Play is undoubtedly an essential daily activity of early childhood and theoretically impacts children’s social, emotional, and cognitive development (e.g., Sylva et al., 2004; Weisberg et al., 2013a; Zosh et al., 2018). Yet what teachers believe about play are thought to impact if and how they implement it within their pedagogy (Pyle and Danniels, 2017; Avornyo and Baker, 2018), and therefore whether it will assist in the quality of ECEC provision (Wen et al., 2011). Play beliefs are defined broadly as tacit, often unconsciously held assumptions about play and its relation to children’s development (Haight et al., 1997; Fogle and Mendez, 2006). In extant studies, play beliefs are synonymous with values, conceptions, perceptions, perspectives, images, views, thoughts, judgments toward children’s play (Pajares, 1992; Fives and Buehl, 2012; Baustad et al., 2018; Nilsen, 2021). Adults’ beliefs contribute to positive or negative judgments about the significance of play (Fogle and Mendez, 2006). In particular, ECEC teachers’ perspectives on play are important for the enactment of the most dominant type of ECEC approaches: play-based curricula (Avornyo and Baker, 2018). Acting as a contextual “filter” (Clark and Peterson, 1986), teachers’ beliefs to some extent affect when teachers interpret, screen their classroom practices, and adapt the subsequent ones (Wen et al., 2011). Teachers’ beliefs toward play and learning influence the curricula and their pedagogical practices, as well as children’s early learning and development (Kansanen, 1991). Their beliefs are particularly important if ECEC is to have the life-long impact indicated by longitudinal research (e.g., McCoy et al., 2017), as teachers can only implement a balance of guided and free play if their values support this (Parker and Neuharth-Pritchett, 2006; Wen et al., 2011; Yin et al., 2021).

In a study by Hyvonen (2011), teachers’ play beliefs were comprised of teachers’ views on the effects of play on early childhood education, preference for the specific kinds of play, both of which determine how teachers arrange play activities. In most previous studies, the roles of teachers are varied by how teachers interact with children for better development in play activities. Some theorists argued that teacher participation may interfere with child play (e.g., Isaacs, 1929), and decades of teacher education and training have reinforced the idea that teacher involvement will only impair the maximal value of play for learning (Enz and Christie, 1993; Pyle and Danniels, 2017). Asian ECEC teachers have traditionally valued play less than their European-heritage counterparts (Bubikova-Moan et al., 2019), with beliefs in some countries (e.g., China) seeing play and learning as completely incompatible, meaning many activities—including play—are adult-guided (Wu et al., 2018). Chinese ECEC curriculum has succumbed to the influence of Western education since it began “opening up” to globalization, both in policy and practice, since the 1980s (Zhu and Zhang, 2008, p. 174), most palpably since the 1989 “Kindergarten Work Regulations and Procedures” (p. 174). This policy emphasized the value of child-initiated activity and its integration with group and other activities (Qi and Melhuish, 2017). By the 1990s, the Reggio Emilia Approach (Edwards et al., 2011), The Zone of Proximal Development (Vygotsky, 1978) and the Project Approach (Helm and Katz, 2011)—each of which acknowledge the importance of child-initiated play—were all recognized in Chinese ECEC policy (Qi and Melhuish, 2017). Such national changes sit within an international context of empirical (Siraj-Blatchford, 2009; Weisberg et al., 2013b) and policy (Bennett, 2005; Rogers, 2013) trends toward a more active teacher role in children’s play, which often calls for greater incorporation of academic content within children’s learning activities. However, these rapid reforms were challenging for Chinese teachers (Zhu and Zhang, 2008, 2018). As noted in the famous “Preschool in Three Cultures” research, the reforms’ success “depends on teachers’ understanding of how and why to teach children in the ways the Guidelines suggest” (Tobin et al., 2009, p. 85). The views of parents, who may not agree with such child-centered approaches to early learning (Zhu and Zhang, 2008; Fung and Cheng, 2012), are likely to also contribute to Chinese ECEC teachers’ current beliefs.

Other than differing beliefs about how much they value academic skills (e.g., literacy, numeracy) and child-initiated play, it is likely there is variation in beliefs how much teachers should involve themselves in children’s play, as many scholars suggest this would abilities and promote their cognitive, social, and language development (Smilansky and Shefatya, 1990; Enz and Christie, 1993; Weisberg et al., 2013b; Pyle et al., 2021). Various authors identify specific roles ECEC teachers should move between when involved in play such as stage manager, play leader, director, co-player, and uninvolved co-player (Enz and Christie, 1993; van Hoorn et al., 2014), model, mediator, allower, and afforder (Hyvonen, 2011). Indeed, the roles can be considered within the discourse of play regulation and support (Rengel, 2013), but do imply greater value on adult-determined content (e.g., academics) than a purely child-initiated approach. This might suggest that Chinese ECEC teachers have shifted from a teacher-directed curriculum to a child-centered one, only to be shifted more toward a view which values play and academics equally. Research has yet to confirm if this is so.

Previous studies have elaborated powerful arguments for the importance of play in children’s education and identified variability in roles that ECEC teachers take in child play (Fisher et al., 2013; Weisberg et al., 2013b; Zosh et al., 2018). A vital first step in ensuring teachers take the most educationally valuable role in play is to determine the beliefs underpinning their practices. Given the culturally situated nature of these beliefs (Bronfenbrenner and Morris, 2006), a key piece in this first step is understanding the cultural beliefs of the most populous country in the world. China has received profound influence from Confucian heritage culture (CHC), which values academic training in early childhood and highlights traditional teaching approach on children’s development (Zhu and Zhang, 2008). In this cultural context, Chinese teachers may be regarded as an educational authority whose direction has long been widely encouraged, meaning the value of child-initiated, play-based activity has been overlooked (Li, 2012). Coupled with the lack of “faith” that many parents have in play, it is likely that there is some diversity in teachers’ beliefs, as has been found in qualitative work in Chinese heritage (e.g., Fung and Cheng, 2012, p. 24) and European heritage contexts (e.g., Pyle and Danniels, 2017; Colliver, 2019). However, limited studies have examined the diversity in Chinese ECEC teachers’ beliefs toward children’s play. This study, therefore, aims to explore the variation in Chinese ECEC teachers’ play beliefs.

To examine the intra- cultural variation of Chinese ECE teachers’ beliefs, we situated the study within a cultural-historical theory methodological framework, which aims to provide a “holistic” view of the phenomenon studied (Chaiklin, 2012, p. 209). For Rogoff (2003) “the interpersonal, personal, and cultural-institutional aspects of the event constitute the activity… [n]o aspect exists or can be studied in isolation from the others” (p. 58). For the purposes of the current research questions, this approach is about seeing teachers’ beliefs not as isolated from their context, but rather informing and informed by the larger societal, cultural, historical spheres, as well as the sub-cultures of each ECE setting (or preschool), the children attending, and their families (Bronfenbrenner and Morris, 2006). To do so, cultural-historical research examines activity and motive in one sphere as its units of analysis, maintaining the others “in the background” (Matusov, 2007, p. 324). Vygotsky, the grandfather of cultural-historical theory, saw this process as akin to studying the molecules within a drop of water to make inferences about the ocean; a “form of analysis (fundamental to) development of theories of thinking …relies on the partitioning of the complex whole into units… those units in which the characteristics of the whole are present” (Vygotsky, 2004, p. 37). To Rogoff’s (1995) three main planes of analysis—the individual, interpersonal, and cultural-institutional—Hedegaard (2008, p. 46) added the societal as a broader plane again. For the purposes of investigating intra-cultural variation in teachers’ beliefs, it is appropriate to examine teachers’ beliefs at the “institutional” level, as the beliefs are tied to the ECEC institution as their workplace (Colliver, 2019). In keeping with the focus on beliefs, it is possible to focus on participants’ values within a cultural-historical approach, as they represent motives at the institutional level (Hedegaard, 2009), and motives are the key unit of analysis in cultural-historical research (Fleer, 2008). Values are often considered to be part of teachers’ play beliefs (Nilsen, 2021).

Akin to these four levels of analysis, psychometric research can be conducted using variable-centered, person-centered, and person-specific approaches (Howard and Hoffman, 2018). Just as an institutional-level analysis will examine variation within a subcultural group of society (e.g., Chinese ECE teachers), despite its misleading name, a person-centered approach will determine if subgroups exist within a given population (Howard and Hoffman, 2018). This approach contrasts societal-level psychometric research, which most often uses variable-centered approaches to examine homogeneous groups of observations based on the combination of some independent variables (Bubikova-Moan et al., 2019). Variable-centered approaches may also examine the relationship of variables with play-learning beliefs (Mooi and Sarstedt, 2011; Avornyo and Baker, 2018). Notwithstanding, variable-centered analyses tend to ignore the intra-population variability, for they assume that associations are similar across all the participants (Howard and Hoffman, 2018). For example, Avornyo and Baker (2018) investigated Ghanaian early years stakeholders’ play beliefs along the single dimension (i.e., the role of play in children’s learning), roughly categorizing their play beliefs according to levels of education, and examined whether their education status could predict play-learning beliefs individually. However, it was noted that components of teachers’ beliefs systems are often not isomorphic, and thus there exists the potential for belief variables to combine in novel ways that shape outcomes. Finally, person-specific approaches are used to examine relationships between variables at the individual level, and are inappropriately fine-grained for research on Chinese ECEC teachers’ beliefs.

In contrast with studying ECEC teachers’ play beliefs in isolation, person-centered approaches allow unique arrangements of play beliefs to be examined in association with their predictors or outcomes. Unlike variable-orientated procedures, the person-centered methodology can look at multiple ideas of addressing a specific issue, and better analyze multidimensional and correlated data (Bámaca-Colbert and Gayles, 2010). ECEC teachers with different cultural backgrounds emphasize different characteristics and significance of children’s play (Sandberg and Samuelsson, 2003; Wu and Rao, 2011; van der Aalsvoort et al., 2015). In addition to culture, childhood experience is also considered key to beliefs and perceptions that are associated with adult beliefs and behavior (Thompson et al., 2008). Prior studies have reported variability in roles that teachers enact in children’s play, as well as fluctuations in the role types (Johnson et al., 1999, p. 201; Pyle and Danniels, 2017). A more nuanced understanding could be garnered about the types of play beliefs among sub-populations of individuals identified based on within-group similarities and between-group differences in associations among variables. Person-centered approaches to analysis are well positioned to provide such an understanding (Howard and Hoffman, 2018).

To overcome apparent disadvantages of variable-centered analysis (Li and Chen, 2017; Yang and Li, 2018; Lin et al., 2020), this study applied a person-centered approach via latent profile analysis (LPA) to explore the variation in ECEC teachers’ play beliefs.

Acknowledging the diversity and complexity of prior investigations of ECEC teachers’ play beliefs, the current study uses a person-centered approach via LPA to identify various profiles of individuals based on their play beliefs. Theoretically, we intend to contribute to the field through empirical evidence and substantiating the person-centered approach in play belief research. Researchers can acquire a more parsimonious but nuanced understanding of the play beliefs of different backgrounds. Practically, it informs early education authorities and practitioners of a convenient way to identify patterns of teachers’ play beliefs and provide them with differentiated support.

Overall, the present study aims to: (a) identify patterns of ECEC teachers’ play beliefs; (b) examine how these patterns are related to teachers’ backgrounds; The following research questions guided this investigation:

This study was conducted in Fuzhou, the capital city of Fujian province, located in southeast China. Participants were informed that they could withdraw participation anytime and informed consent was given at the start of the survey. A total of 674 ECEC teachers were recruited from a variety of kindergartens, in both urban and rural areas to ensure sample diversity. As shown in Table 1, 172 teachers (25.5%) had more than 10 years of teaching experience. The majority (65.4%) of the teachers had an associated or lower degree and 34.6% of the teachers had a bachelor or above degree. Almost all (90.1%) teachers specialized in preschool education. Around half (49.7%) had no teaching qualification.

The cultural influences nature on teachers’ beliefs makes cultural-historical theoretical framework most appropriate for the current investigative focus (Haight et al., 1997; Hedegaard, 2008). Within the tiered structure of influences of activity (societal, institutional, interpersonal and individual levels; Rogoff, 1995; Hedegaard, 2008), the institutionally nested nature of Chinese ECEC teachers’ beliefs determined our institutional-level focus on values and beliefs, as discussed earlier. This level of analysis fits with the person-centered approach described within Howard and Hoffman’s (2018) three levels of approach to psychological research, which, it is argued, should be chosen for its fit with the research question(s).

In the present study, LPA was used to discover diverse patterns of teachers’ play beliefs. As a person-centered statistical approach, LPA allows researchers to find previously unnoticed subgroups among varied populations (Vermunt and Magidson, 2002; Vermunt, 2010). In comparison to traditional cluster analysis, LPA uses rigorous model-fit indices to determine the number and character of the profiles, resulting in higher sensitivity to latent taxonomy detection (Cleland et al., 2000). To get the proper latent profile solution, we employed the following commonly used model-fit indices (Morin et al., 2011; Foti et al., 2012): The Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and sample-size adjusted BIC (SSA-BIC), which should be smaller than those of other profiles; the Entropy value should be higher; the Vuong-Lo-Mendell-Rubin likelihood ratio statistics (VLMR) value and Bootstrap Likelihood Ratio Test (BLRT) value should be significant (p < 0.05). In addition to the above fitting indices, the solution of the model should also consider the theoretical limitations.

First, the mean scores of the two CTPBS subscales “Academics over Guided Play” (AGP) and “Free Play and Socio-Emotional Skills Support” (FPSSS) were employed as indicators to identify the latent profiles of teachers’ play beliefs. Next, a series of LPA models were conducted in Mplus Version 8.0 using the full information maximum likelihood (FIML) as the estimation method (Muthén and Muthén, 1998–2010). Then, teachers’ background information variables were used as predictors of the latent profile variable. Logistic regression was conducted in Mplus Version 8.0 with the indicators of years teaching experience, education major, education level, and teaching qualification to evaluate how these factors were associated with the Chinese ECEC teachers’ play beliefs patterns.

The first research question in this study examined Chinese ECEC teachers’ play beliefs profiles. Latent profile models ranging from 1 to 5 classes were estimated. As shown in Table 4, a two-profile model was adopted as it had the highest entropy (0.90), and a significant VLMR and BLRT value (p < 0.001). Despite having the lowest AIC, BIC, and SSA-BIC value, the four-profile structure’s VLMR value was insignificant. In consideration of the model indices and the interpretability of the profiles, a two-profile solution was adopted as the final model in the present study.

As shown in Figure 2, Profile 1 was distinguished by a high level of AGP (M = 4.51, SD = 0.04) and a low level of FPSSS (M = 2.43, SD = 0.02). Profile 2 was distinguished by high scores on both AGP (M = 4.42, SD = 0.03) and FPSSS (M = 4.48, SD = 0.02). The LPA results identified two profiles: (1) High AGP and Low FPSSS (91.1%), (2) High AGP and FPSSS (9.0%).

The second research question of this study was to identify whether teachers’ years teaching experience, education major, education level, and teaching qualification predicated Chinese ECEC teachers’ play beliefs profile. Binary logistic regression was conducted in Mplus using the above indicators to evaluate how these factors related to the likelihood of membership in a Chinese ECEC teachers’ play beliefs profile.

As shown in Table 5, comparing the High AGP and Low FPSSS profile (Profile 1) to the High AGP and High FPSSS profile (Profile 2), there was a significant difference in years teaching experience. Teachers with 10 years and above of teaching experience were more likely to belong to latent profile 1 (Logits = 1.624, p < 0.01, OR = 5.257) than those with 0–3years of teaching experience. There were no significant differences (p > 0.05) in teachers’ education major, education level, and teaching qualification comparing Profile 1 to Profile 2.

In collaboration with ECEC teachers and experts in China, this study developed, revised, and validated the Chinese Teachers’ Play Beliefs Scale (CTPBS). First, a 27-item pilot version of the measure was culturally adapted by the final author based on the current parental play belief scale (Fogle and Mendez, 2006), then the first author translated it from English into Chinese. A group of Chinese ECEC teachers were asked to comment on the scale’s wording and suitability and ECEC experts were consulted with to confirm the scale’s content validity. In the process of EFA, 14 items were removed because of unacceptably low communities, their lack of significant loadings on the two factors, or cross-loading among factors. The final version of CTPBS was a 13-item scale, which included two constructs: “Academics over Guided Play” (AGP) and “Free Play and Socio-Emotional Skills Support” (FPSSS), which had similar factors of parental play belief scale to Fogle and Mendez (2006) study. However, in line with recent international trends toward intentional teaching in play (Bennett, 2005; Siraj-Blatchford, 2009), the CTPBS further explored not only attitudes on the benefits of child-centered free play, but also adult’s participation in child play, and the corresponding focus on skills learned from free and guided play (see Zosh et al., 2018). The items of the CTPBS also seemed to be more generalized than those in the Chinese Parent Play Beliefs Scale (CPPBS; Lin and Li, 2018), which was derived from a detailed list of young children’s play and preacademic activities. While CTPBS may share some concepts similar to CPPBS—play for fun and play for learning, their differences probably exist because concrete items were appropriate for teachers (as professionals intimately familiar with this nomenclature) but not for parents.

Using a person-centered approach via LPA, the present study revealed a two-profile model: Profile 1, who placed the highest value academic learning (e.g., counting, recognizing letters, reading), a slightly lower value on guided play, and an even lower value on free play and socio-emotional skills (e.g., language, communication, feeling expression, social skills); and Profile 2, who placed a high value both on academic and socio-emotional learning, as well as on guided and free play, albeit to a lesser extent than on the learning. Chinese parents may value adult-guided activities over free play because they believe adult-guided activities are more effective for children’s academic learning, which they value more (Rao and Li, 2009; Fung and Cheng, 2012; Sun and Rao, 2017). Interestingly, the item –“It would be more beneficial to read to children than to join in on their play”—which loaded well (0.713) onto the first subscale, AGP, suggested that adult-led instruction such as reading was valued more than even guided play, similar to this value hierarchy shown by many Chinese parents (Fung and Cheng, 2012; Sun and Rao, 2017). Similarly, Q11, “It is more important for children to have good academic skills than to play well with others” suggests a valuing of academics over play that provides a nuance to the valuing of guided play that may be quite culturally specific to Asia, as suggested in the systematic review of Bubikova-Moan et al. (2019). Like profiles of Chinese parental play beliefs where parental involvement moderated children’s academic-related play only if those parents valued play for learning, and not if they valued play just for fun (Lin and Li, 2018; Lin et al., 2019), profiles in the current study showed distinct intracultural groups of ECEC teachers who value free play differently. In contrast to the findings of prior studies, the majority of ECEC teachers valued their role in play for children’s academic learning above the value of free play for child-initiated learning of socio-emotional skills (Chakravarthi, 2009; Rengel, 2013; Clevenger, 2016; Avornyo and Baker, 2018). Chinese ECEC teachers’ valuing of their active role in play contrasts patterns found in western-heritage countries, where teachers are reluctant to “hijack” play’s inherent value (e.g., Pyle and Danniels, 2017, p. 274; Colliver, 2019). Similarly, Turkish teachers with the most experience (> 15 years) were most likely to be uninvolved in play (Ivrendi, 2017). The majority of Chinese ECEC teachers still appear to value academic learning—and to a lesser extent, guided play—over free play and socio-emotional skills. However, those with less experience, most likely newer teachers, valued guided and free play equally, suggesting they have “caught up” with contemporary trends toward valuing intentional teaching and guided play (Bennett, 2005; Siraj-Blatchford, 2009; Weisberg et al., 2013b), and possibly the contemporary emphasis on valuing and enacting both guided and free play equally (Sylva et al., 2004; Zosh et al., 2018). However, the challenge for Chinese policy is to ensure this valuing of free play and child autonomy is enacted well, as research suggests that it takes years of experience to know how to convert these beliefs to practice (Parker and Neuharth-Pritchett, 2006; Wen et al., 2011; Yin et al., 2021). While there is much rhetoric regarding the moral imperative to uphold children’s human right to play (Colliver and Doel-Mackaway, 2021), it requires skill and expertise to follow children’s interests and initiatives in a thoughtful and educational manner (Edwards et al., 2011; Hedges and Cooper, 2016).

In China, the central status of play as the principal activity in kindergarten was enshrined in the 2001 Guidelines for Kindergarten Education Practice (The Ministry of Education of the People’s Republic of China, 2001). With educators responding positively to the policies, great changes have taken place in kindergartens across China. Anji play, a successful experiment in child-initiated outdoor play at preschool, is being expanded across the country and learned by ECEC educators throughout the world (Wang, 2018). The current results are consistent with the idea that the benefits of a child-centered, play-based curriculum have more recently become widely accepted by Chinese ECEC teachers. Thus, Profile 2 in this study unsurprisingly valued guided and free play equally. Many preschool educators, however, may be influenced to the expectations of the families that they serve to nurture early learners and prepare young children for later academic learning (Fung and Cheng, 2012; Sun and Rao, 2017), and this tendency appears to be reflected particularly in Profile 1, high in AGP, academic learning over guided play support, and low in FPSSS. Moreover, since the importation of notions and practices imported from European-heritage countries, there has exited a conflict and tension between “host” traditional values and “guest” ones, where adults adopt different patterns of acculturation and adjustment of their culturally based beliefs about play (Berry, 1997; Rao et al., 2010). Confucian heritage culture appears to deeply rooted in China, with some teachers still tending to view recreation as akin to laziness, and playing as not being useful, with high expectations of children’s academic achievement (Lee, 1996; Li, 2003; Rao and Chan, 2009). If the 1:1 ratio of adult- and child-initiated activities, and 1:1 ratio of adult- and child-guided play (Sylva et al., 2004), is also optimal in the Chinese context, these background influences remain a significant challenge in the enterprise of improving Chinese ECEC provision: for more experienced teachers because they hold views less consistent with contemporary research evidence, and for less experienced teachers because they are more likely to hold less managerial and therefore influential roles.

Studies have suggested that teachers’ educational background and teaching experience are related to teaching beliefs (Spodek, 1988; Charlesworth et al., 1993; Calderhead, 1996). The current study found that years of teaching experience predicted the profiles of ECEC teachers’ play beliefs. However, in contrast with those with less teaching experience, teachers with a decade or more teaching experience were less likely to have a free play and socio-emotional skill focus. There were no significant differences in teachers’ education major, education level, and teaching qualification comparing Profile 1 to Profile 2.

These findings contrast those of previous studies, which have suggested that more experienced Chinese teachers have more positive views on the protection of children’s right to play and an endorsement of child-centered free play (Liu, 2014). Since child play is regarded as the principal activity in ECEC, how to better support and guide child play has become the fundamental component and focus of teacher training and professional learning. On the other hand, previous studies (e.g., Wu et al., 2018) have generally shown Chinese ECEC teachers have valued play less than those from Western-heritage countries (as highlighted by Q6 and Q12 of the AGP), with the growing influence of western ideas, it seems likely that more recently trained teachers would espouse values more supportive of free play and holistic learning. It is possible that the current study is showing newer teachers have had more contemporary training reflecting an equal emphasis on guided and free play, and holistic learning as per the balancing of academic and socio-emotional learning. It is notable that previous research has shown that more teaching experience can be a predictor of higher teacher self-efficacy, which is related to the autonomy of teachers’ professional development (Cheung, 2008; von Suchodoletz et al., 2018) as well as their likelihood of enacting a play-based curriculum, which may involve minimal adult intervention in play (Yin et al., 2021). On the other hand, Chinese teachers with less teaching experience may be more likely to enact the recent ECEC education reforms emphasizing play-based, individualized and active learning (Liu and Feng, 2005). However, implementing these values in practice may require experience these teachers do not yet possess (Parker and Neuharth-Pritchett, 2006; Yin et al., 2021), or specialist in-service training (Wen et al., 2011; Hedges and Cooper, 2016; Colliver, 2019).

The changing values of Chinese ECEC teachers suggested by the current data are likely reflective of the rising influences of the Zone of Proximal Development (Vygotsky, 1978) and Reggio Emilia (Edwards et al., 2011) and Project (Helm and Katz, 2011) approaches in China in the late twentieth and early twenty-first centuries (Qi and Melhuish, 2017), the relative value of free play was significantly lower for the group who received teacher training more than 10 years ago (although such approaches can be effectively used to balance child- and teacher-oriented activities; Veraksa et al., 2021). A review of the prominent ECEC literature published in Mandarin over the past two decades reflects a redirection of value orientation in ECEC teachers’ beliefs about free play and learning socio-emotional skills. Affected by Western ECEC concepts and the Developmentally Appropriate Practice (DAP) approach (McMullen et al., 2005), Chinese literature in the early twenty-first century paid more attention to the value of child-centered play, children’s autonomy and holistic learning, including an emphasis on social and emotional skills (e.g., Qiu, 2001; Jing and Chen, 2002). However, partly due to parental expectations (e.g., Sun and Rao, 2017) and some empirical evidence (e.g., Hu et al., 2017), there has been a recent redirection in the play literature toward striking a balance between child-initiated and teacher-led play, with more emphasis on teachers’ active role in supporting and guiding play for learning (e.g., Lin et al., 2019; Deng, 2021). It is possible that this change of values, and potentially from their influence on teacher education, is reflected in the significant belief differences found here between teachers with more or less experience.

Several limitations need to be noted when interpreting the current findings and addressed in future studies. First, the sample was limited to teachers in Fuzhou, a highly developed city in China. There is a need for future studies to involve ECEC teachers from less-developed areas of China, to examine play beliefs more comprehensively in contemporary China. Second, some researchers have argued that the improvement of the statistical ability of LPA would be brought from more class indicators and a larger sample size (Nylund et al., 2007). Therefore, it is desirable that further studies would be conducted with larger samples using more class indicators. Third, the data about play beliefs were reported by ECEC teachers, which may lead to self-report bias or socially desirable response sets. In further research, apart from teachers’ self-reports, researchers might consider using other-informant ratings or observation. Fourth, previous studies have investigated parental and teachers’ play beliefs, but few have examined whether ECEC administrators and teachers differ in how they view child play, which is needed to be explored in future studies. Fourth, although the Parent Play Belief Scale was here adapted to reflect contemporary international trends toward balancing guided and free play (Bennett, 2005; Siraj-Blatchford, 2009), many of the items did not meet communality criteria outlined in the section “Materials and Methods”. This resulted in a smaller scale than would be anticipated if using factors derived from the original scale (Fogle and Mendez, 2006). Finally, our study did not explore and verify the cause-effect relationship between teachers’ play beliefs and child outcomes, which calls for further studies considering these variables (e.g., Lin and Li, 2018).