Teacher Credibility Emerges as Key to Chinese Thesis-Writing Empowerment

A cross-sectional survey of 446 third-year English majors at two Chinese universities found that students’ perceptions of teacher credibility had the strongest and most consistent statistical association with empowerment in high-stakes thesis-writing courses. Credibility was most strongly associated with perceived course impact, while teacher care was most strongly associated with perceived course meaningfulness and teaching clarity with perceived course competence. The study used a Chinese adaptation of the 5Cs Scale, a framework measuring five positive teacher interpersonal behaviors: care, clarity, closeness, confirmation and credibility. The results suggest that students did not respond to these behaviors uniformly: different forms of teacher support were associated with distinct psychological resources as students negotiated academic writing, evaluation and the consequences attached to graduation.

Participants were enrolled in compulsory English thesis-writing courses that met once a week for two class hours. The courses addressed topic selection, proposal defenses, literature reviews, thesis structure, theoretical application, textual analysis and academic language expression. Students were 20 to 22 years old; 215 participants were male and 231 were female, and most had passed either the Test for English Majors Band 4 or the College English Test Band 4. The researchers distributed 500 questionnaires online and obtained 446 complete, valid responses, an 89.2% response rate. Students rated the questionnaire on a seven-point scale ranging from strongly disagree to strongly agree. Because all measures were collected in a single survey, the results describe statistical associations in this sample rather than showing that particular teacher behaviors caused changes in empowerment.

Adapting the measurement instrument required more than straightforward translation. The original set contained 133 items covering teacher behaviors, learner empowerment and learning effects. The researchers excluded the learning-effects scale because it was not central to the study and reduced the burden of a long online questionnaire. Five experts, including specialists in English education and educational psychology, participated in translation, back-translation, cultural adaptation and content validation. Two bilingual experts independently translated the English items into Chinese, while two bilingual experts unfamiliar with the original questionnaire back-translated the Chinese version. Ambiguous or inconsistent items were discussed and revised. Cultural changes included removing an item that used the contrast “competent” and “stupid,” deleting an overlapping care item and rephrasing “my teacher uses my knowledge in the class” as “my teacher adopts my view on the thesis in class.” The final Chinese instrument contained 49 items: 31 addressing teacher behaviors and 18 addressing learner empowerment.

Expert ratings supported the content of the adapted measure. Item-level content validity indices ranged from 0.80 to 1.00, and the average scale-level content validity index was 0.94. A pilot study involving 50 students examined item discrimination and factor loadings. Three items were removed after exploratory factor analysis, including one reverse-coded item. In the formal sample, internal consistency was high across the measured dimensions, with Cronbach’s alpha values ranging from 0.865 to 0.926. Corrected item-total correlations ranged from 0.682 to 0.797. A Harman single-factor test indicated that the first unrotated factor accounted for 35.09% of the variance, below the study’s stated 40% reference threshold. That screen suggested common method bias was not a serious concern in these data, although it cannot exclude all forms of bias associated with self-report questionnaires.

Exploratory and confirmatory factor analyses supported the adapted scale’s structure in the surveyed population. Exploratory analysis of half of the formal sample produced 10 factors with eigenvalues above 1, explaining 71.30% of the total variance. Factor loadings ranged from 0.585 to 0.836, with no reported cross-loadings above 0.40. Confirmatory analysis of the other half fit the proposed model well: the chi-square-to-degrees-of-freedom ratio was 1.381, the root mean square error of approximation was 0.029, the comparative fit index was 0.970, the incremental fit index was 0.970, the Tucker-Lewis index was 0.968 and the Parsimony Normed Fit Index was 0.839. Composite reliability values were all above 0.86, average variance extracted values exceeded 0.58, and heterotrait-monotrait ratios ranged from 0.377 to 0.659. The researchers also modeled credibility as a second-order factor with competence, goodwill and trustworthiness as its three subdimensions; correlations among those subdimensions ranged from 0.49 to 0.66.

Structural equation modeling then estimated 15 proposed paths from the five teacher behaviors to the three dimensions of learner empowerment. Impact represented students’ belief that they could influence the course and have a voice in its outcomes. Meaningfulness concerned the perceived value, interest and personal relevance of course content and tasks. Competence reflected confidence in performing required tasks and achieving course goals. The structural model showed good fit, with a chi-square-to-degrees-of-freedom ratio of 1.401, a root mean square error of approximation of 0.030 and a comparative fit index of 0.971. All 15 paths were positive and statistically significant at p < 0.05, but their standardized coefficients differed. The strongest association for each outcome was teacher credibility to impact, with β = 0.296; teacher care to meaningfulness, with β = 0.320; and teaching clarity to competence, with β = 0.242.

The result that most sharply diverged from previous work concerned the relative standing of confirmation and credibility. In an earlier Iranian English-as-a-foreign-language university context, the study cited by the researchers reported confirmation as the strongest predictor of empowerment, with β = 0.29, and credibility as the weakest, with β = 0.14. In the Chinese thesis-writing sample, credibility had the largest standardized coefficient for all three empowerment outcomes, ranging from 0.196 for competence to 0.296 for impact. Closeness had the smallest association with impact, at β = 0.121. The contrast is consistent with the possibility that the relative importance of teacher behaviors may vary with educational culture, task demands and students’ immediate psychological concerns, but the Iranian and Chinese studies were not direct experimental comparisons. The present data cannot isolate cultural, institutional or task-related causes, and the coefficients describe associations rather than demonstrated causal effects.

One tentative interpretation is that thesis writing intensifies students’ need for assurance that their instructors possess relevant expertise and will provide dependable guidance. The credibility factor combined perceived competence, goodwill and trustworthiness. A student who views a teacher as knowledgeable, responsible and genuinely concerned may be more willing to propose an argument, negotiate a writing direction or trust feedback on a demanding project. By contrast, care had the clearest link to meaningfulness, suggesting that encouragement and attention to individual difficulties may help students experience a required assignment as relevant to their own development. Teaching clarity had the strongest association with competence, consistent with the possibility that explicit criteria, task decomposition, examples and actionable feedback make a complex writing process easier to navigate. These are interpretive mechanisms proposed by the authors; the survey itself measured relationships among students’ perceptions rather than the instructional acts directly or the resulting psychological changes over time.

Several limitations constrain what can be concluded. The cross-sectional, self-report design precludes causal inference and cannot show whether teacher behavior changes empowerment or whether students’ perceptions of teachers and empowerment influence their survey responses. The sample came from two comprehensive universities, so results may not generalize to specialist language institutions, vocational colleges, other countries or different educational systems. The study did not measure actual writing performance, revision quality, teacher-reported data, students’ use of artificial intelligence tools or qualitative responses, and it included no interviews or open-ended questionnaires. Longitudinal, experimental and cross-cultural studies would be needed to test whether the observed pathways are stable over time, causal or generalizable. Even with those caveats, the research provides a technically grounded account of how students’ perceptions of three forms of teacher support map onto different dimensions of empowerment in an academically demanding Chinese English-major setting.

Subject of Research: Psychometric validation of the Chinese 5Cs Scale and its associations with learner empowerment in English-major thesis writing.

Article Title: Localizing the 5Cs Scale in China: when teacher credibility outperforms confirmation in high-stakes thesis writing

Article References: Sun, Y., He, G., & Niu, Z. (2026). Localizing the 5Cs Scale in China: when teacher credibility outperforms confirmation in high-stakes thesis writing. Frontiers in Psychology, 17, Article 1955402. https://doi.org/10.3389/fpsyg.2026.1955402

Image Credits: AI Generated

DOI: 10.3389/fpsyg.2026.1955402

Keywords: 5Cs Scale, teacher credibility, learner empowerment, thesis writing, English majors, China, positive psychology, structural equation modeling, teacher-student relationships, Localizing, Scale, teacher