuman

S. B.

Cunningham

(2009). The impact of professional development and coaching on early language and literacy instructional practices. American Educational Research Journal, 46(2), 532–566. https://doi.org/10.3102/0002831208328088

22.

Pianta

Hamre

Downer

Burchinal

Williford

Locasale-Crouch

Howes

La Paro

Scott-Little

(2017). Early childhood professional development: Coaching and coursework effects on indicators of children’s school readiness. Early Education and Development, 28(8), 956–975. https://doi.org/10.1080/10409289.2017.1319783

23.

Pianta

R. C.

(2012). Implementing observation protocols: Lessons for K-12 education from the field of early childhood. Center for American Progress.

24.

Pianta

R. C.

Burchinal

Jamil

F. M.

Sabol

Grimm

Hamre

B. K.

Downer

LoCasale-Crouch

Howes

(2014). A cross-lag analysis of longitudinal associations between preschool teachers’ instructional support identification skills and observed behavior. Early Childhood Research Quarterly, 29(2), 144–154. https://doi.org/10.1016/j.ecresq.2013.11.006

25.

Pianta

R. C.

Hofkens

(2023). Defining early education quality using CLASS-observed teacher–student interaction. Frontiers in Psychology, 14. https://doi.org/10.3389/fpsyg.2023.1110419

26.

Pianta

R. C.

La Paro

K. M.

Hamre

B. K.

(2008a). Classroom assessment scoring system™: Manual K-3. Paul H. Brookes Publishing Co.

27.

Pianta

R. C.

Mashburn

A. J.

Downer

J. T.

Hamre

B. K.

Justice

(2008b). Effects of web-mediated professional development resources on teacher–child interactions in pre-kindergarten classrooms. Early Childhood Research Quarterly, 23(4), 431–451. https://doi.org/10.1016/j.ecresq.2008.02.001

28.

Powell

D. R.

Diamond

K. E.

Burchinal

M. R.

Koehler

M. J.

(2010). Effects of an early literacy professional development intervention on head start teachers and children. Journal of Educational Psychology, 102(2), 299–312. https://doi.org/10.1037/a0017763

29.

(2009). A comparative research about interaction between kindergarten teacher and children in learning and non-learning activity [in Chinese]. Studies in Preschool Education, (3), 40–43. http://doi.org/10.13861/j.cnki.sece.2009.03.008

30.

Šašić

SŠ

Šimunić

Ivković

Ključe

(2018). The correlation of perceptions of professional roles and teacher beliefs with the quality of teacher interaction. Journal of Research in Science, Mathematics and Technology Education, 1(2), 207–227. https://doi.org/10.31756/jrsmte.125

31.

Schachter

R. E.

(2015). An analytic study of the professional development research in early childhood education. Early Education & Development, 26(8), 1–29. https://doi.org/10.1080/10409289.2015.1009335

32.

Scott-Little

La Paro

K. M.

Thomason

A. C.

Pianta

R. C.

Hamre

Downer

Burchinal

Howes

(2011). Implementation of a course focused on language and literacy within teacher–child interactions: Instructor and student perspectives across three institutions of higher education. Journal of Early Childhood Teacher Education, 32(3), 200–224. https://doi.org/10.1080/10901027.2011.594489

33.

Simons

Kushner

Jones

James

(2003). From evidence-based practice to practice-based evidence: The idea of situated generalisation. Research Papers in Education, 18(4), 347–364. https://doi.org/10.1080/0267152032000176855

34.

VanVoorhis

C. R. W.

Morgan

B. L.

(2007). Understanding power and rules of thumb for determining sample sizes. Tutorials in Quantitative Methods for Psychology, 3(2), 43–50. https://doi.org/10.20982/tqmp.03.2.p043

35.

Varol

Farran

D. C.

Bilbrey

Vorhaus

E. A.

Hofer

K. G.

(2012). Improving mathematics instruction for early childhood teachers: Professional development components that work. NHSA Dialog, 15(1), 24–40. https://doi.org/10.1080/15240754.2011.636488

36.

(2001). A new probe into the role of teachers and teacher development (pp. 215–216). Education Science Press.