In recent decades, the quality of education has become a key topic of concern among parents, educators, and policymakers. The ongoing economic, social, and information integration holds strong implications for how the prospect of the future is changing, in that globalization, automation, and knowledge economy are driving a broad emphasis on investment in people. Conceptually, a wide range of inputs fed into the process of learning, such as pupil effort, teacher quality, and school facilities, while instructional and organizational practices are key determinants affecting how efficiently these inputs combine and generate meaningful learning engagements. To that end, evidence across international contexts underscores the critical importance of effective teaching for bolstering educational success, encouraging meaningful interaction, and sustaining lifelong learning (Altinok and Kingdon, 2012; Hanushek and Rivkin, 2012; Chetty et al., 2014). More critically, the weight of the influence of teachers on learning outcomes has been shown to be more deterministic in developing countries, where availability of home educational resources and learning opportunities are scarce (Bau and Das, 2017; Liu and Steiner-Khamsi, 2020). Consequently, policymakers have become interested in evaluating how the educational credentials of teachers can serve as promising predictors of effective teaching and in understanding how such observable indicators of teacher quality affect the learning of children (Liu, 2021).

Nonetheless, the empirical literature on the value of the educational credentials of teachers has long been divided. In one camp, scholars argue that educational credential reflects positively on the level of the academic proficiency of a teacher, in addition to the reflections of sufficient cognitive and non-cognitive skills required to design, execute, and engage in teaching activities (Harris and Sass, 2011). The theoretical basis for this view rests on that the human capital development of teachers precedes in importance than that of their pupils, and that their educational mindset, instructional effectiveness, and professional learning all depend on a foundational level of accumulated skills throughout their own teacher education preparation. As a case in point, Jackson and Bruegmann (2009) summarized the attained level of observable human capital development of teachers, such as degree level, licensure and certification status, and score on licensure examination, into a single index and found positive associations with math and reading achievement of pupils. In broad strokes, teaching is a complex task that requires a well-rounded set of cognitive and non-cognitive abilities to excel and not to mention intensive demands for pedagogical skills and subject knowledge for classroom success. Studies supporting this view have persistently shown that the human capital attainment of teachers strongly predicts on-the-job performance and subsequent pupil learning gains in the classroom (Bastian, 2019; Noell et al., 2019).

Conversely, there is an opposing research cluster that casts doubt on the positive link between the educational credentials of teachers and the academic achievement of pupils. In particular, scholars argue that observable teacher traits are inconsistent at best when used as the predictors of effective teaching (Hanushek and Rivkin, 2006; Chingos and Peterson, 2011). These studies directly put in question the usefulness of existing teacher-education arrangements, as well as making uncertain the effectiveness of educational policy initiatives aimed at improving teacher educational attainment and teacher preparation. Accordingly, such controversial division in this literature strongly motivates rigorous empirical evaluations to assess the value-added of the educational credentials of teachers in learning terms.

To address these aforementioned research contentions, this study elects to focus on the case of China, where swift teacher education reforms and reorganization policies have been taking place in light of public demands for a better quality of education (Zhou and Reed, 2005). A key factor linking the Chinese context to the broader international debate on the usefulness of the educational credentials of teachers as a predictor of effective teaching has been the keen policy interest and emphasis of China in improving teacher educational attainment and the quality of teacher education (Liu and Xie, 2021). Historically, the educational credential requirements of teachers in China varied considerably by instructional level. For instance, primary schools have traditionally set high school diploma, or shizhuan, as a basic requirement for the entry of a new teacher, whereas secondary schools often require at least a 2-year associate's degree, or zhuanke, to be eligible for employment (Ingersoll, 2007).

In recent years, national motivation to increase the supply of teachers who hold higher levels of educational credentials has been strong. In 1999, the State Council initiated Action Plan to Revitalize Education in the Twenty-first Century and widely publicized its Decisions to Deepen Educational Reform and Improve Quality-Oriented Education in a Holistic Way, which served as policy guidelines in generating national consensus to improve teacher education (Zhou, 2014). As mandated in the Ten-Five Teacher Education Reform and Development Outline, the Ministry of Education (2002) posits that all new teachers in compulsory education should hold at least a bachelor's degree, which led to a sharp increase in the percentage of new teachers who hold bachelor-level educational credentials. In addition, the Ministry of Education initiated Free Teacher Education Program (FTEP) in 2007, recruiting more than 46,000 bachelor-degree teacher candidates in its first cycle (Qian et al., 2020). As a result, the proportion of teachers who hold bachelor's degrees drastically increased within a relatively short time frame. According to the Ministry of Education (2016), the percentage of lower-secondary teachers with at least a Bachelor of Education (BEd) degree expanded quintuple times from 12.4% of the total teaching force in 1999 to 59.4% in 2009, and by 2016, approximately 4 out of 5 lower-secondary teachers have at least a BEd degree.

Despite the rapid movement to bachelorize teachers in China, its benefits realized in pupil learning terms have been unclear. For instance, a cluster of scholars increasingly scrutinize this policy movement to improve the quality of a teacher by “bachelorizing” teachers and question whether attaining a higher level of educational credential contributes to effective teaching (Jin, 2007; Zhou et al., 2011; Hu, 2015). In general, there still exists an open debate in academic circles regarding the effects on learning, resulting from the rapid movement to bachelorize teachers, which echoes the broader international discussion regarding the usefulness of the educational credentials of teachers as a predictor of effective teaching. In particular, little is known empirically about the actual benefits of promoting mass teacher education at the bachelor level (Zhou, 2014). This study, through employing an empirical analysis leveraging pupil fixed-effect modeling, evaluate the extent to which the educational credentials of teachers may serve as a predictor of effective teaching and quantify their contributions to learning.

Prior studies have indicated that there exists an open debate as to whether observable teacher background traits, such as the educational credentials of teachers, are predictive of teaching effectiveness (Chingos and Peterson, 2011). On the one hand, studies have argued that the human capital development of teachers is vital in determining how well they translate what they know into what pupils learn, particularly in making instructional experiences meaningful to pupils who may possess wide-ranging learning needs (Darling-Hammond et al., 2005). On the other hand, critics of teacher education have asserted that pre-service teacher education is only useful to the extent that teachers are minimally qualified to be a teacher but may not necessarily translate into effective teaching (Ballou and Podgursky, 2005); therefore, many ponder the policy implications since there is some indication that the higher levels of the educational credentials of teachers do not predict elevated learning gains (Hanushek and Rivkin, 2006). Not to mention that academic skeptics have scrutinized recent policy engagements in China, which promote mass teacher education at the bachelor level (Zhou, 2014; Hu, 2015).

The research objective in this study was to evaluate the extent to which the educational credentials of teachers may serve as a predictor of effective teaching and quantify their contributions to pupil's educational success. In broad strokes, this study is centered on the Chinese educational context and employs a pupil fixed-effect modeling approach to relate differences in the educational credentials of teachers to variations in pupil learning gains all within the same pupil, while eliminating problematic confounders at the pupil-, teacher-, and school levels. In broad terms, the Chinese context provides a useful analytic case where policy engagement with improving teacher educational attainment has been swift and strong. More specifically, analytic findings that are generated using baseline and follow-up CEPS datasets indicate that returns to studying under teachers holding higher levels of educational credential are substantial, as calculated in terms of pupil learning gain. Particularly, pupils who study with teachers who hold at least a BEd degree are observed to gain 0.042 higher SDs of measured learning between baseline and follow-up, compared with when the pupil studies with teachers who do not hold such credentials. This effect translates to as much as 1–2 months of additional learning gain per academic year (Organisation for Economic Co-operation Development., 2016). Considering the potentially compounding effect over time, the number of additional learning gains that better-educated teachers can deliver could be quite significant, particularly if pupils are systematically exposed to better-educated teachers in the course of their learning careers.

Findings in this study add rigorous evidence to a growing body of literature highlighting the criticality of the human capital attainment of teachers through programs of tertiary teacher education preparation for boosting instructional effectiveness and bolstering pupil learning and development (Clotfelter et al., 2007; Winters et al., 2012). Conceptually, focusing on improving teacher education and bettering teacher preparation at the tertiary level, well before teachers arrive in the classroom, is an attractive policy option because existing evidence supports that instructional effectiveness matters for learning outcomes (Allen et al., 2013). At the theoretical level, it is crucial to recognize that the human capital development of teachers is a deterministic factor affecting that of their pupils and that how well teachers are educated reflects how well pupils can be reasonably expected to learn. The importance of the level of human capital development of teachers in meeting the increasingly complex demands of instructional success in the classroom should not be underemphasized. As such, a joint policy emphasis should be placed on improving the quality of teacher education and incentivizing teacher candidates in pursuing higher levels of educational credentials. In practice, however, many countries find it a fiscal and logistical challenge to procure large populations of highly qualified teachers, given that the personnel-related costs are already taking up significant portions of educational budgets (Baumol, 1993; Levin, 2011). To that end, while this study does not directly offer fiscal solutions, its findings critically counter against opponents of teacher education and show that better-educated teachers indeed lead to improved learning gains, which can translate into large growth prospects and societal gains in the long run (Hanushek, 2011).

Additionally, closer attention must be paid to understand the inner workings of the Chinese model of teacher education, when contextualizing and translating findings in this study to a culturally and educationally different circumstance. Most importantly, neither all teacher education programs are of the same quality nor do they offer teacher candidates the same professional preparation to make a meaningful difference in the classroom. For instance, Li (2012) observed that the Chinese model of teacher education has been historically a hybrid system that involves a combination of specialized teacher training colleges and comprehensive research universities, and the model emphasizes core values such as independence, adaptability, and diversity stemming from Confucian epistemology and pragmatism. In addition, studies elsewhere have highlighted the importance of cultural relevance in supporting teacher candidates to self-reflect, critique, and link their unique teacher education experience to effective instructional practices in the classroom (Acquah et al., 2020). Finally, in order for teacher candidates to truly benefit from high-quality teacher education programs, studies have shown that they must invest in themselves through hard work, as indicated through undergraduate course grades and course hours (Kukla-Acevedo, 2009).

Nonetheless, two important limitations in this study are also worth mentioning and can perhaps motivate future research in addressing them. First, it is critical to acknowledge that this study only evaluates the extent to which the educational credentials of teachers may serve as an indication for effective teaching, in that its contribution to pupil learning gains is empirically calculated. There are many additional dimensions of human capital development traits of teachers, which are worth considering, such as teacher certification (Harris and Sass, 2009) and teaching experience (Kane et al., 2008), as well as teacher rank in the Chinese context, where the findings in this study, as well as that in the studies by Hannum and Park (2007) and Chu et al. (2015), have shown to positively influence pupil learning. These positive signals of teacher quality are not only indicative of the human capital development of teachers but also more importantly predict their instructional effectiveness when they enter the classroom. Second, the CEPS study does not include in its research design to randomly assign teachers to pupils, and therefore, there are reasonable concerns that sample selection bias at both the school- and classroom level may be present and confound findings. For instance, if more qualified teachers are systematically assigned to teach classes enrolled with high performers, or if more qualified teachers are compensatorily assigned to teach in low-performing classes, the relationship between the educational credentials of teachers and the pupil learning gains becomes ambiguous. Notwithstanding, the pupil fixed-effect modeling approach attempted to address some of these issues to the possible extent, by leveraging within-pupil, between-subject variation, in addition to including a rich set of pupil- and teacher-level control variables. In this regard, however, the estimated effect of the educational credentials of teachers could reflect a combined effect, to some extent, comprised of unobserved peer characteristics at both pupil- and teacher levels. Consequently, future research could readily take advantage of the ongoing longitudinal CEPS and build on findings in this study to explore how sample selection bias may be at play.

China Education Panel Survey dataset is publicly-available at https://doi.org/10.18170/DVN/KURJUU.

Written informed consent to participate in the study was obtained by National Survey Research Center of Renmin University, PR China. All research ethics approvals were granted by the Institutional Review Board of Renmin University, PR China.

The author confirms being the sole contributor of this work and has approved it for publication.

This study was financially supported by the National Social Science Foundation of China (Grant No. CJA200256).

The author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.