Development of Chinese college students’ perception teacher differential behavior scale and its reliability and validity test

Abstract

Objective

To compile a scale of Chinese college students’ perception of teachers’ differential behavior and to provide a reference for college students to establish correct life values, promote college students’ physical and mental health, and reduce teachers’ differential treatment.

Methods

Open-ended questionnaires and expert interviews were used to conduct interviews and correspondence with 58 college students, ten psychologists, and six psychologists to form an initial questionnaire. Then, the scale’s exploratory factor analysis, confirmatory factor analysis, and reliability and validity test were conducted on 7053 college students from 18 universities in 6 provinces (municipalities directly under the Central Government).

Results

The Chinese college students’ perception of teachers’ differential behavior scale has two dimensions: teacher prejudice and preference. Each dimension includes three aspects: emotional feedback, behavior orientation, and opportunity privilege, and each aspect have a total of 4 items. The consistency test coefficients of each dimension and each factor of the prepared scale are all above 0.7, and the split-half reliability is above 0.6. Confirmatory factor analysis shows that the six-factor structural model fits well (χ²/df = 4.287, RMSEA = 0.066, CFI = 0.950, TLI = 0.919). Using the generalized anxiety disorder scale and the patient health questionaire-9items as empirical criteria, each factor in the scale demonstrated significant correlations with both the GAD scale and the patient health questionaire-9items.

Conclusions

The Chinese college students’ perception of teachers’ differential behavior scale has a two-dimensional six-factor structure and has good reliability and validity. It can be used as an effective tool to measure Chinese college students’ perceived teacher differential behavior.

Introduction

Suicide has emerged as the second leading cause of death among Chinese adolescents [1]. The primary factors contributing to adolescent suicide include a lack of internalized values and external environmental pressures [2, 3]. During the phase of steady personality maturation, young college students are particularly susceptible to the influences of campus life [4]. Teachers’ behavioral orientation, particularly their differential treatment, constitutes a fundamental factor contributing to behavioral deviations in adolescents [5]. Differential teacher behavior, also known as unfair treatment, predominantly involves varying emotional or behavioral responses exhibited by teachers in their routine teaching and interactions with students [6]. Typically, such behaviors stem from teachers’ preconceived notions or attitudes shaped by their understanding of students’ diverse family backgrounds, academic performances, and personality traits, leading to differentiated treatment among students [7]. This differentiation manifests as favoring certain students while discriminating against or suppressing others [8]. Research indicates that teachers’ differential behavior significantly impacts students’ academic achievements, psychological traits, and social development. Teacher favoritism or neglect can influence students’ academic performance and motivation positively or negatively, respectively. Positive emotional support can enhance students’ self-esteem and confidence, while negative feedback may exacerbate anxiety and stress levels. Moreover, equitable treatment and constructive interactions from teachers foster students’ sense of belonging, enhance social skills, improve peer relationships, and promote the development of positive behavioral norms [9, 10].

Regarding the factors influencing teachers’ differential behavior, some scholars posit that students’ varied performances, encompassing their personalities, abilities, and family backgrounds, constitute the primary drivers [11]. However, a prevailing view among scholars suggests that teachers’ differential behaviors predominantly stem from the teachers themselves. Factors such as their personality traits, goal expectations, and behavioral orientations influence their overall assessment of students, leading to differential treatment [12, 13]. Research indicates that teachers with high self-efficacy are more inclined towards treating students as independent learners, aiming to influence their learning styles and address disruptive behaviors [14,15,16]. Conversely, studies have also found that teachers experiencing higher levels of job burnout and stress exhibit reduced tolerance towards students’ individualized performances, often resulting in differential behaviors [17]. In terms of perceiving teachers’ differential behaviors, studies highlight that students’ responses and experiences primarily reflect teachers’ instructional practices and daily educational interactions. Students’ emotional and behavioral changes, along with their perceptions, serve as indicators of teachers’ conscious or unconscious differential behaviors [18]. Therefore, students’ perceptions offer an objective means to quantify teachers’ differential behavior. As direct recipients of teacher actions, students observe firsthand the variations in classroom interactions, emotional responses, and resource allocations. Their experiential insights provide authentic and varied feedback, crucial for revealing the nuances of teachers’ differential treatment, whether intentional or unintentional. Consequently, such feedback forms a robust and genuine empirical basis for further research endeavors.

There are limited measurement tools available for assessing teachers’ differential behavior both domestically and internationally. Brophy et al. pioneered the use of a dual-observation system to quantify teachers’ differential behaviors based on interaction frequency and emotional communication in the classroom [19]. Subsequently, Weinstein developed the Teacher Treatment Inventory (TTI) [20] from students’ perspectives, widely utilized abroad primarily for evaluating teachers’ instructional practices, albeit lacking emotional factors. Chinese university education emphasizes quality and advocates for integrating cultural education with ideological and emotional development. Babad et al. expanded the dimension of emotional support in teachers’ differential behavior, developing the Perceived Teacher Behavior Questionnaire (PTBQ) [21]. However, this questionnaire is critiqued for its limited item count, uneven distribution, and incomplete coverage, resulting in limited adoption. Chinese scholar Fan Liheng addressed these gaps by formulating the Teacher Differential Behavior Scale, tailored to the realities of primary and secondary school students in China. This scale, evaluated from the student’s viewpoint, encompasses guidance, feedback, opportunity, and emotional aspects [22]. Nonetheless, the characteristic evaluation of teachers’ differential treatment lacks an objective measurement of students’ perceptions of such behaviors. Moreover, significant distinctions exist between college students and those in primary and secondary education. College students navigate a critical developmental phase from late adolescence to early adulthood, pivotal for shaping individual personalities, values, and lifestyles. This phase underscores academic advancement and independence in living, fostering heightened psychological and social maturation that enhances their ability to discern and respond to environmental stimuli. College students’ increased autonomy necessitates managing academic pursuits, personal responsibilities, social interactions, and crucial life decisions. Their autonomy also enables them to appraise the impact of differential teacher behavior more accurately, offering a nuanced perspective for research. Consequently, developing a measurement tool, such as the “perception of teacher differential behavior among young college students,” tailored to Chinese socio-cultural contexts, becomes imperative. Such a tool not only aids in understanding and improving college students’ psychological well-being but also guides enhancements in teaching effectiveness and educational quality. Understanding students’ evaluations of teacher behavior assists educators in refining instructional methods and attitudes, fostering a more equitable educational environment, and mitigating unintended disparities. Furthermore, research into college students’ perceptions of differential teacher behavior informs evidence-based educational policy formulation and implementation. By comprehensively exploring students’ experiences and needs, policymakers can devise rational strategies to promote educational equity, quality, and holistic student development effectively.

From the perspective of college students’ self-perception, this study selected 7,053 college students from 18 universities across 6 provinces and municipalities as participants. Employing both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA), the study aimed to delineate the structure of the Teacher Differential Behavior Scale. The findings are anticipated to serve as a foundational tool in mitigating teachers’ differential treatment, fostering the cultivation of appropriate life values among college students, and promoting their overall physical and mental well-being. Acknowledging the prolonged duration of university education and the substantial variability in individual developmental trajectories, the study encompassed both undergraduate (bachelor’s) and postgraduate students.

Materials and methods

Participants

From September to October 2022, using convenience sampling, six provinces (Shanghai, Jiangsu, Shandong, Anhui, Yunnan, and Gansu) were selected based on their GDP. Within these provinces, 18 colleges and universities were chosen, representing different levels of higher education (key undergraduate, general undergraduate, and junior college). Each institution distributed 400 questionnaires, totaling 7,200, with 7,053 valid responses recovered, resulting in a recovery rate of 97.96%. The sample comprised 3,032 males and 4,021 females. Participants were randomly divided into two groups, one of 3,526 and the other of 3,527, based on the initial serial number of their questionnaire (odd or even) to meet the research requirements. These groups were subjected to exploratory factor analysis (EFA) and confirmatory factor analysis (CFA), respectively, to determine the structure of the Perceived Teacher Differential Behavior Scale. The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Kangda University of Nanjing Medical University (NO.2022-1) and the Medical Ethics Committee of Binzhou Medical University (NO.2022 − 280), and all participants signed the informed consent.

Questionnaire preparation process

Open-ended questionnaire and expert interviews

The initial perception teacher differential behavior Scale was developed using literature review and open-ended questionnaires. The open-ended questionnaire covered four main aspects: (1) What do you think teacher differential treatment refers to, based on your own experience? (2) Have you ever experienced differential treatment from a teacher, and if so, in what ways? (3) How do you think teachers’ differential behaviors manifest? (4) What qualities do you believe define a good teacher? Convenience sampling selected 58 college students from five universities in China, including 26 students from double first class universities, 18 from regular undergraduate institutions, and 14 from vocational colleges. Semi-structured online interviews were conducted with these participants, with their consent for audio recording obtained. The responses from these open-ended interviews were used to identify dimensions of perceived differential teacher behaviors and to formulate initial survey questionnaires. Additionally, a convenience sample of 10 psychology experts and 6 clinical psychologists participated in a questionnaire survey to define the construct, delineate its content, and establish measurement indicators for perceived teacher differential behavior. After two rounds of expert consultations, it was agreed that teachers’ differential behavior is primarily reflected in the fairness and impartiality of their treatment of students, encompassing two dimensions: prejudice and preference. Students perceive teachers’ differential behaviors mainly through emotional feedback, opportunity privilege, and behavior orientation.

Prepare the initial questionnaire

Based on the results of open-ended interviews and expert consultations, and drawing on Brophy and Good’s teacher expectancy cycle model, Weinstein’s teacher treatment inventory, and Babad’s teacher behavior questionnaire, we plan to develop a college students’ perception teacher differential behavior scale. This scale will encompass two dimensions: teacher bias and teacher favoritism. Each dimension will initially include items addressing teachers’ emotional feedback, opportunities granted, and behavior orientation. After compiling the initial questionnaire, a pilot survey will be conducted among sophomore college students to ensure that the content of each item is appropriate and comprehensible. Following revisions to the wording of the questionnaire, the initial test questionnaire will consist of 25 items, using a 7-point Likert scale (1 = strongly disagree, 2 = disagree, 3 = somewhat disagree, 4 = neutral, 5 = somewhat agree, 6 = agree, 7 = strongly agree).

Efficiency standard measuring tool

Research suggests that differential treatment by teachers can lead to psychological stress and anxiety among students. Those who perceive themselves as being treated negatively or unfairly may experience heightened concerns about their academic performance and social interactions, potentially triggering or exacerbating symptoms of anxiety. Additionally, studies have shown that students who perceive differential teacher behavior may also experience reduced self-esteem, feelings of helplessness, and a pessimistic outlook. Students who are exposed to such negative environments over the long term may develop symptoms of depression as a result. To establish the construct validity of the perception of teacher differential behavior scale, anxiety and depression scales are used as external criteria for validation. A significant positive correlation between scores on the perception of teacher differential behavior scale and those on anxiety and depression scales would indicate a substantial relationship between the perception of teacher differential behavior and students’ psychological well-being. Utilizing anxiety and depression scales as external criterion measures helps verify whether the perception of teacher differential behavior scale effectively distinguishes students who perceive differential treatment and facilitates a deeper understanding of the potential negative psychological impacts of such treatment [23,24,25].

Generalized anxiety disorder (GAD-7)

The scale, developed by Spitzer et al. [26], contains seven items designed to evaluate various aspects of anxiety: tension and anxiety, uncontrollable worry, excessive worry, inability to relax, inability to sit still, irritability, anger, and fear. The scale employs a scoring method ranging from 0 to 3, with responses indicating frequency from “not at all” to “almost every day.” Higher scores reflect greater levels of anxiety, with scores of 10–14 indicating moderate anxiety and 15–21 indicating severe anxiety. In this study, the scale demonstrated a Cronbach’s alpha coefficient of 0.965, indicating high internal consistency.

Patient health questionaire-9items (PHQ-9)

Developed by Kroenke et al. [27], this scale measures respondents’ feelings of depression over the past two weeks. The scale comprises nine items, assessing aspects such as loss of interest, low mood, sleep disturbances, fatigue, eating disorders, feelings of inferiority, difficulty concentrating, psychomotor retardation, and suicidal ideation. Each item is rated on a four-point Likert scale, from “never” (0 points) to " almost every day " (3 points). The total score, derived from summing the individual item scores, indicates the severity of depression, with higher scores reflecting greater levels of depression. In this study, the PHQ-9 scale demonstrated a Cronbach’s alpha coefficient of 0.950, indicating high internal consistency.

Data analysis

SPSS 22.0 was used for item analysis, exploratory factor analysis, and reliability and validity test, and Mplus 8.3 was used for confirmatory factor analysis of the scale.

Results

Project analysis

With a sample of 3526 college students as participants, correlations between scores of each dimension and the total score were examined in the initial questionnaire. The Pearson correlation coefficients between each item and the total score exceeded 0.3, indicating a moderate level of correlation and suggesting good discriminant validity. Additionally, all items demonstrated critical ratio (CR) values in the discriminant validity test that reached significance at the 0.001 level, affirming their robustness in distinguishing between constructs.

Factor analysis

Exploratory factor analysis

Based on a sample of 3526 participants, Bartlett’s test of sphericity was conducted to assess the suitability of factor analysis, yielding a Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy at 0.962, indicating appropriateness for factor analysis. Subsequently, principal component analysis was performed using the maximum variance method, and a rotated factor loading matrix was obtained. Each factor was required to meet criteria including communalities greater than 0.3, eigenvalues exceeding 1, factor loadings above 0.40, and loading predominantly on a single factor. Following these criteria, 24 items were retained, forming two dimensions, with factor loadings and communalities detailed in Table 1. Exploratory factor analysis extracted two common factors, contributing to a cumulative variance of 76.576%. Factor one was named “teacher preference” and included 12 items, and factor two was named “teacher prejudice” and included 12 items.

Table 1 Exploratory factor analysis results of understanding teachers’ differential behavior

In addition, the Bartlett test was conducted on the two dimensions of “teacher preference” and “teacher prejudice,” yielding KMO values of 0.954 and 0.966, respectively. Exploratory factor analysis was then performed, extracting three dimensions for both “teacher preference” and “teacher prejudice.” The contribution rates for these dimensions were 82.143% and 91.785%, respectively. Combining the two dimensions resulted in the formation of six common factors. The first factor was named “preference-emotional feedback” (F1), the second “preference-behavior orientation” (F2), the third “preference-opportunity privilege” (F3), the fourth “prejudice-emotional feedback” (F4), the fifth “prejudice-behavior orientation” (F5), and the sixth “prejudice-opportunity privilege” (F6). Each factor included four items, as detailed in Tables 2 and 3.

Table 2 Exploratory factor analysis results of teachers’ preference behavior

Table 3 Exploratory factor analysis results of teachers’ prejudice behavior

Teacher preference refers to the evident favoritism or more positive attitudes and behaviors displayed by teachers towards certain students during interactions. This preference may be based on students’ personal characteristics, academic performance, personality traits, or other factors. Teacher prejudice typically involves unfair or inconsistent behaviors and attitudes exhibited by teachers towards different students. Specifically, teacher prejudice can be defined as differential treatment based on students’ characteristics such as academic achievements, gender, race, socioeconomic background, observed during classroom interactions or other student engagements. This differential treatment may be intentional or unintentional. Emotional feedback primarily encompasses the emotional responses and attitudes displayed by teachers during interactions with students. Such emotional feedback includes expressions, behaviors, and verbal responses exhibited by teachers towards different students. Positive emotional feedback involves praise, encouragement, support, care, enthusiastic interactions, while negative emotional feedback may include criticism, rebuke, indifference, neglect, negative body language, among others. Behavior orientation pertains to specific actions and strategies employed by teachers towards different students in teaching or other contexts. These actions and strategies reflect teachers’ attitudes, expectations, and levels of attention towards students, manifested through adjustments in teaching methods, variations in interaction styles, and differential allocation of resources. Opportunity privilege primarily involves teachers providing certain students with greater opportunities or special treatment compared to others during the educational process. These opportunities and privileges encompass academic opportunities, extracurricular activities, leadership roles, and preferential use of resources.

Confirmatory factor analysis

Using another sample of 3,527 college students, confirmatory factor analysis (CFA) was performed on the six-factor structural model of perceived teachers’ differential behaviors to test its stability. Figure 1 illustrates the standardized path coefficients. The factors “behavior orientation” and “opportunity privilege” under prejudice were combined into a single factor, resulting in “preference-behavior orientation,” “prejudice-behavior orientation,” and two other factors, “preference-emotional feedback,” and “prejudice-emotional feedback.” The CFA compared the fit indices of both the four-factor and six-factor models. The results indicated that the fit indices for the six-factor structural model were excellent, and the fit indices for the four-factor structural model were also satisfactory. Among them, the fit indices for teacher preference are as follows: χ²/df = 4.549, CFI = 0.990, TLI = 0.958, RMSEA = 0.070, SRMR = 0.062; for teacher prejudice, the fit indices are χ²/df = 8.352, CFI = 0.928, TLI = 0.907, RMSEA = 0.071, SRMR = 0.038, indicating good structural validity of both primary indicators of perception teacher differential behavior. In the second-order model, the fit indices for preference-emotional feedback are χ²/df = 5.577, CFI = 0.994, TLI = 0.983, RMSEA = 0.065, SRMR = 0.010; for preference-behavior orientation, they are χ²/df = 4.807, CFI = 0.998, TLI = 0.993, RMSEA = 0.061, SRMR = 0.006; for preference-opportunity privilege, they are χ²/df = 12.567, CFI = 0.942, TLI = 0.926, RMSEA = 0.073, SRMR = 0.069; for prejudice-emotional feedback, they are χ²/df = 6.919, CFI = 0.997, TLI = 0.990, RMSEA = 0.068, SRMR = 0.007; for prejudice-behavior orientation, they are χ²/df = 8.880, CFI = 0.982, TLI = 0.947, RMSEA = 0.071, SRMR = 0.018; for prejudice-opportunity privilege, they are χ²/df = 6.228, CFI = 0.997, TLI = 0.990, RMSEA = 0.067, SRMR = 0.005; indicating that the structural validity of the second-order model also meets appropriate levels (Table 4).

Fig. 1

Standardized roadmap of the six-factor structural model for understanding teachers’ differential behavior

Table 4 The fitting index and reliability of each dimension, six-factor and four-factor confirmatory factor analysis of the comprehension teacher Differential Behavior Scale

Reliability analysis

Using a sample of 3,527 college students, both internal consistency reliability and split-half reliability tests were conducted. The results demonstrated that the internal consistency reliability of the Comprehension Teacher Differential Behavior Scale was 0.917. Specifically, the internal consistency reliability for the teacher preference dimension was 0.956, with a split-half reliability of 0.933. For the teacher prejudice dimension, the internal consistency reliability was 0.980, with a split-half reliability of 0.957. The internal consistency reliability and split-half reliability for each factor ranged from 0.790 to 0.963 and from 0.624 to 0.963, respectively (Table 4).

Validity analysis

Construct validity

Confirmatory factor analysis demonstrated that the six-factor structure of the Perceived Teacher Differential Behavior Scale was valid. Using data from a sample of 3,527 college students, item scores within each factor were summed. Pearson correlation analysis was then employed for pairwise comparisons within the factors. The analysis revealed that the correlations between the factors were moderate to high (correlation coefficients ranging from 0.307 to 0.767), which were lower than the correlations between each factor and its respective dimension (correlation coefficients ranging from 0.323 to 0.938). These results indicate that the six factors are independent and exhibit strong internal consistency. Furthermore, by employing the Average Variance Extracted (AVE) method to measure Discriminant Validity, it was determined that the square root of the AVE values exceeded the Pearson correlations with other related constructs, thereby confirming the presence of discriminant validity (Table 5).

Table 5 Presents the correlation matrix between each factor of the Perceived Teacher Differential Behavior Scale and the total score of the respective dimensions

Empirical validity

The empirical validity analysis utilized the Generalized Anxiety Scale and the Depression Scale as criteria. The findings indicate a negative correlation between teacher preference and anxiety (-0.268) and depression (-0.287), whereas teacher prejudice shows a positive correlation with anxiety (0.342) and depression (-0.352). Specifically, correlations between each factor within teacher preference and anxiety/depression range from − 0.312 to -0.216, while those within teacher prejudice range from 0.317 to 0.361, all with a significance level of P < 0.001 (Table 6).

Table 6 Validation of the correlation between each factor and dimension of the comprehension

Invariance test

In this study, we conducted a thorough examination of measurement invariance to verify the consistency of the scales across various demographic variables (gender, age, grade, and type of school). Through the implementation of Multi-Group Confirmatory Factor Analysis (MGCFA), this research investigated whether the factor structures of the preference and bias scales related to teacher differential treatment remained invariant across different sub-samples. Specifically, the invariance testing was divided into three levels. Configural Invariance: Initially, it was confirmed that the basic structure of the model remained unchanged across all groups. This stage involved no constraints on factor loadings or intercepts, primarily verifying whether different groups have a common understanding of the structure of the scales. Metric Invariance: Building on the established configural invariance, constraints were applied to the factor loadings in the model to ensure that the understanding and sensitivity of responses to the items of the scale were consistent across sub-samples. The testing of metric invariance involved comparing the fit indices between the constrained and unconstrained models, commonly using indicators such as the Chi-square Difference Test, Comparative Fit Index (CFI), and Root Mean Square Error of Approximation (RMSEA). Scalar Invariance: Further constraints were applied to the intercepts to test whether different groups had the same baseline scores on the scales. Scalar invariance is crucial for interpreting the scoring of the scales, ensuring that the scores can be compared fairly across different groups.

The results of the study indicate that across groups defined by gender, age, grade, and school type, both the teacher prejudice and preference scales, as well as their dimensional indices, demonstrated robust configural and metric invariance, with Comparative Fit Indices (CFI) exceeding 0.90 and Root Mean Square Error of Approximation (RMSEA) values below 0.08. Scalar invariance was also confirmed in most groups, evidenced by the stability of the Incremental Fit Indices, which showed minimal changes in the CFI (not exceeding 0.01) before and after the imposition of intercept constraints. The RMSEA and Standardized Root Mean Square Residual (SRMR) values exhibited consistent stability, indicating that the scales provide consistent and comparable scoring standards across different demographic groups. Through these systematic invariance tests, this research ensured that the scales possess equivalent measurement properties across diverse groups, thereby providing a solid foundation for further analysis and ensuring the broad applicability and accuracy of the research findings.

Discussion

To enhance the validity and reliability of the Comprehension Teacher Differential Behavior Scale, this study adheres to established procedures and standards in psychological measurement. Initially, the study draws upon dimensions and items from analogous scales, adapting them to reflect the realities of Chinese college students. Utilizing open-ended questionnaires and expert interviews, the study constructs initial questionnaires and specific items based on observed instances of comprehending teachers’ differential behaviors. Subsequent stages involve item analysis, exploratory factor analysis, confirmatory factor analysis, and empirical testing to assess the scale’s reliability and validity. These processes culminate in the formalization of a scale designed to measure Chinese college students’ perceptions of teacher differential behavior.

To ensure the scientific rigor of the scale, this study rigorously examined the Chinese college students’ perception of teacher differential behavior through a comprehensive, multi-perspective, multi-dimensional, and multi-channel approach. Exploratory factor analysis revealed that students’ perceptions of teachers’ behavior can be categorized into two distinct dimensions: teacher preference and teacher prejudice. Subsequent exploratory factor analyses conducted on these dimensions identified three factors each: emotional feedback, behavior orientation, and opportunity privilege. This structural framework aligns closely with the theoretical underpinnings of the study, resulting in the construction of a two-dimensional, six-factor scale comprising twenty-four items. In contrast to Weinstein’s teacher treatment scale abroad, which features dimensions such as “work and prescribed orientation,” “negative feedback and teacher guidance,” and “high expectations and opportunity choice,” this study introduces differentiated dimensions based on positive and negative impacts of teachers’ behavior. Specifically, emotional feedback corresponds to aspects of “negative feedback” and “high expectations,” while behavior orientation aligns with “work and prescribed orientation” and “teacher guidance.” Opportunity privilege encompasses facets of “opportunity choice,” expanding upon the original scale by addressing both positive and negative feedback in emotional interactions, diverse aspects of teachers’ behavior orientation, and multiple dimensions of opportunity privilege, including tolerance, understanding, discrimination, and isolation. These enhancements contribute to a more comprehensive and scientifically grounded measurement of teacher differential behavior perceptions among Chinese college students.

In this study, we rigorously designed and tested a psychometric scale to explore the impacts of teacher prejudice and preference on the mental health of university students, analyzing the responses from a diverse sample of 7053 participants. The student sample was stratified across various demographic variables, including age, gender, academic discipline, and institution type, ensuring that the findings could be generalized across the broader student population. This diversity is crucial in understanding the varied perceptions of teacher behavior and its psychological implications. The initial item analysis indicated all items had correlation coefficients with the total scale score greater than 0.3, confirming good discriminant validity, which suggests that the items were well-suited to differentiate between perceptions of teacher prejudice and preference. In the exploratory factor analysis, item loadings on their respective factors—teacher prejudice and teacher preference—ranged from 0.6 to 0.9. This high level of loading demonstrates that the items are highly representative of the underlying constructs they are intended to measure. Following EFA, confirmatory factor analysis was conducted to validate the factor structure proposed by the EFA. The results were highly encouraging, with all principal components showing significant pairwise correlations, suggesting that the dimensions of teacher prejudice and preference are interrelated yet distinct constructs. The Composite Reliability (CR) for each factor exceeded 0.7, and split-half reliability scores were above 0.6, indicating that the scale is reliable for repeated use in diverse settings. These metrics underscore the internal consistency and stability of the scale. Structural validity was assessed using fit indices such as the Comparative Fit Index (CFI) and Tucker-Lewis Index (TLI), both of which exceeded 0.9, while the Root Mean Square Error of Approximation (RMSEA) and Standardized Root Mean Square Residual (SRMR) were maintained below 0.8. These results affirm the scale’s structural integrity and its ability to accurately measure the constructs. Discriminant validity was further supported by moderate to high correlations among factors, which were distinct enough to validate each factor’s uniqueness. Empirical validity was robust, as demonstrated by significant correlations of the teacher prejudice and preference dimensions with students’ reported levels of anxiety and depression, highlighting the real-world applicability and relevance of the scale. The scale’s invariance across different demographic groups was tested to ensure that it measures the same constructs in the same way across different groups of students. This is crucial for applying the findings universally across the student population. The results confirmed that the scale is invariant, suggesting that the impacts of teacher prejudice and preference on student mental health are consistent across genders, ages, and other demographic variables. This study’s findings highlight the significant impact of teacher differential treatment on student mental health and underscore the necessity for educational institutions to address these behaviors proactively. The validated scale provides a valuable tool for identifying and quantifying the perceptions of teacher behavior, thereby facilitating targeted interventions aimed at improving the educational environment and student well-being.

In higher education teaching, teacher preference and teacher prejudice are critical dimensions that directly influence students’ learning experiences and educational equity. Addressing these issues requires systematic improvements in emotional feedback, behavior orientation, and opportunity privilege by university educators. Firstly, teachers’ emotional feedback significantly impacts students’ confidence and motivation. It is crucial to avoid teacher preference and prejudice in emotional feedback. Teachers should strive to provide consistent emotional feedback to all students, avoiding overt favoritism or neglect towards any individual. For instance, in praising, encouraging, or criticizing students, feedback should be based on specific behaviors and efforts rather than personal preferences or preconceived biases. When providing constructive criticism, teachers should maintain a supportive attitude and avoid undue harshness or leniency based on biases. Specific guidance should be offered to help students improve rather than merely dismissing their efforts. Secondly, teachers’ behavior orientation in teaching methods, classroom interactions, and student life directly affects students’ learning opportunities and engagement. Teachers should ensure balanced interaction with all students in the classroom, avoiding excessive attention or neglect towards certain individuals. Diverse questioning and interaction methods should encourage active participation from every student. In personalized teaching, teachers should assess each student’s needs and potential fairly, providing differentiated guidance and support based on academic requirements rather than personal preferences or biases. Ensure all students have equal access to personalized learning resources and assistance. Furthermore, opportunity privilege involves teachers’ decisions in resource and opportunity allocation, which profoundly impact students’ academic and career development. Teachers should adhere to transparent and fair standards when distributing academic resources, extracurricular opportunities, and recommendation letters. Open opportunities and selection criteria to all students ensure equal competitive opportunities. Active encouragement and support for students from diverse backgrounds and abilities should be promoted in various academic and extracurricular activities, avoiding limitations on certain students’ development opportunities due to biases. For example, teachers can implement open application and selection procedures to provide more students with opportunities to showcase their talents and potentials. Overall, university educators should remain vigilant against the unfair impacts of teacher preference and prejudice on students through fair and consistent emotional feedback, balanced classroom interactions, and transparent opportunity allocation. By creating a more just and inclusive learning environment, educators not only enhance educational equity and quality but also stimulate students’ motivation and potential for growth in a fair environment. Continuous reflection and improvement in teaching practices are essential for achieving educational equity and excellence in higher education.

Limitations

While the Comprehension Teacher Differential Behavior Scale demonstrates robust reliability and validity, certain areas for enhancement should be acknowledged. Firstly, the study’s sample selection was limited to 18 colleges and universities without considering students’ majors, potentially introducing sample bias. Secondly, the inclusion of key universities, public universities, and junior colleges excluded institutions without graduate programs, leading to a disproportionately small sample size of graduate students. Exploring the differential behavior between postgraduates and their supervisors is crucial due to the unique dynamics in this relationship. Future research could develop a separate scale focusing on differential behavior in postgraduate supervision contexts. Furthermore, teachers’ differential behavior reflects contemporary characteristics, necessitating ongoing refinement of its definition in alignment with current educational contexts.

In this study, exploratory factor analysis was employed primarily to investigate the dimensions and interrelationships among questionnaire items based on data, while also examining and improving the quality of the items. The study divided the same sample of respondents into two groups using odd and even numbers for exploratory and confirmatory factor analyses, respectively. This approach demonstrated feasibility by enabling internal cross-validation of data within a relatively small sample, thereby conserving resources and time in data collection and partially controlling for internal consistency within the sample. However, this design also carries significant limitations. Because modifications to items and dimensions cannot be made after the exploratory factor analysis stage, suboptimal factor structures may enter the confirmatory factor analysis stage, potentially affecting model fit and the accuracy of research conclusions. Furthermore, this grouping method may compromise sample representativeness, particularly in smaller sample sizes, making it challenging to fully capture the characteristics of the entire population and impacting the robustness and external validity of the results. Therefore, despite its practicality under resource constraints, future research should consider conducting exploratory and confirmatory factor analyses on independent samples and allowing for item adjustments as necessary to enhance the validity and reliability of research findings.

Conclusions

The perception teacher differential behavior scale for Chinese university students is a crucial assessment tool designed to reveal and quantify teachers’ differential treatment of students during the educational process. Through systematic measurement, this scale aids educational researchers and administrators in gaining deeper insights into how teacher preferences and prejudices manifest in educational practices and their impacts on students. Comprising dimensions such as emotional feedback, behavior orientation, and opportunity privilege, each dimension consists of four items.

The scale covers two main dimensions: teacher prejudice and teacher preference, each encompassing variations in teachers’ behaviors across emotional feedback, behavior orientation, and opportunity privilege. These measurements not only illuminate teachers’ differential attitudes and behaviors towards students but also provide essential data for identifying unfair practices in education. Teachers’ emotional feedback directly influences students’ psychological well-being and academic motivation, with fair and consistent emotional feedback enhancing students’ self-esteem and learning enthusiasm. Behavior orientation determines students’ participation opportunities and learning experiences in the classroom, where balanced interaction and personalized teaching support facilitate optimal student development. Opportunity privilege concerns whether students can access academic resources and development opportunities equitably, and transparent and equitable opportunity allocation helps eliminate hidden inequalities in education. By employing the perception teacher differential behavior scale, educational institutions can identify instances of teacher preferences and prejudices during instruction, enabling targeted interventions to enhance educational equity and quality. This approach not only improves teaching practices, fostering fairness and equity in educational environments, but also enhances students’ motivation and academic performance, thereby contributing to overall educational advancement.