Research Trends in K–12 MOOCs: A Review of the Published Literature

Massive Open Online Courses (MOOCs) appeared in the area of educational technologies in 2008. Until 2013, academic research into MOOCs focused mainly on their application to adults as well as students or graduates of tertiary education. However, since 2013, the rising number of K–12 students enrolled in higher education MOOCs made MOOCs a de facto reality in pretertiary education and triggered universities, governments, and MOOC providers to (a) develop MOOCs specifically designed for pretertiary education, and (b) research their potential and value in K–12 educational settings. This resulted in a notable number of K–12 MOOCs and pilot research works in the literature that focused on the potential of MOOCs in compulsory education settings, as well as on their ability to reshape and transform the current educational K–12 framework. This work seeks to (a) trace, analyze, and review the existing literature on K–12 MOOCs, (b) identify representative MOOC implementations, (c) classify and organize research trends and patterns, and (d) reveal MOOCs’ potential value and impact on K–12 settings. The research used a narrative literature review methodology in order to critically review and qualitatively analyze twenty-one research publications in a systematic manner. Analysis of relevant works demonstrated that MOOCs, under a set of prerequisites, can be effectively incorporated into and positively affect pretertiary education.


Introduction
Massive open online courses (MOOCs) are online courses aiming for unlimited participation and open access to knowledge via the Web (Kaplan & Haenlein, 2016). Since their first appearance in 2008, their popularity has increased rapidly, and now it is estimated that more than 11 thousand such courses provide free and open educational content to more than 100 million learners worldwide (Shah, 2018).
During these years, research on MOOCs has mainly focused on their application to adults as well as students or graduates of tertiary education, while research on the potential, role, value, and benefits of this new learning tool in pretertiary education was absent until 2013 (Ferdig, 2013).
Since 2012, which was called the year of the MOOC (Pappano, 2012), the growing popularity of MOOCs in higher education triggered discussions about their potential role in lower educational levels (Briggs & Crompton, 2016), but it was the rising number of K-12 students who enrolled in MOOCs designed for tertiary education that made them a reality in compulsory education (Atkeson, 2014;Guo & Reinecke, 2014;Stoltzfus, Scragg, & Tressler, 2015). This increment of K-12 students' participation in MOOCs resulted in a concurrent increase in research interest on K-12 students' perceptions of and experiences with this new, digitally advanced learning tool, as well as the benefits from MOOCs' use in pretertiary education in general (Briggs & Crompton, 2016;Dermirci, 2014;Horn, 2014;Norris & Soloway, 2012).
The use of an advanced digital learning tool such as the MOOC in K-12 education is beneficial, not only due to the digital advantages it provides, but also due to the remarkable potential to provide enriched learning opportunities (Briggs & Crompton, 2016). However, current MOOC types aimed at adult learners often don't meet the needs of K-12 students. The MOOC model for K-12 education must be consistent with pedagogical methods that best fit the ways Κ-12 students learn, in order to meet the needs of the K-12 educational setting (Bock & O'Dea, 2013;Briggs & Crompton, 2016;Locke, 2013).
Anecdotal research has demonstrated that MOOCs can be used in pretertiary education as a supplementary resource to traditional instruction, taking a blended approach (Atkeson, 2014;Pannoni, for younger audiences, as well as for K-12 educators. Day by day, more and more K-12 MOOCs are made available and can be integrated into school practice and policy, making research on their potential role and value in pretertiary education much easier than in the past. At the same time, an increasing number of academic papers on K-12 MOOCs appears in the peer-reviewed literature. Since 2013, the few pilot research works that have been carried out have highlighted: (a) the considerable potential MOOCs have to play in K-12 education, and, at the same time, (b) the need for further research and experimentation in order to identify and evaluate the opportunities, benefits, risks, uses, and actual value of this tool in the settings of K-12 education (Yin, Adams, Goble, & Madriz, 2015;Wartell, 2012).
Aiming to contribute to this promising research area, this paper seeks to trace, analyze, and review the existing literature on K-12 MOOCs, identify trends and patterns, and classify the academic research on pretertiary educational MOOCs.
Through this process, our work addresses the following research questions: Q1. What are the most representative K-12 MOOC implementations?
Q2. What are the research trends in the K-12 MOOC field? Q3. Do MOOCs have a positive impact on K-12 education?
The paper is organized into four sections. Section 1 presents the rationale and defines the objectives and scope of this research. Section 2 presents the methodology used to gather and analyze the literature.
Sections 3 and 4 present, analyze, and discuss the research findings and conclusions based on the criteria established by the research questions.

Methodology
In this section, we describe the methodology used for our work and, more specifically, the methods by which we gathered literature on K-12 MOOCs (data collection) and examined, classified, and analyzed that literature (data classification and analysis).
In our research, we used a narrative literature review methodology to identify and summarize what has been previously published, develop a comprehensive overview of K-12 MOOC research, and reveal new study areas not yet addressed (Ferrari, 2015). Such reviews benefit researchers in planning future studies, as well as developing convenient summaries of the literature on a particular issue (Petticrew & Roberts, 2008). Ferrari (2015) claims that the quality of a narrative review may be improved by borrowing elements from the systematic review methodologies. In this respect, the following three key activities for systematic research proposed by Gough, Oliver, and Thomas (2012) were applied to our approach: 1. identifying relevant research 2. critically reviewing and qualitatively analyzing the identified research works in a systematic manner, using evidence tables (Green, Johnson, & Adams, 2006), and 3. synthesizing research findings into a set of conclusions.

Data Collection
Literature discovery searches were systematically conducted, using the key words "MOOC," "MOOCs," "Massive Open Online Course," and "Massive Open Online Courses" in conjunction with "K12," "K-12," "secondary education," "compulsory education," and "high school." To be included in the corpus, each identified document had to • focus on MOOCs used in pretertiary education and address K-12 students; • have been published in a peer-reviewed journal or in conference proceedings, and present primary research studies; • have been published or been made available online between January 2013 and March 2020; and • have been written in the English language. Next, we focused our research on educational technology resources, and we used the LearnTechLib Library (https://www.learntechlib.org/) and the Educause Library (https://library.educause.edu/).
Subsequently, we conducted a forward referencing process, similar to the one proposed by Veletsianos and Shepherdson (2016) and Liyanagunawardena et al. (2013), in order to identify relevant research that had cited the papers we had already located. We first located in Google Scholar each of the papers included in our corpus. Google Scholar provides information on how many times a paper is cited and allows researchers to view all papers citing the original. If these new articles met our inclusion criteria, we included them in our corpus. Finally, we examined the references of the papers in our corpus, which were published after 2013, in order to identify any we might have missed.
This data collection process resulted in the identification of twenty-one distinct published works in total, including seven papers published in academic journals and fourteen papers in conference proceedings.

Data Classification and Analysis
Each of the twenty-one identified publications on K-12 MOOCs was read and analyzed by the authors in a systematic and consistent manner. During this process, for every publication, the authors' team identified and systematically recorded: 1. the focus of the research work 2. the aim of the research work 3. the research questions that each work attempted to address 4. the research methodology 5. the aspects related to K-12 MOOC implementation, and 6. the research findings, identified results and conclusions.
The collected data were organized to address the three research questions. In this respect, each piece of evidence drawn from the twenty-one papers was extracted in the same fashion to help decrease potential bias. Evidence tables were used in order to identify similarities and differences in the analyzed studies.
These tables are presented in the following section.

Review of the Literature and Discussion on the Research Findings
Research Question 1: What are the Most Representative K-12 MOOC Implementations?
The main goal of this study was to trace and classify the existing K-12 MOOC implementations that emerged during the data collection process. The classification analysis focused on three key aspects: (a) the country where the MOOC is implemented, (b) the teaching subject of the MOOC, and (c) the provider and the learning platform where the MOOC is hosted. Table 1 shows the K-12 MOOC implementations classified based on these criteria. Dinosaur Paleobiology MOOC Canada The data collection process showed that a significant number of K-12 MOOC implementations are STEM-related courses. More specifically, in Spain, Dziabenko and Persano Adorno (2017)  Most of the collected research papers deal with computer-science MOOC courses. In Germany, Grella, Staubitz, Teusner, and Meinel (2016)

Research Question 2: What are the Representative Trends in K-12 MOOCs?
The second goal of this study was to identify research trends and patterns in the field of K-12 MOOCs, through analysis of the research aims and objectives of existing studies. The classification analysis focused on the role that the K-12 MOOC played in the educational process and its resulting value in the K-12 setting.
Analysis of the collected papers as shown in Table 2 revealed three different ways in which MOOCs were successfully implemented into existing school infrastructures: (a) as foundation or advanced placement (hence AP) courses to help students in their preparation for higher education, (b) as supplementary courses to the existing school curriculum, or (c) as additional advanced courses, to offer new teaching subjects at a level beyond the K-12 setting.

MOOCs implemented into the existing school infrastructures as foundation
courses to help students in their preparation for higher education. The data identified four MOOC implementations as foundation courses that support students preparing for college.

Among them, Canessa and Pisani (2013) implemented a High School Open Online Course (HOOC)
specifically designed to support the training and basic scientific knowledge of young students. The course helped prepare students for higher education in math and physics. The main goal of the research was to uncover students' and parents' experiences and opinions about the ΗOOC. The researchers attempted to identify the implications and effects on engagement when high school students were allowed to watch again, at their own place and pace, the same lessons of physics and mathematics held in the classroom. Besides that, the researchers tried to obtain feedback from students on the use and effects of implementing a MOOC in the school setting. There was a quantitative evaluation of the course from students and parents, and qualitative analysis of students' experiences after attending.
In a similar approach, Najafi, Evans, and Federico (2014) integrated a university preparatory economics MOOC not explicitly developed for high school students into a school-based course.. They investigated the engagement of high school students with the learning and assessment components of the course.
More precisely, they examined the potential absence or presence of teacher's support in students' learning outcomes, engagement, and persistence during the course. They used a quantitative approach, with pre-and post-questionnaires comparing MOOC-only and blended-mode students.

Tomkins, Ramesh, and Getoor (2016) researched implementation of a preparatory MOOC for Advanced
Placement exams in the K-12 context. The students were provided with a year-long computer science high school MOOC. The aim of their research was to understand the applicability of the MOOC model to high schoolers and, at the same time, to gain a better understanding of the teachers' roles as coaches.
The research used a machine-learning model to predict students' achievements on AP exams based on their course, forum data, and learning environment.
Finally, Koutsakas et al. (2020) developed and openly offered a computer programming MOOC specifically designed to prepare secondary students for national exams for Greek higher education. The research aimed to observe and analyze students' motives for participation and reasons for early dropout, learning expectations and attitudes toward MOOCs, and students' behaviour within the context of the MOOC. Additionally, the research attempted to determine whether students with special learning disabilities participated in the course and how, if at all, the MOOC impacted them (Politis, Koutsakas, & Karagiannidis, 2017). The research used quantitative analysis of students' learning analytics, and beforeand after-MOOC answers (Koutsakas et al., 2020;Koutsakas et al., 2018).

MOOCs implemented into the school infrastructures as a supplementary resource
to the existing curriculum. The data collection process identified twelve MOOC implementations that provided supplementary resources to the existing school curriculum. Five of them were STEMrelated MOOCs and seven were computer programming MOOCs.
The first of the five STEM MOOCs was developed by Khalil and Ebner (2015), which provided a physics, In another research work that adopted a similar approach, Grover, Pea, and Cooper (2016) provided a computer programming MOOC specifically designed for K-12 that aimed to introduce middle-school students to algorithmic thinking and programming. The aim of the research was to identify the variation across middle-school students in learning of algorithmic flow of control, the factors that influence these learning outcomes and the conditions under which the adoption of MOOCs can support learning and teaching in computer science in this educational level. The research used a two-step iterative empirical approach, with the first iteration taking place in a traditional face-to-face classroom setting, and the second in an online version of the MOOC that used a blended model of learning.
Furthermore, Hermans and Aivaloglou (2017) provided an introductory Scratch computer programming MOOC specifically designed for high school students in order to teach programming and software engineering concepts to K-12 students. The research aimed to investigate students' difficulties in programming and engineering concepts, age-related differences in students' performances, and predictions on course completion. The research used a qualitative research methodology, including evaluation and feedback on a weekly basis.
In a similar approach, Staubitz, Teusner, and Meinel (2019) offered two MOOCs specifically designed for high school students in order to support teaching and learning computer programming (Python and Java). The aim of their work was to investigate the use of MOOCs as an instrument to support computer science teachers in secondary schools. The research conducted both quantitative and qualitative analysis.

MOOCs implemented into the existing school curricula as additional (advanced)
courses to offer new subjects beyond the K-12 level. The data identified three MOOC implementations developed as additional (advanced) courses to offer new subjects at a level beyond the K-12 setting.
The K-12 MOOC presented in the research of Yin, Adams, Goble, and Madriz (2015) used an undergraduate university level MOOC in order to investigate students' learning experience. This MOOC implementation was slightly different from the previous MOOCs, as it was targeted at both K-12 students and their parents. The MOOC was about dinosaurs and was provided as an option beyond the school curriculum. The research aimed to gather, analyze, and evaluate the complex realities of children's everyday experiences in the MOOC. The research used a qualitative methodology and generated data via in-depth phenomenological interviews with 12 child-parent couplets from around the world. Nigh, Pytash, Ferdig, and Merchant (2015) implemented and offered a cMOOC to high school students interested in teaching. Apart from providing a new subject outside the existing school curriculum, the cMOOC addressed students' professional development. It was also offered to pre-and in-service teachers who actively participated and enriched the learning community with their experiences. The purpose of the research was to observe and analyze in-depth K-12 students as they participated, while exploring pre-and in-service teachers participating for professional development reasons. The research collected and analyzed both quantitative and qualitative data.
Finally, Kurhila, and Vihavainen (2015) offered a university computer science MOOC as an extracurricular elective course for high school students, aiming to alleviate the lack of computer science education in Finnish schools. The research aimed to analyze Finnish high school students' behaviour in a university-level computer science MOOC, examine the differences between school students and others, and measure how the participants perceived the difficulty and educational value of a MOOC. The research used a quantitative methodology.

Research Question 3: Do MOOCs Have a Positive Impact in K-12 Education?
The third goal of this study is determine whether MOOCs have a positive impact on the K-12 education field. In order to identify the potential of MOOCs in compulsory education and the practical and conceptual benefits that MOOCs have to offer to both teachers and students, we followed a classification analysis that emerged from the research findings and conclusions of the existing K-12 MOOC implementations.
Although much of the K-12 MOOC research is still in the pilot stage, all scientific papers and articles found in the corresponding literature emphasize the potential of MOOCs to reshape the existing school infrastructure by providing a unique learning experience to both students and teachers. On the other hand, the existing MOOC implementations uncover the complex realities and issues of incorporating MOOCs into pretertiary education. High school students have a particular modeling demand as compared to adult learners. Thus, we have to examine the set of prerequisites under which the adoption of MOOCs may have a considerable impact on K-12 education.
Researchers examined in detail students' perceptions, expectations, opinions, attitudes, improvement, and satisfaction regarding participation in MOOCs in order to provide insights into the potential impact of MOOCs on K-12 education. An analysis of the collected papers revealed that all students strengthened their knowledge in the teaching subject (Grover, Pea, & Cooper, 2016;Najafi, Evans, & Federico, 2014), improved their performance, and systematically scored higher in their exams (Canessa & Pisani, 2013;Najafi, Evans, & Federico, 2014). MOOCs could be a valuable tool to support students in their studies (Canessa & Pisani, 2013;Author, 2020), through the provision of valuable content and different learning strategies that could contribute to a positive transformation of education in general (Grella et al., 2016;Nigh et al., 2015). These courses can provide both students and teachers different learning opportunities and experiences and support new relational configurations (Nigh et al., 2015;Yin et al., 2015). Most students had a very positive overall experience with MOOCs (Author, 2020; Staubitz, Teusner, & Meinel, 2019), and they found them quite challenging (De Kereki & Manataki, 2016).
While MOOCs could be a feasible option for K-12 students, there are several concerns about the use of this new learning tool in K-12 education. The analysis of the collected papers showed these concerns could be mitigated in order to provide enriched opportunities for both teachers and students (Staubitz, Teusner, & Meinel, 2019). The main concern is the ability of young students to assume more control and responsibility for their learning in a self-study manner. While a significant number of K-12 MOOC implementations show promise regarding the integration of MOOCs into existing school infrastructures, only highly motivated students thrive in these MOOCs (Canessa & Pisani, 2013;Kurhila & Vihavainen, 2015;Najafi et al., 2014;Tomkins et al., 2016).
In order to address this inequality, research has demonstrated that the best way to implement a MOOC in K-12 education is to blend the MOOC into the traditional classroom, using a didactical approach (i.e., a blended or flipped classroom model) in order to motivate students (Grella et al., 2016;Khalil & Ebner, 2015;Najafi et al., 2014;Panyajamorn et al., 2016;Tomkins et al., 2016). A teacher's presence in these courses is a core element for student achievement because it may positively influence students to remain on track (Khalil & Ebner, 2015;Koutsakas et al., 2018;Najafi et al., 2014;Tomkins et al., 2016).
Additionally, MOOCs of shorter duration increase the completion rate, and the use of badges of accomplishment encourage students to continue to participate in MOOCs (Khalil & Ebner, 2015;Staubitz, Teusner, & Meinel, 2019). Finally, using social media to promote interaction, collaboration, and contribution to the course showed a further way of mitigating some of the concerns (Koutsakas et al., 2018;Panyajamorn et al., 2016).
Using MOOCs in K-12 education is quite challenging but offers an unmatched learning experience to both students and teachers. The existing K-12 MOOC implementations are quite promising and highlight the positive impact that MOOCs offer both teachers and students.

Conclusion
The proliferation and success of MOOCs in higher education have led to discussions about the potential of this modern learning tool in K-12 education. The unique digital advantages of MOOCs in conjunction with the rising number of K-12 students who enrolled in typical MOOCs for tertiary education leverage both MOOC providers and standalone researchers to develop MOOCs explicitly targeted at K-12 learning environments. This paper attempted to shed light on the existing literature about K-12 MOOCs in order to identify the most representative MOOC implementations, to classify and organize research trends and patterns, and, consequently, to reveal MOOCs' potential value and impact in K-12 education.
We used a qualitative narrative literature review methodology in order to develop a comprehensive review of K-12 MOOC research.
The analysis of the relevant work on K-12 MOOCs demonstrated that a significant number of K-12 MOOC initiatives are STEM and computer science courses, implemented mostly in Europe and the USA.
As far as we can deduce from the collected papers, the vast majority of K-12 MOOCs has been designed explicitly to meet the needs of K-12 students in order to extend and enhance the existing school curriculum. The analysis of these studies identified students' motivations, expectations, opinions, experiences, and satisfaction regarding their participation in MOOCs in order to gain insight into the issue of using MOOCs in K-12 learning environments. It appears from the review that the most effective option is to embed MOOCs into the existing school infrastructures in a blended approach. It is concluded that MOOCs, under a set of prerequisites, can effectively be incorporated into and positively affect pretertiary education. Teacher's presence, social interaction, and guided learning in a blended school environment seem to be core elements to ensure MOOCs benefit both students and teachers.
Incorporating MOOCs in the area of K-12 education is still in its infancy and many questions remain unanswered. Further research in the field of design, development, implementation, provision, and evaluation of such courses in compulsory-age education is needed in order to uncover the complex realities of K-12 students and teachers participating in the world of MOOCs.