Sustaining Content and Language Integrated Learning in China: A Systematic Review

Abstract

The past decades have witnessed the rise and boom of Content and Language Integrated Learning (CLIL). As this dual-focused pedagogical approach continues to grow with remarkable vitality and sustainability, however, there is a paucity of knowledge regarding its development and application in China, necessitating a systematic review of previous research to apprise the academia of what has been done in the education and research agenda of CLIL. Grounded in four databases (i.e., Web of Science, Scopus, Education Resources Information Center, and Google Scholar), 28 research articles were extracted from 2013 to 2022 with certain exclusion and inclusion criteria considered. Thematic analyses primarily demonstrated that: (1) most studies were conducted in higher education providers, with English being the predominant instructional language; (2) diverse subject matters were embraced, denoting the coordination of “self” and “other” intercultural identities; (3) affective evidence concerning perceptions of CLIL and performance evidence regarding language learning outcomes tended to be the research nuclei, de-emphasizing the other research topics (e.g., assessment, translanguaging, heterogeneity in learning, learning materials, learning process); (4) the research findings were various, with most of them corroborating the benefits of CLIL to different aspects of learning; (5) CLIL learners were the principal source of data, and diverse research paradigms were displayed with appropriate designs as per the research objectives. These findings, by comparison with the knowledge and experience obtained from the Western world, imply that the scope of the CLIL agenda should be extended and deepened in China, encouraging educators and researchers to consider and probe a wider range of issues of interest.

1. Introduction

The last decades have witnessed the birth and rise of Content and Language Integrated Learning (CLIL). As an umbrella term that refers to an “educational approach in which an additional language is used for the learning and teaching of both content and language” [1], CLIL distinguishes itself from other foreign language teaching (FLT) approaches, with undeniable potential to benefit learners in knowledge construction, language learning, affective learning, cognitive development, and intercultural understanding. Furthermore, it lends itself to sustainable education and can be integrated at various levels because its dual-focused nature reflects a holistic approach to education, embraces cultural differences, and makes connections to other disciplines and students’ daily lives [2]. CLIL was once considered an increasingly acknowledged approach to bilingual education in Europe [3], while its extensive spread has created an irresistible educational tide for other sociocultural contexts to swim with, one of which is China.

It is no exaggeration to maintain that CLIL is not new in China, and some clues about it have been found in the English immersion programs provided at the beginning of this century in some developed Chinese areas [4]. However, it is the last decade that has seen an upsurge in CLIL research and implementation [5]. Based on the experience and knowledge gained from the Western world, scholars summarize that CLIL in China should be an approach “in which a target language is expected to be used as much as possible and as appropriate as possible for the teaching and learning of both content and language in integrated ways for multiple educational purposes”, with an outlook on yuren (students’ comprehensive development of knowledge, capabilities, and qualities), kecheng (designing and developing curriculums based on the yuren objectives), jiaocai (designing authentic learning materials), jiaoxue (emphasizing both content and language goals in teaching), and jiaoshi (addressing teachers’ needs of professional development) [6]. This interpretation, along with many of the others [7,8,9], has established a solid theoretical foundation for the local application of CLIL. However, CLIL is still an obscure topic of interest in China, with the academia unaware of what has been done in the boom of content-based FLT programs and the growing body of research [10]. This necessitates an updated and systematic review of solid empirical evidence to shed light on an enlightening CLIL agenda explicit on its development patterns, the understanding of which can help to retain the sustainability and vigor of this pedagogical approach.

2. Literature Review

2.1. CLIL and Sustainability

FLT is playing an essential role in sustainable development, an emerging topic of growing interest in the international community. Particularly, learning a foreign language (L2) is related to Sustainable Development Goal (SDG) 4, which focuses on ensuring quality education and promoting lifelong learning opportunities for all [11,12]. This is because learning an L2 helps individuals communicate and interact with people from different cultures and backgrounds and can also lead to better job opportunities and improved cognitive ability [2]. Foreign language education is also related to other SDGs. For instance, it can help to promote intercultural understanding, which is important for peace, justice, and strong institutions, the essence of SDG 16 [13]. By learning about the culture, history, and language of other countries, individuals can gain a deeper appreciation and understanding of diverse perspectives, which can help to build bridges between communities and foster greater cooperation.

CLIL, one of the most popular FLT approaches, has a close relationship with sustainable education as both aim to enhance education and support the personal and professional development of individuals. SDG 4 aims to ensure inclusive and equitable quality education and promote lifelong learning opportunities for all [11]. CLIL is a teaching approach that integrates content and language learning, allowing students to learn a subject while also developing their language skills [2]. By using this approach, students are able to engage with subject-specific content in a meaningful and relevant way, increasing their motivation and involvement in the learning process.

CLIL can contribute to the achievement of SDG 4 by providing students with the necessary skills and competencies to fully participate in the knowledge-based economy and society and to lead productive and fulfilling lives. CLIL helps to develop critical thinking, problem-solving, and intercultural skills [9], which are all important competencies for lifelong learning and personal development. Particularly, CLIL can provide students with exposure to different perspectives and cultures, promoting intercultural understanding and respect [1]. This is an important aspect of SDG 4 because it aims to promote education as a means of promoting peace, tolerance, human rights and sustainable development.

2.2. CLIL in the Western World

Since the introduction of CLIL, it has been implemented across diverse socioeducational contexts. Against this background, we have seen different languages used for classroom instruction and as the focus of learning—not only English as a lingua franca but also other languages (e.g., minority languages)—presenting a multilingual perspective [14]. Meanwhile, numerous CLIL teachers have demonstrated their creativity and initiative to theme FLT through various content matters and transition CLIL from primary and secondary education to higher education (HE). An example is that CLIL, though principally implemented in secondary education in some European areas via the medium of well-established subjects (e.g., science, history, and mathematics), has been embraced in university English-Medium Instruction (EMI) programs with diverse academic disciplines [15]. The experience of these successful educational programs has exemplified how CLIL can be achieved for learners of different languages, with various content areas, and across different educational levels, positioning CLIL in a whole education agenda [16].

The CLIL research agenda has been evolving with the development of educational theories and practice. Since Coyle [17] suggests that “for CLIL research to ‘mature’, the nature and design of the research must evolve to identify CLIL-specific issues whilst drawing on a much wider frame of reference”, researchers have attempted to explore and extend the research agenda. Coyle et al. [1] propose that an ideal CLIL study should collect and examine performance evidence (student academic performance in content and language learning), affective evidence (student feelings, emotions, beliefs, and perceptions), process evidence (the learning process and classroom discourse), and material and task evidence (the learning materials and activities used in the classroom). Although this is a brief agenda, many researchers have embraced it and considered it to be valuable in organizing and evaluating a CLIL study [18]. Another classic agenda is formed by Dalton-Puffer and Smit [19], who, under the umbrella of policy, classroom discourse, and classroom pedagogy, summarize the themes of CLIL research activities, including CLIL goals, beliefs about CLIL, stakeholders’ views on CLIL in different languages, use of macro language skills, academic language use, comparison of CLIL lessons with non-CLIL lessons, and explicit language teaching episodes. To a large degree, this proposal can be seen as the extension of Coyle et al.’s [1] research framework, as both of them have reflected the necessity to seize a range of opportunities for scientific research, classroom enquiries, top–down support and bottom–up practice, and different research paradigms.

Nevertheless, in the current age of CLIL “conundrum” and “controversy” with predominant suspicion of the effectiveness of CLIL, Pérez-Cañado [3] argues that an updated overview of what is hot and what is not on the CLIL research agenda should be presented. From this perspective, she highlights the diversity in bilingual education, pluriliteracy development in CLIL, and replication, extension, and meta-analysis of previous research as the real issues for ongoing and upcoming CLIL studies, regarding egalitarianism, target language (TL) exposure, and effects of CLIL as the nonissues that should be weeded out from the agenda. With solid theoretical and empirical evidence reinforcing this proposal, Pérez-Cañado’s [3] idea does sound reasonable, though some researchers have lent dissenting voices by, for example, necessitating and justifying the ongoing research about the psychological experience in CLIL [20] and the impacts of CLIL on every possible aspect of education [21]. However, whether that is the classic agenda formulated by CLIL pioneers a decade ago or the latest one that may work well at the very moment, an undeniable problem with these conceptions is that they have been formulated based on the educational practice in the Western world, particularly in Europe. Whether they can work as expected in other socioeducational contexts, such as China, is still in question. This is also a global concern about whether the experience of Europe can be used directly in other contexts regardless of the different “sociopolitical, cultural and linguistic parameters that are involved in language policy implementation” [22].

2.3. CLIL in China

Regardless of the concern that the CLIL experience learned from the Western world may not be applicable in Asian contexts due to the divergent social, political, cultural, and linguistic parameters involved in FLT [16], CLIL has indeed gained currency in China’s educational context, with domestic development and reformation of educational policies offering top–down support and encouraging bottom–up practice [10]. Nevertheless, without a systematic snapshot of domestic CLIL programs with regard to which languages and subject matters have been chiefly embraced and how CLIL has been implemented across differing contexts and educational levels, teachers and researchers tend to have limited understanding and knowledge of CLIL, which constitutes significant obstructions to the development of CLIL in China. Occasionally, it is even believed that transitioning from existing approaches to FLT to CLIL would be bizarre and unpromising conduct in China [23]. This situation necessitates a summary of previous CLIL programs to exemplify how they have been organized and what learning effects have been exerted.

Actually, some Chinese scholars have attempted to scrutinize what has been done in previous research. Despite the consensus that Chinese CLIL researchers should engage more in classroom research to provide empirical evidence rather than repetitive interpretations of CLIL theories [24,25,26], how empirical CLIL research should be organized and move forward is still an issue of debate. Hu [10], based on a scoping review of CLIL-related publications, maintains that the vitality of Coyle et al.’s [1] research framework should be aroused because it not only corresponds to the scope of most previous CLIL research in China but also illuminates what types of research designs can be used. However, this assumption is tentative without a theoretically strong framework underpinning the researcher’s review and to some extent is a repetition of previous work without considering the special features of CLIL in China. Some Chinese scholars have also intended to nudge researchers in the directions justified by systematic reviews to be the future of CLIL studies. For example, Mi [24] contends that the focus of CLIL research should switch from the learning outcomes to the learning process, involving various stakeholders; Liu [25] asserts that future research should direct attention to the development of students and teachers as the most direct stakeholders of CLIL; Wang and Xing [26] also maintain that future CLIL research should involve students and teachers, but special attention must be paid to differentiation and teacher development. Nonetheless, a fundamental defect of their assumptions is that they have equalized Content-Based Instruction (CBI) and CLIL in the reviewing process, which have been proved to be disparate in China’s context with differing philosophies and practice [6,27]. This renders their research proposals less persuasive. In order for CLIL to continue developing and retaining sustainability, the academia should be conversant with what the latest research has evidenced. Likewise, in order to make CLIL alive in China, the context-specific features of previous research must be pointed out, the understanding of which is essential to the local evolvement of CLIL. Unfortunately, there is a paucity of knowledge about CLIL in China both in the domestic academia and in the international academic community [7], leaving a gap to be filled. To this end, this review aims to inquire into the features of recent empirical research on CLIL in China, with three research questions (RQs) as follows:

• RQ1: What are the languages and content subjects taught in the studied CLIL programs, and which educational level are they contextualized at?
• RQ2: Which areas of interest do the reviewed studies emphasize, and what are the research results?
• RQ3: What types of research designs are presented in the studies, and what are the primary data sources?

3. Method

As shown in Figure 1, this systematic review followed the method of Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), consisting of four consecutive steps (i.e., identification, screening, eligibility, and included). Researchers have widely utilized PRISMA because of its comprehensiveness, which assists in an explicit and well-articulated review of literature [28].

3.1. Identification

In the initial process of identifying documents for review, four databases were chosen, including Web of Science (WoS), Scopus, Education Resources Information Center (ERIC), and Google Scholar, as they are supposed to be popular and comprehensive databases for systematic reviews [29]. The following research strings (see

Table 1. Search strings used in the study.

Database	Search String
WoS	TS = ((“CLIL *” OR “Content and Language Integrated Learning *”) AND (“China *” OR “Chinese *”))
Scopus	TITLE-ABS ((“CLIL *” OR “Content and Language Integrated Learning *”) AND (“China *” OR “Chinese *”))
ERIC	(“CLIL” OR “Content and Language Integrated Learning”) AND (China OR Chinese)
Google Scholar	“CLIL” and “China”

) were used to identify relevant publications from 2013 to 2022 as the past decade witnessed a growth of CLIL publications in China [5]. Although there were similar content-based FLT programs in the research context under the name of, for instance, EMI and CBI, it was asserted that they were distinct from CLIL and were underpinned by dissimilar theories and philosophies [10]. Thus, only the publications themed as CLIL were considered. As a result, 1071 records were found in the databases. It should be mentioned that although Google Scholar is considered a more comprehensive database than the others, there has not been a “reliable and scalable method to extract [accurate] data” from it [30]. This has brought about a huge number of search outcomes, though only the first 1000 records were displayed [31].

3.2. Screening

With duplicates (n = 69) excluded from the initially identified articles, the rest (n = 1002) were first screened by the titles, abstracts, and keywords based on the following inclusion and exclusion criteria (see

Table 2. Inclusion and exclusion criteria.

Inclusion	Exclusion
Empirical study	Not empirical or primary research
A prime focus on CLIL	A narrow focus on CLIL/unrelated to CLIL
Studies conducted in China	Studies conducted in other contexts

). A total of 853 records were then excluded because they did not offer primary evidence collected from empirical research, had a minor focus on CLIL, or were contextualized out of China. With several publications (n = 7) unable to be retrieved from the databases, the full texts of the remaining articles (n = 142) were assessed for eligibility. As a result, 73 records contextualized in contexts (e.g., Hong Kong, Taiwan) other than Mainland China were excluded due to the different socioeducational systems, 7 records were excluded because they were non-peer-reviewed materials (e.g., theses), and 34 records were excluded because they were either review articles or opinion articles without offering empirical research evidence. The quality of the remaining publications was also assessed to ensure the research validity. In accordance with the quality assessment tool used in [32], the researchers coexamined whether a particular study answered the RQs properly and whether the presented evidence was trustworthy given the research methodology, results, and conclusion. All the publications were considered acceptable for further review.

3.3. Included

After the screening process, 28 reports were eligible for this systematic review because they presented rigorous empirical evidence collected from China’s educational context. In order to have a comprehensive review of the status quo of the CLIL research in China, different types of publications were included as long as they satisfied the eligibility criteria. They included journal articles (n = 22), conference papers (n = 4), and book chapters (n = 2).

3.4. Data Analysis

The selected articles were thematically analyzed in a software program, NVivo. They were coded from three perspectives, namely, details of the studies to indicate what was done in a research study, how the research was undertaken, and the results of each study [33]. The detailed findings can be seen in Appendix A. As per the RQs, the major themes to be located included the languages and content subjects taught in CLIL, educational level, areas of research interest, general research findings, research designs, and data sources. It should be noted that CLIL research covers a wide range of fields and topics of interest. Therefore, they were analyzed and categorized in accordance with the classic research agendas reviewed above.

4. Results

4.1. RQ1: What Are the Languages and Content Subjects Taught in the Studied CLIL Programs, and Which Educational Level Are They Contextualized at?

Given the indispensable role of language learning and content learning in CLIL, the languages and content subjects taught in the reviewed studies were examined. A majority of the research focused on English as a foreign language (EFL) learning in CLIL [34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59], with only one study contextualized in the context of learning Japanese as an L2 [60]. The study [61] was a special one, as it focused on both EFL and languages other than English (LOTE) teachers. Global language proficiency was the focus of most studies, with only a few ones foregrounding macro language skills, including writing [37,46,50,60], reading [55], listening [53], and speaking [36,46,53]. Two comprehensive studies were [41,58], in which the researchers probed into the effects of CLIL on both global English proficiency and macro language skills. However, more in-depth research dwelling on specific language knowledge was rare, with only the study [46] probing into lexical richness and complexity in English writing.

Regarding the content subjects taught in CLIL, the review displayed a high level of diversity. Common discipline areas included sociology [38,46], business [42,43,48,50,51,57], humanities [45,55], and science [40,49], while a variety of branch subjects and topics were found, such as biochemistry experiments [34], traditional Chinese medicine [36], English literature [39], law [41,58], research methodology [44], education [47], the “Belt and Road” Initiative (i.e., China’s foreign policy of building economic corridors in Eurasia) [52], Chinese culture [53,59], e-government [56], and Japanese etiquettes [60]. Some studies were contextualized in a series of CLIL programs. Thus, more than one content subject was presented by the authors. For instance, the content subjects in the study [35] included e-commerce, international trade, and marine power plant; the researchers of [54] reported over 10 subjects and themes in a content-based curriculum, such as popular science, linguistics, international trade, and Chinese classics; and 6 subjects were mentioned in the study [61], including economics, law, education, history, literature, and science.

Almost all the reported CLIL programs were organized by higher education providers (HEPs) in China for undergraduate students. Although most studies emphasized EFL learning, the contexts were slightly different, with [34,35,36,39,41,45,48,49,50,51,52,53,56,57,58,59] dealing with college English (CE) (i.e., a compulsory English course for undergraduates whose main course is not related to English language studies), [38,46,47,54,55] working with students specializing in English studies, [42,43] focusing on “3 + 1” programs (i.e., undergraduate programs wherein students spend 3 years studying in China and 1 year abroad at a partner institution), and [44] centering on EMI in a Chinese–American jointly established university. The research [60] was also contextualized in undergraduate studies, yet the language taught was Japanese. A few studies also focused on EFL but were placed on different educational levels, such as doctoral programs [40] and secondary school education [37].

4.2. RQ2: Which Areas of Interest Do the Reviewed Studies Emphasize, and What Are the Research Results?

Students’ perceptions of CLIL tended to be a major theme of the reviewed studies, though they adopted different research methods and designs. Many studies generally demonstrated learners’ positive attitudes to CLIL, portraying it as an engaging pedagogical approach to improving certain skills or academic proficiency in language learning and/or content learning [39,40,42,45,47,48,49,51,53,55,56,58,60]. However, in the study [42] themed as technology-enhanced CLIL (TECLIL), the researchers presented learners’ favorable attitudes to CLIL but suggested that some students might have a skeptical view, assuming that TECLIL was of little effect on their learning and that the features of the online tools used to support TECLIL discouraged them from active learning. Some studies specifically examined motivation as a possible outcome of CLIL, though it might overlap with learners’ perceptions because both mirrored positive feelings in CLIL. Specifically, in the study [50], the researchers investigated motivation for L2 learning and its relevant constructs (i.e., attention, satisfaction, relevance, and confidence); in the study [41], the researchers referred to the sociopsychological theory of motivation and examined learners’ motivational intensity, desire to learn English, and attitudes towards learning English. Nevertheless, the results were different, with the former study presenting that CLIL could strengthen students’ learning motivation of different types, contradicting the latter that implied that CLIL could not influence learners’ attitudes to learning and that academically weak learners might be demotivated in CLIL. Also characterized by motivation, the study [48] was an exception, with special attention to learners’ motivation, especially practical motivation, for registering for CLIL courses.

Additionally, performance evidence collected from CLIL programs was another focus of research, and almost all the relevant studies proved CLIL to be beneficial in improving TL proficiency [34,36,37,38,41,43,49,57,58,60]. As mentioned above, these studies underlined either global language proficiency or certain macro language skills, while the effects of CLIL on both of them were captured in the studies [41,58], which presented a comprehensive examination and account of the language learning outcomes in CLIL. Among these studies, an in-depth one is [38], which probed into the effects of CLIL on lexical complexity and richness in EFL writing. By contrast, research on the effects of CLIL on content proficiency is rare, with only [38,43] exploring this theme. The study [44] was a special one, as it centered around the correlation between students’ perceived learning outcomes and their satisfaction with CLIL. Some research focused on other aspects of learning or desirable skills, such as effective use of learning strategies [40], cognitive thinking skills [38,46], critical thinking [37,38,39,49], learner autonomy [37], creative thinking [51], and content proficiency [58], but these topics tended to be less popular in the research agenda.

Another group of researchers focused on the CLIL process, presenting a thorough description and analysis of the studied program and the learning tasks occurring in it. For example, in the study [40], the researchers found that student-centered and task-based activities engaged students into effective learning; in [43], the researchers disclosed that technology use in CLIL facilitated learning interactions and scaffolded content and language learning and that the different discourse patterns between academically strong and weak learners led to desirable learning outcomes; in [47], differently, the researcher recognized that the choice and organization of tasks and the cultural conflicts arising from FLT might impede learning, though CLIL activities were productive enough to promote learning; in [46], the researchers found that although CLIL could be a dual challenge of content learning and language learning, they could be mediated by cognitive development in the learning process; in [49], the researcher mentioned the effects of formative assessment on the CLIL outcomes and then brought attention to how effective assessment and learning activities facilitated learning; in [57], although the researchers pointed out the importance of scaffolding in CLIL, they also offered a detailed description of the learning process and activities designed to show that an effective CLIL course should consider students’ different learning needs and accommodated scaffolding accordingly to improve learning. To a large extent, these studies were placed in a formal learning context. This was different from the study [48], the learning process and activities in which were somehow placed in an informal learning context, underscoring students’ after-class activities (e.g., revision, translating English materials to Chinese) to cope with the cognitive challenge of learning content and subject simultaneously in CLIL.

A minor issue of interest worthy of mention was the heterogeneity in CLIL, with four studies falling into this category. One was [41], wherein the researchers compared the learning outcomes of learners of different academic levels. This study showed that both the academically strong and weak learners made considerable progress in language learning, but only the advanced learners reaped the affective benefits of CLIL. This study, to some degree, overlapped with [58], which also indicated that both strong and weak learners could improve their academic proficiency in CLIL when teachers created an authentic and differentiated learning environment. Another study was [45], in which the researcher spotlighted the heterogeneity among different types of universities (i.e., Project 985 universities, Project 211 universities, non-Project 985/211 universities, comprehensive universities, universities of science and technology, and universities with industry characteristics). The findings disclosed the stakeholders’ mixed views on CLIL and the features of the CLIL courses across different types of HEPs. The last study was [54], which revealed a tendency of polarization development in learners’ language development, as well as an increased excellence rate and failure rate in a standardized English test. Besides, lower-grade learners tended to benefit more from CLIL, with higher-grade learners not experiencing any positive or negative effects of CLIL.

The remaining studies also examined the issues related to CLIL, whereas the themes did not necessarily belong to the above categories. For example, in the study [35], the researcher stressed the analysis of the textbooks used for CLIL and argued that the existing textbooks were not suitable for CLIL due to a lack of consideration of the language of, for, and through learning and learners’ cognitive, cultural, and communicative development needs; in [52], the researchers intended to combine CLIL with massive open online courses (MOOC), analyzed available documents (e.g., lesson plans and curriculum guides) about CLIL, and summarized that a CLIL-based MOOC design should consider the educational objectives of knowledge, capacity and quality, discipline subjects, media of transmission and delivery methods, learning task, and assessment; in [55], the researchers examined learners’ attitudes towards the use of the first language (L1) in CLIL classrooms and looked into successful translanguaging strategies employed by teachers to progress learning; in [59], similarly, the researcher analyzed CLIL learners’ attitudes towards L1 use but probed into this issue by focusing on the correlations between certain learners’ factors and different situations of L1 use; in [61], the researcher focused on Chinese CLIL teachers’ competencies and presented a heterogenous professional profile in accordance with their characteristics (e.g., language taught, content taught, affiliation, education background, and years of teaching).

4.3. RQ3: What Types of Research Designs Are Presented in the Studies, and What Are the Primary Data Sources?

Many studies adopted mixed-methods designs under the warrant that the combination of quantitative and qualitative data could contribute to an in-depth understanding of the studied phenomena, while a further review revealed more specific details about the research designs. Sequential explanatory designs were often used in classroom research, prioritizing the collection and analysis of quantitative data and using qualitative data to explain the quantitative results [34,38,39,40,43,48,53,58]. This group of studies generally centered around the academic and affective learning outcomes in CLIL, which were further explained by classroom discourse and learners’ perceptions. Conversely, some researchers took an exploratory approach, which was exemplified by using quantitative data on student attitudes and performance to confirm qualitative findings about assumingly beneficial CLIL ideas or conceptions of CLIL activities [34,49]. Some researchers did not specify in the reports the types of design they employed, whereas the way they collected and discussed the data validated that they adopted a convergent parallel approach to the research. Specifically, in the study [38], although the researchers claimed that they used a classroom discourse approach, both quantitative and qualitative data were used to demonstrate the features of effective scaffolding in CLIL; in [42], the researchers reported two mixed-methods studies about TECLIL but brought the findings together to illustrate how learners could engage in an online CLIL environment to develop language and content proficiency; in [45], the researcher presented the status quo of CLIL programs across different levels of universities by merging the results about the types of CLIL programs and stakeholders’ motivation for and understandings of CLIL. The other studies in this category adopted action research methods, which featured the design of CLIL curricula based on the problems facing language learners and the evaluation of CLIL methodologies [47,55,57].

Another main group of researchers adopted a quantitative paradigm, with most studies characterized by quasi-experimental designs. Some studies included a nonrandomized control group (CG) and organized both a pretest and post-test [36,37,50]; some only had an experiment group (EG) but also conducted a pretest and posttest [41,56,60]; in [54], the researchers included a nonrandomized CG but only used the test scores after the intervention to investigate the effects of CLIL. Some studies adopted a survey approach in large-scale CLIL programs, such as [44,51,59,61], though they had very different research purposes. In comparison, there are only two qualitative studies. One is [52], wherein the researchers, against the backdrop of the “Belt and Road” Initiative, analyzed the documents (e.g., CLIL lesson plans and curriculums) collected from different universities and educational platforms, based on which they summarized the features of effective CLIL programs with top–down support and proposed a preliminary lesson planning format for CLIL. The other one is [35], which adopted a case study design and gathered data from interviews with students and teachers.

Due to the different objectives of these studies, the instruments were slightly different. However, the common ones utilized to collect quantitative data were questionnaires, evaluation tests of content proficiency, language proficiency and some other skills, performance scores of classroom activities, and so on. For qualitative research, the researchers mainly used classroom observations, interviews, documents (e.g., student works, lesson plans, curriculum guides, reflective journals), open-ended questionnaires, and so on. There were some innovative ways of acquiring data, such as online interaction records and forum log information used in [42,43] to analyze TECLIL learners’ discourse patterns in the supporting software, while most research adopted the traditional and most used data collection methods. To a large degree, CLIL students were the principal source of data [34,35,36,37,38,39,40,41,43,44,45,46,47,48,49,50,51,53,54,55,56,57,58,59,60]. Even though some studies specified that other CLIL stakeholders were also the subjects, such as classroom teachers [35,39,45,61] and policymakers [45], this body of knowledge was rather limited.

5. Discussion

First, the language taught in the studied CLIL programs was mainly English, though there was a wide variety of educational contexts (e.g., CE, English major, EMI, “3 + 1” programs) to accommodate these English and content-integrated studies. By contrast, the number of studies underscoring LOTE was limited, with only two exploring Japanese learning in university. This circumstance corresponds to the one illustrated in [10,25,62] that English language education (ELE) tends to be the predominant theme of most previous and ongoing CLIL research in China, with LOTE education receiving little attention from researchers. Undoubtedly, ELE is more important than ever before in China’s socioeducational system, with a national endeavor to facilitate internationalization at home [63]. Accordingly, continuous educational reforms have been carried out, with top–down support and bottom–up initiatives to refine the ELE curriculum. From this perspective, it makes sense that CLIL as an emergent pedagogical approach in China has been mostly embraced in EFL programs, which are usually considered the mainstream of language education. However, the negligible amount of LOTE CLIL research forces the rumination about whether LOTE, which is indeed a niche subject, has really received less support and attention than English has [64], though China has entered the golden age of multilingual development and education [62].

In contrast, the subject matters taught in the reviewed CLIL programs were miscellaneous, including not only the well-established ones (e.g., science and humanities) but also those concerning current affairs (e.g., the “Belt and Road” Initiative). Meanwhile, the content knowledge communicated by the CLIL programs was associated with the TL (e.g., English literature, Japanese etiquette) and the knowledge about China (e.g., Chinese traditional medicine, Chinese culture). This reflects that CLIL accommodates various content subjects, topics, and concepts as the objects of “real communication” to achieve the natural use of the TL [65]. None of the reviewed studies specifically foregrounded culture as another important component of CLIL, whereas it has weaved its way throughout the content subjects and topics. Thus, embracing a variety of subject matters about China and the TLs fosters the development of a “self” and “other” identity and pluricultural understanding [1]. This phenomenon can be also interpreted as that the sustainability of CLIL is being kept in China by supplementing it with Zhongguo Tese (i.e., Chinese characteristics, which refer to the knowledge about China and are learned from China, and the context-dependent characteristics and needs). This reflects the proposal that the development of CLIL in China should be embedded in the idea of Ronghui Zhongwai [62], which is to integrate the knowledge and experience learned from both China and foreign countries to localize CLIL.

Regarding the central issues in previous research, this review has presented a miscellaneous agenda. First, researchers have drawn primary attention to affective evidence, exploring the perceptions, attitudes, motivation, and other affective feelings of CLIL stakeholders, especially students. Given that the psychological experience of CLIL is still an underresearched area with much to be explored [66], this body of studies is important to understand the psychological process and outcomes of CLIL. Performance evidence is also a significant topic, focusing on a variety of learning outcomes in CLIL. However, language learning outcomes tend to be the most discussed issue, with other topics (e.g., content learning, thinking skills, learning strategies) less emphasized. This somehow confirms the presumption that almost all the CLIL research in China has put effort into language learning, with little attention directed to the other outcomes of learning [10]. It is suggested that CLIL research on the language learning outcomes and the affective side of learning has been predominant in the past few years, somehow neglecting the issues that are of the same importance to increasing the understanding of CLIL [67]. This systematic review, on the one hand, disproves this assumption because many studies have also examined the learning process of CLIL to explain its effects. On the other hand, some issues (e.g., learning materials, assessment, L1 use, heterogeneity in learning, technology use) have been indeed de-emphasized, and much remains to be explored beyond the currently limited amount of research. Centering on these research issues, the reviewed studies have presented various findings. However, it is not exaggerating to summarize that the current CLIL agenda has presented a rose-tinted scenario of CLIL application in China, depicting it as a profitable approach that will ultimately help to reach the learning goals. Compared with the Western world, where CLIL research has entered the “conundrum” and “controversy” period with research findings shedding doubt on the effectiveness of CLIL [3], CLIL in China is still in its infant “craze” period. Although some research has questioned the so-called panacean functioning of CLIL and necessitated a critical outlook on it, an extremely optimistic panorama is still dominating the current understanding of content-based FLT in China.

Notwithstanding the achievement and understanding that have been obtained about CLIL in China, the pertinent research agenda cannot be considered extensive and profound enough, especially when compared with the upsurge of CLIL research in the Western world. First, with CLIL being welcomed across different levels of education in other parts of the world [68], it is surprising to see that CLIL has been chiefly applied in China’s HE context. Indeed, most CLIL practice in China has been witnessed in HEPs instead of the other levels of schooling and education as a result of the complex educational policies and social context [10]. However, a comprehensive research agenda should invite researchers to consider the educational issues specific to the education of different levels and contexts. Moreover, to a large extent, a majority of the reviewed studies are within the scope of performance evidence, affective evidence, process evidence, and material and task evidence, though some types of evidence have captured more attention than others. Having said that, however, it does not mean that this classic research proposal is problematic. Instead, it has underpinned most previous CLIL research and is still doing so in ongoing studies [18,67]. It simply signifies that more concrete issues of interest deserve to be explored in Chinese CLIL programs. To this end, a CLIL research agenda with Chinese characteristics, which is still in its infancy, should first be based on the experience obtained from the CLIL programs in other contexts and consider the various topics of interest mentioned in the existing agendas, regardless of whether they are hot issues or nonissues. These issues are countless and are still evolving with the development of FLT theories and practice. A proper summary can be the one recently proposed in [69], which highlights three interwoven dimensions future CLIL research should dwell on, namely, designing CLIL, preparing for CLIL, and implementing CLIL. According to the authors [69], CLIL design involves “policies, curriculum development, and approaches that lend support to CLIL growth”; CLIL preparation centers upon preservice and in-service teachers’ professional development; CLIL implementation tends to be more complex; and relevant issues include but are not limited to language proficiency, teaching materials, assessment, translanguaging, and L1 literacy. However, it must be admitted that the issues involved in these dimensions are more than what has been mentioned above. Chinese researchers must base themselves and their research upon the evolvement of FLT theories and the Chinese characteristics used to complement CLIL.

The other problem with the CLIL agenda in China is a lack of in-depth studies, with most reviewed research simply “touching CLIL and going”. For example, many studies examined the effects of CLIL on language proficiency, while few emphasized the balanced development of listening, speaking, reading, and writing, which is not only an opportune research proposal [69] but also the current FLT trend being followed in China by policymakers, teachers, and students [70]. Admittedly, a study focusing on a particular language skill is sometimes more targeted than an extensive study covering several ones. However, it must be acknowledged that each language skill has different types of dimensions, knowledge, microskills, or macroskills associated with it, though sometimes they are interwoven with each other [71]. Likewise, several reviewed studies examined learners’ general motivation for CLIL, while motivation in FLT is a complicated variable composed of various constructs [72], the investigation of which is essential to evaluating the efficiency of CLIL and enhancing the quality of learning. Language learning and affective learning are simply two typical examples of the relatively cursory research agenda in China, and there are admittedly other cases wherein the researchers did not probe into the specific skills or capabilities of interest. In this case, in line with the appeal made in [73], CLIL researchers are encouraged to not only extend the research agenda but also deepen it to offer subtler understandings of the issues of interest.

The reviewed articles have also presented various types of research designs and methods of data collection as per the research objectives. A pragmatic paradigm has been found in most studies, followed by a world of view that is either positivist or interpretivist. This finding contradicts the assumption that mixed-methods CLIL studies are few in China [62] but accords with the research trend summarized in [10]. Noticeably, quasi-experimental designs have been commonly used in the reviewed studies to determine the effects of CLIL, which is a desirable phenomenon considering that CLIL studies “with pre-, post-, and follow-up assessments are still rare” [74]. However, compared with a comprehensive portfolio of CLIL evaluation measures, such as the one in [1], the research agenda in China also seems limited from a methodological perspective, encouraging researchers to extend the scope of research subjects (e.g., learners, teachers, materials, and tasks), nature of data (e.g., testing, informational assessment, portfolios, questionnaires, interviews, observations, motivational evidence, transcripts of verbal reports, material analysis, task analysis) and methods of analysis (e.g., criterion referenced, statistical, comparative, qualitative, discourse analysis). Undoubtedly, extending the breadth of research designs is closely linked with the aforementioned need to examine various issues related to CLIL. Regarding research design itself, whether that is quantitative, qualitative, or mixed methods, what really matters is the design rigor and appropriateness [1], the significance of which is that reliable data can be collected to demonstrate whether CLIL works and, if so, what works and why it works.

However, no matter how CLIL has been implemented and studied, it is vital for educators and researchers to commit to sustainable and quality education so that the vigor of CLIL can be maintained. To achieve it, teachers and educators play a vital role [75,76], who should be “empowered, adequately recruited, well-trained, professionally qualified, motivated and supported within well-resourced and efficient educational systems” [11]. Additionally, they need to optimize the subject matters key to FLT in CLIL, especially those that can enable learners to lead healthy and fulfilled lives and make informed decisions as qualified citizens, so that quality CLIL fosters creativity and knowledge and ensures the acquisition of language skills as well as high-level cognitive, interpersonal, and social skills. More importantly, we need to strengthen inputs, processes, and evaluation of outcomes and mechanisms to measure CLIL, with teacher–researcher collaboration, cooperation, coordination, and monitoring of the implementation of the CLIL agenda to be achieved in a shared network.

6. Conclusions

This paper has illustrated the essential features of CLIL studies conducted in China’s educational context, the significance of which lies in the fact that this systematic review could inform the academia of the patterns and trends of CLIL application and research. To summarize, most reported CLIL programs have been contextualized in HEPs, with language learning focusing on EFL, and content learning themed on different subject matters and topics associated with a “self” and “other” intercultural identity. The research objectives are multifaceted, while most studies have dwelled on performance evidence, affective evidence, and process evidence collected through different research designs, subjects, and instruments, presenting an inspirational scenario wherein CLIL is a promising and engaging pedagogical approach. Nevertheless, the potential of CLIL should be exploited across different educational levels and in a multilingual context, and educators and researchers should move beyond ELE in HE and embrace CLIL in more aspects of FLT. Meanwhile, the scope of the CLIL research agenda in China should be extended to cover a vast variety of issues of interest—not only the topics that have been well examined (e.g., learning outcomes, psychological experience) but also those underresearched areas (e.g., learning materials, assessment, L1 use, heterogeneity in learning, technology use)—and deepened to elicit a more sophisticated understanding of these issues. This is the direction that future CLIL research in China should take to inform the broad academia about how CLIL has been localized in China’s socioeducational context and developed in its own right with evidence-based implications drawn.

A non-negligible limitation of this review is that the articles were only extracted from the databases commonly used in the English world, without taking into account the databases (e.g., China National Knowledge Infrastructure, Wanfang Data) that are widely used in China and have indexed a vast amount of literature about CLIL and its development in the local context. This limitation, however, clearly opens up a new research opportunity to be seized in the future, particularly for comparing the research trends reflected by the literature in different languages, mainly Chinese and English, and identifying new directions for CLIL research. The challenges, opportunities, implications, and future directions identified in this systematic review, along with the ones to be pinpointed in future exploration, are key to the effectiveness and sustainability of CLIL and will spark more interest and awareness about the experience of various stakeholders in CLIL contexts.

A final note of this paper is an appeal for a more objective, precise, and context-specific understanding of CLIL in China. For the past years, we have been attempting to show researchers and scholars from other countries the very essence of CLIL in our own contexts. A popular (mis)conception that we have been advised on is that CLIL is often implemented in Chinese HEPs under the name of EMI, though our knowledge and experience inform us that CLIL and EMI are not two sides of the same coin in most cases. Although there has been a transition from CLIL in schools to EMI in universities in some European areas, voices from China are desirable to understand localized CLIL practice in the light of the divergent sociopolitical, linguistic, and cultural parameters involved in FLT. Therefore, we also hope that a thorough and context-specific comparison of CLIL and EMI will be made in the future to determine and justify whether EMI is a subsubject versus CLIL.