Idea Improvement and Socially Shared Regulation Matter in Cross-Cultural Online Collaborative Learning

Abstract

Cross-cultural collaborative learning has been paid more and more attention in recent years. To promote productive cross-cultural collaborative learning, idea generation and improvement, and socially shared regulation is crucial. The study aimed to identify the differences in idea generation and improvement as well as socially shared regulation between high- and low-performance groups in cross-cultural online collaborative learning. In this study, 24 culturally diverse university students composed of eight groups conducted cross-cultural online collaborative learning to solve problems collaboratively. Epistemic network analysis and lag sequential analysis methods were employed to analyze data quantitatively and qualitatively. The results revealed that different strategies shape different learning performances. High-performance groups adopted more cognitive, social, and regulation processes than low-performance groups. The results extend the existing literature by indicating that idea elaboration, refining or building on ideas, and appraisal is strongly connected to new ideas. In addition, transitions from monitoring and controlling to adapting metacognition in collaborative learning activities are the main difference in socially shared regulation between high-performance and low-performance groups. This study shed light on how to engage culturally diverse students to generate and improve ideas as well as jointly regulate collaborative learning.

Keywords

cross-cultural collaborative learning idea generation idea improvement socially-shared regulation

Introduction

In the globalized world, information and communication technologies have enabled online learning possible and promising. Online learning is becoming more and more essential means of education in recent years since learning has gradually been shifting from individual learning to learning with and in a group (Järvelä et al., 2015). One of the well-studied strategies for effective online learning is online collaborative learning, a method that encourages learners to learn together at anytime and anywhere. Online collaborative learning supported by computer technologies creates a learning environment that enables students from different cultural backgrounds to learn together irrespective of distance termed cross-cultural collaborative learning. One of the aims of online collaborative learning is to promote 21st-century skills such as collaboration skills, cross-cultural competences, and ICT skills (Child et al., 2016; Laal et al., 2012; Larson et al., 2011). Furthermore, online collaborative learning emphasizes shared goals, shared tasks, and shared strategies in a shared environment with the support of technology (Collis, 2004; Jeong et al., 2019; Jeong & Hmelo-Silver, 2016). Online collaborative learning groups vary from small to large group sizes (Kumi-Yeboah, 2018; von Davier & Halpin, 2013).

According to Yamazaki and Kayes (2004), instructors and educators need to view cross-cultural collaborative learning as a means of building relationships with others through interaction, valuing individuals from different cultures, listening and observing the host culture, and making sense of complex information and translating personal thoughts into the host culture. During collaborative learning, individuals share ideas to build on their knowledge through social interaction and communication (Zheng, 2017). However, knowledge building does not evolve easily during collaborative learning (Kuhn, 2015). For an effective learning process and outcomes, team members’ ability to regulate their processes to accomplish a given task is important (Ucan & Webb, 2015). Hence, scholars (such as Iiskala et al., 2015; Khosa & Volet, 2014) have investigated how collaborative learning groups regulate their discussion to build on their knowledge.

Although there are numerous studies on socially shared regulation and idea generation and improvement in online collaborative learning, there is a lack of studies on how learners generate and improve ideas as well as conduct socially shared regulation in cross-cultural collaborative learning context. Therefore, understanding how cross-cultural groups regulate online collaborative learning to accomplish group tasks will significantly contribute to the study area. This study aims to investigate how learners jointly regulate cross-cultural collaborative learning as well as generate and improve ideas to build on their knowledge in a cross-cultural online collaborative learning context.

Literature Review

Cross-Cultural Collaborative Learning

Culture could be viewed as an influence on learning perception and knowledge construction, where an individual’s country and educational background can shape their strategies and presentation of ideas (Dai, 2019). A lack of understanding of each other’s cultures is likely to result in conflict (Hofstede, 1986). Culture plays a significant role in how people think, behave, and interact with others from a different culture (Gyasi et al., 2021). According to White (2002), cultures evolve through social interaction and shared knowledge but, differences in culture might lead to different patterns of learning, use of learning activities (Marambe et al., 2012), and learning behaviors. Task understanding and English language proficiency could be a challenge as a result of cultural background (Jeannin, 2013).

Several studies confirm the effectiveness of the application of different instructional strategies on learners in a cross-cultural collaborative learning setting. For instance, Spires et al. (2019) conducted a study on cross-cultural collaborative inquiry between Chinese and U.S. high school students. They found that it promoted learning and engagement. Similarly, Herlache et al. (2018) adopted a role-play strategy to increase students’ knowledge, negotiation skills, attitude, and confidence in cross-cultural collaborative learning. However, cross-cultural groups face challenges which have motivated scholars to adopt different strategies to curb these challenges. For instance, students from individualistic cultures tend to focus on completing tasks while students from collectivist cultures usually focus on building relationships (Liu & Chen, 2020). Therefore, Kirchner and Razmerita (2015) involved 75 students from Denmark and Germany in their study with different technology tools such as MOODLE, Skype, Facebook, Prezi, Dropbox, WhatsApp, Email, and GoogleDocs/Drive to support collaborative learning. Similarly, Spires et al. (2018) engaged American and Chinese high school participants to utilize problem-based inquiry to promote cross-cultural collaborative learning.

Furthermore, other studies also examined differences in learning performance by comparing cultural backgrounds. For example, Popov et al. (2014) engaged students with individualistic and collectivist backgrounds in paired collaborative learning and found that individualistic students performed better than collectivist students. In addition, several cross-cultural studies focused on students’ perception (Lee & Markey, 2014; Lin & Gao, 2020; MacLeod et al., 2017) or a cross-cultural competence (Liu, 2007; Puteh et al., 2020; Su, 2008; Syzenko & Diachkova, 2020). Other studies also focused on the steps to design cross-cultural collaborative learning (Kumi-Yeboah, 2018; Yang et al., 2014). These studies have contributed immensely to the research domain, however, there are still aspects to discover. For instance, the differences between high and low performing groups during cross-cultural collaborative learning still remain lacking. Previous studies explored the impact of culture and effects of intervention on students’ behaviors and perception (Lee & Markey, 2014; Su, 2008). For instance, S.-J. Chen et al. (2006) investigated the cultural influences on students’ online learning and social behaviors and their perceptions. Other studies that analyzed students’ group discussion data and investigated students’ behaviors among cross-cultural groups (e.g., Iiskala et al., 2015). However, previous studies have not investigated how learners generate and improve ideas in cross-cultural collaborative learning context. Hence, there is a need to further explore the ideation processes of cross-cultural groups in collaborative learning activities that could promote learning performance.

Connectivism

Connectivism as a theory was first introduced by Siemens (2004), as a model that views learning as knowledge distributed within a network that is socially and technology enhanced with the ability to recognize and interpret patterns. It is one of the prominent learning theories that has gained recognition in the e-learning research domain (Goldie, 2016). Knowledge is viewed as a distributive process that is dynamic and can be retrieved from interactions with individuals, technologies, and communities. Connectivism asserts learning as a process of pattern recognition and knowledge as a network (Apostolidou, 2022). Thus, knowledge is a network formed by actions and experiences (Tham et al., 2021). Connectivism is one of the theoretical basis for digital learning where connections between the learner, learning resources, and digital information are promoted by technologies (Saadatmand et al., 2012). Connectivists views knowledge as something that is emergent and comes through interactivity rather than content (Downes, 2019). Therefore, engaging learners from diverse cultural background connected digitally to each other to share and improve ideas will shed more light not only in the area of cross-cultural collaborative learning domain but also the ideation research domain.

Idea Generation and Idea Improvement

Collaborative learning activities engage students in tasks that allow for idea generation and improvement (Lam, 2019). Idea generation and improvement involves cognitive, metacognitive, and social processes. Interactional perspective-taking behaviors such as questioning, elaboration, and defending ideas are productive for idea generation (IG) whereas perspective taking through mutual recognition bridges social cultural background issues, resulting to idea generation (Hawlina et al., 2017).

Idea generation is a cognitive process that involves knowledge retrieval from long-term memory and integration into the working memory through the moderation of social and motivational factors (Nijstad & Stroebe, 2006; Paulus & Brown, 2007). Idea generation involves setting an environment and implementing creativity techniques that will help students produce, express, and merge ideas (Escandon-Quintanilla et al., 2015). Cognitive and metacognitive skills are significant thinking processes that enables students to think divergently and solve problems (Kelley, 2008). Research on group learning have indicated that social interaction among peers promote cognitive activities through which learning takes place (O’Donnell & King, 1999). Accordingly, there are tasks such as comprehension require recall and application of concepts, however, tasks such as problem solving require high order thinking (King, 2002).

Furthermore, metacognitive processes enables team members to select, evaluate, and correct ideas which are significant for creative ideas (Jia, 2019). Metacognition can be referred to as an individual’s ability to be aware by knowledge and control their cognitive processes (Nelson & Narens, 1990). According to Ellamil et al. (2020), metacognition is likely to play different roles in idea generation and improvement such that a low level metacognition could make individuals process more information and consequently construct new ideas. It involves regulation of cognitive processes by planning, monitoring and controlling, and evaluate cognition.

Moreover, generating ideas involves social process as previous studies have found that through collaboration among group members innovative works and ideas were birthed (Glăveanu, 2018; Kay Siu, 2012). Some group communication patterns such as giving attention to shared ideas, taking mutual perspective, refining, and building on ideas lead to effective and productive team work (Paulus et al., 2010). When students collaborate in an online learning environment, it is significant to examine idea generation and improvement to reveal how students’ performance is influenced by CL (Ng et al., 2022). According to Breen (2015), it is a democratic process were team members share different perspectives based on experiences and individual observations. Previous studies have focused on the various techniques for improving creativity and performance. For instance, Acar et al. (2019) explored the constraints and mechanisms that inhibits the generation of creative ideas. Wang (2019) focused on developing a taxonomy for idea generation techniques that promote creativity. According to Scardamalia (2002), idea improvement is a principle that directs students and teachers’ effort in knowledge building activities. In knowledge building process, team members engage in idea sharing, giving appraisals about important ideas, and further searching for new information to improve ideas (Kawakubo et al., 2021). Previous studies found a difference in knowledge building processes among high and low-performance groups (Scardamalia, 2002; Zheng et al., 2021). A recent study investigated how high school students generate ideas in a scientific task and discovered that high-performance groups applied divergent thinking and regulation to plan and monitor their ideation processes (Sun et al., 2022). Sun et al. (2022) found that high performing groups applied cognition and regulation strategies more than the low-performing groups. Therefore, investigating idea generation and improvement processes among university students would add more knowledge to existing literature. However, very few studies examined idea generation and idea improvement strategies in cross-cultural collaborative learning context. That is, how cross-cultural groups generate and improve ideas to build on knowledge is still understudied. Moreover, the cognitive, metacognition, and communication processes of cross-cultural collaborative groups remain unclear. Therefore, investigating how students apply these strategies to generate and improve ideas would contribute to adding existing knowledge.

Socially Shared Regulation

Regulation of collaborative learning as a social activity is a meta-cognitive goal-oriented action where collaborative teams apply different strategies to monitor and control their behaviors, emotions, direction of the discussion, and meta-cognition through interactions (Hadwin et al., 2019; Miller & Hadwin, 2015). Socially shared regulation refers to the processes that team members monitor and control, reflect and evaluate their collaborative learning activity (Järvelä et al., 2015). Socially shared regulation of learning (SSRL) occurs at a group level where learning is negotiated by learners with control during the collaborative learning process (Isohätälä et al., 2017). An individual’s ability to regulate self-learning and group-learning is a significant 21st century skill (Järvelä et al., 2015). Hence, scholars have explored different significant aspects of social regulation in collaborative learning. For example, Malmberg et al. (2015) explored how groups progress in SSRL in a computer-supported collaborative learning environment. The findings suggested that the SSRL focus shifted from regulating external challenges such as time management toward the cognitive and motivational aspects such as task completion. Furthermore, another study compared the differences in regulation between same cultural groups and mixed cultural groups, thus Canadian, Chinese, and Canadian and Chinese pairs (Shi et al., 2013). Their findings indicated that Canadian pairs generated high self-regulation behaviors which was similar to the mixed pairs. In addition, Zheng et al. (2017) developed a tool that supports the social regulation of collaborative learning to promote goal setting, planning, and knowledge building. A minimal number of studies have visually analyzed socially shared regulation (Isohätälä et al., 2017) among high and low performing cross-cultural groups during collaborative learning. Moreover, very few studies have investigated the SSR behavioral patterns of cross-cultural groups in relation to idea generation and improvement processes.

High and Low Performance Groups

Research in the field of collaborative learning have identified differences between the high and low performing groups. Team members from diverse backgrounds use different approaches during collaboration. For instance, in a problem-solving group activity, Lau et al. (2021) identified high occurrences of task related keywords and behaviors such as spending more time on problem solving, followed by personal learning and last collaboration among top 5 teams. Whereas, the bottom five performing teams focused on collaboration, personal learning, and last problem-solving. Sampson and Clark (2011) identified high-performance groups voiced more content-related ideas, discussed, and attended to team member’s ideas before accepting or rejecting, challenged each other’s ideas more than seeking for information, adopted scientific strategies for justifying evaluated ideas, and utilized provided resources to guided and evaluate their written arguments compared to the low-performance groups. Wang and Kuo (2016) identified differences in cognitive styles (visual and verbal) between the high and low-performance groups and they found that groups with verbal cognitive styles showed better annotation behaviors and quality had high performance compared with the groups with visual cognitive styles. Furthermore, high performing teams displayed more confusion and frustration which transforms into delight and neutral emotions whiles low performing teams displayed more boredom based on facial expressions (Sharma et al., 2021).

To identify group performance in collaborative learning activities, scholarly studies used group products, group discussion transcripts or recordings for tasks and social related activities or post-test (Janssen et al., 2005; Liu & Chen, 2020; Popov et al., 2019). To illustrate, Zhang et al. (2022) used the discourse data of students to evaluate their problem solving competencies and processes with ENA, which revealed that both high and low performance groups maintained positive communication, but the high performance groups begun with positive communication and ended by negotiating ideas whiles the low performance groups begun with sharing resources or ideas and ended with regulating problem solving activities. Therefore, to evaluate group performance, measurement are required based on what is being evaluated.

Research Purposes and Questions

Generating and improving ideas involves multiple processes in different dimensions during collaborative learning. First, idea generation is a cognitive process of sharing and merging of ideas (Escandon-Quintanilla et al., 2015). Second, idea generation involves social communication processes as it promotes idea sharing and social interaction (Paulus et al., 2010). Third, the metacognitive processes of groups may lead to information processing and building on ideas (Ellamil et al., 2020). In cross-cultural collaborative learning context, research on how students’ complete problem-solving tasks through social communication processes to generate and improve ideas is underdeveloped. To the best of our knowledge, there are no studies that have investigated both cross-cultural collaborative learning ideation processes and socially shared regulated behaviors. Therefore, there is a need to explore the features of idea generation and improvement processes that could promote good performance in cross-cultural collaborative learning. Discovering the differences in socially shared regulation behaviors of cross-cultural groups as they generate and improve ideas would be an up-to-date contribution to the research domain. Therefore, this study aimed to explore whether the high-performing cross-cultural groups differ from the low-performing groups in idea generation and improvement and socially shared regulation behaviors. In addition, it was deemed appropriate to further examine how these groups collectively regulate cross-cultural collaborative learning to achieve learning objectives. Therefore, the research questions were addressed as follows:

(1) How do learners to generate and improve ideas during cross-cultural online collaborative learning?

(2) Do high- and low-performance groups differ in idea generation and improvement processes during cross-cultural online collaborative learning? If so, what are the variations in their strategies?

(3) Do high- and low-performance groups differ in socially shared regulation behaviors during cross-cultural online collaborative learning?

Method

Participants

This study was conducted with international students enrolled in a university in Beijing, China. The participants were 24 masters and PhD students constituting 11 females and 13 males. Participants were nationals of Pakistan, Canada, Turkmenistan, Ghana, Cameroon, Guinea, Uganda, Tanzania, Botswana, South Africa, and Senegal. All participants were enrolled in an English-taught program. Participants had no prior experience in cross-cultural collaborative learning in an online or social media platform. They were randomly assigned into eight groups with three members per group in this study. All participants gave their consent to participate in this study without compulsion and were assured that their personal information such as names will be anonymized.

Procedure

The procedure of this study included three phases. The first phase is to conduct pre-test online to examine the prior knowledge of participants which lasted for 30 minutes. The second phase is to conduct cross-cultural online collaborative learning through a social media software, namely WeChat that is very popular in China. The cross-cultural online collaborative learning lasted for 2 weeks during which cross-cultural online collaborative learning of eight groups took place on different dates. Before the collaborative learning activity commenced, the first author introduced the activity for 25 minutes. Each collaborative discussion lasted for 3 hours. The cross-cultural online collaborative learning task for each group was to discuss the aspects of instructional design (such as instructional goals, instructional strategies, instructional delivery process, and assessment methods) and design an instructional design plan for English teaching for grade 7 students. The group products were the instructional design plan of each group. The last phase was to conduct the post-test online for 30 minutes after each group submitted their group products.

Instruments

The instruments of this study include pre-test and post-test. In terms of pre-test with a perfect score of 100, there are 10 multiple-choice questions and 5 short answer questions. The post-test a perfect score of 100 also included another 10 multiple-choice questions and another 5 short answer questions. In addition, online discussion transcripts of each group were downloaded from WeChat to Microsoft Excel for coding of the idea generation and improvement analysis, and socially shared regulation analysis. The GSEQ 5.1 software was utilized for the socially shared regulation behaviors of cross-cultural groups. Whereas the Epistemic network analysis (ENA) web tool by Shaffer (2016) was adopted for the visualization of idea generation analysis. Furthermore, all the codes and scoring measures were considered by both authors based on students’ online discussion transcripts and group products. To determine the task performance, a 100 points rubric for the scoring group product (lesson plan) and post-test quiz were designed by the authors. Scores for each group was summed up used to determine groups performance which were categorized into high or low performance groups. The mean score was used to determine high and low performance groups.

Data Analysis Methods

To analyze the collaborative learning data, content analysis, and network analysis methods were deemed appropriate for this study. The coding scheme for idea generation and improvement was adopted from previous studies by Sun et al., (2022) and Zhu, Moreno, et al. (2019), but modified for the purpose of this study which is presented in Table 1. Moreover, the coding scheme for socially shared regulation behavior proposed by Zheng (2017) was adopted for coding each group’s online discussion transcript as illustrated in Table 2. Each group’s product (lesson plan) was scored to determine their performance based on the criteria of Table 3 (Zheng et al., 2019).

Table 1.

Coding Scheme for Idea Generation and Idea Improvement.

Super dimensions	Sub-dimensions	Descriptions	Examples
Cognition	New ideas (NI)	New ideas, comments, or concept from a new viewpoint that has not been mentioned initially.	“We need to enhance critical thinking, collaboration, communication skills, ICT skills along with subject knowledge”
Cognition	Refine/build on ideas (RBoI)	Refine or expand ideas to generate new ideas. Compare and contrast different ideas.	“Doing this, we will clearly have a great percentage of each background of people willing to learn. And the learning ability doesn’t just depend of the studies background but the age and the importance of this learning for the learner to be more captivated on lesson.”
Social communication	Paraphrase/appraisal/claim (PAC)	Facts, opinion or remarks, paraphrased information, or claim about task or topic of discussion.	“I’ll check the learning styles of the students by using VARK test.”
	Question	Seeking or asking for facts about terms, shared ideas, or concepts. Request for detailed explanation of shared ideas or concepts.	“How will you achieve this please?”
	Elaborating ideas (EI)	Giving explanations, examples, reasons, relationships, differences, or results about an idea.	“Commonly there are four types of learners, including visual, auditory, read/write and kinesthetic learners.” “I’ll make a profile of every learner by using VARK to calculate the average style of the learners in whole class.”
	Agreement	Acceptance of shared viewpoints or thoughts.	“Okay!” or “Go ahead!”
	Argument	Argue about how suitable a viewpoint is related to the task.	“This is general. Choose one appropriate for this particular lesson.” “How can you have holiday trips in the middle of the lesson?”
Metacognition	Regulation	Planning, monitoring and controlling group processes.	“Let’s go to the next task.” “How is going? Is there any question?”

Table 2.

Socially-Shared Regulation Coding Scheme.

Super dimensions	Sub-dimensions	Examples
Orienting goals (OG)	Task understanding.	“The main objective of today’s activity is to discuss the components of instructional design and plan a lesson.”
Orienting goals (OG)	Setting goals.	“We are about tackling task one so please read through it and try to write down some points to make the discussion an interesting one.”
Making plans (MP)	Making plans about how to reach the goals, including selecting strategies and setting timelines.	“Before we embark on drafting the lesson plan, I think let us first respond to the tasks embedded in the collaborative learning packet.”
Making plans (MP)	Negotiating the division of labor.	“I think we can begin with assigning responsibilities.”
Enacting strategies (ES)	Advancing and explaining solutions.	“Examples of learner characteristics are learning attitude and motivation as well as learning style.”
Monitoring and controlling (MC)	Monitoring or controlling the whole group’s progress.	“I think for this discussion to not go all part we should kindly follow the word document please.”
	Claiming (partial) understanding or comprehension failure.	“Can you elaborate on this?”
	Detecting errors or checking plausibility.	“I already did the corrections.”
Evaluating and reflecting (ER)	Evaluating current solutions.	“There are few things to correct on the diagram too.”
Evaluating and reflecting (ER)	Reflecting on the groups’ goals and progress.	“Let’s evaluate our lesson plan together.”
Adapting Metacognition (AM)	Making adaptations to goals, plans, strategies, or correcting product.	“Please we should adjust our strategies.”

Table 3.

The Criterion for Scoring the Lesson Plan.

Criterion	Descriptions	Weight
Instructional goals	Instructional goals are clearly defined and described.	10
Learners’ characteristics	The instructional plan is designed based on students’ prior knowledge, skills, learning style, background, and personalities.	10
Instructional content	The instructional content, learning tasks, and key points of instructions are clearly described and analyzed.	10
Instructional strategies	The instructional strategies and tools are carefully designed and applied appropriately based on instructional goals and content.	10
Instructional processes	The instructional process is complete and the instructional procedures are properly arranged. The timeslots are allocated appropriately.	10
	Learning activities are elaborately designed to engage students in learning.	20
	Exercises, assignments, and discussion are designed in line with instructional goals.	20
Evaluation	Appropriate assessment methods are designed and the rubrics are consistent with instructional objectives.	10

Then, lag sequential analysis (Gottman et al., 1977) was conducted with the GSEQ 5.1., to analyze the socially shared regulated behaviors of cross-cultural groups. Epistemic Network Analysis (ENA) was applied to examine idea generation and improvement with the ENA Web Tool version 1.7.0 (Marquart et al., 2018; Shaffer, 2016; Swiecki & Shaffer, 2020). Our ENA model included the following codes: Regulation, Paraphrase/Appraisal/Claim (PAC), Elaborating ideas (EI), Question, Argument, Agreement, Refine/Build on Ideas (RBoI), and New idea (NI). We defined conversations as all lines of data associated with a single value of comments.

Results

RQ1: How Do Students Generate and Improve Ideas in Cross-Cultural Collaborative Learning To?

ENA was used to visualize how students generates and improve ideas during cross-cultural collaborative learning. All the categories from the eight groups’ conversations were used to generate a mean network graph displayed in Figure 1. As displayed in Figure 1, regulation was strongly related to new idea, elaborating ideas, agreement, and paraphrase/appraisal/claims. Similarly, agreement was strongly connected to questions, new idea, and elaborating ideas. In terms of idea generation and idea improvement, Figure 1 shows new idea was highly linked to regulation, agreement, elaborating ideas and refining, or building on ideas. This revealed that regulation was a mechanism to monitor and control their online collaborative learning to inspire new ideas, elaborate of ideas, and agree to proceed or support an idea. Agreement was also found to be directly connected to questions and by tracing back to the conversations, it discovered that agreement was direct response to regulatory utterances. For instance, in group 1, L1 posed a question (“How will you achieve this please?”) to L3 who responded (“Commonly there are four types of learners Visual, Auditory, Read/Write, and Kinesthetic Learner”) with explanations and L1 reverted with a response of agreement (“Okay, …Agree”).

Figure 1.

The mean network of all group idea generation and improvement conversations.

To illustrate a connection of the model in Figure 1 back to the discussion transcripts, the excerpts below are presented in Tables 4 and 5. The extracts illustrate how one of the groups discussed to generate and improve on their knowledge. Episode 1 in Table 4 illustrates how students elicited for ideas through regulation. L19 employed questions as a strategy to elicit for ideas and also regulate the discussion. L20 and L21 shared their ideas in response to the questions asked. After sharing their ideas, L19 harmonized their ideas to generate new ideas.

Table 4.

Episode 1 on Generating New Ideas Through Regulation and Elaboration.

Learners	Dialogue (posts)	Codes	Annotations
L19	What do you think about instructional goals?	Question Regulation	L19 using question to regulate and seek for ideas
L20	A clear statement of observable behaviors that learners are to display as a result of the education they get is an instructional goal.	New idea	L20 shares a new idea with explanation in response to L19.
L21	Performance Approach which focuses on what learner need to be able to do at the end of the instruction.	New idea	L21 shares a new idea with explanation in response to L19.
L19	I think they are the things the instructor looks forward to achieve at the end of lesson.	Refine/build on idea	L19 shares a new idea which is build up on previous ideas from L20 and L21.
L19	Good ideas	Paraphrase/appraisal/claim	L19 giving positive appraisal.
L19	We share the same thought.	Paraphrase/appraisal/claim	L19 giving positive thought.
L19	How about the instructional strategies?	Question Regulate	L19 using question to regulate and seek for ideas.
L21	Techniques teachers use to help students become independent, strategic learners.	New idea	L21 shares new idea.
L19	ok PL21	Agreement	L19 agrees with L21
L20	Teachers utilize instructional tactics to help students become independent, strategic learners. When students choose acceptable techniques on their own and employ them effectively to complete tasks or fulfill goals, these methods become learning strategies.	New idea Elaborating ideas	L20 shares new idea with detailed explanation in response to PL19.
L19	Great	Paraphrase/appraisal/claim	L19 shares positive thought.

Table 5.

Episode 2 on Idea Improvement Through Collaboration.

Participant learner	Dialogue (posts)	Code	Annotations
L19	Is there any relationship between instructional goals and strategies?	Question Regulation	L19 asks another
L21	Yes …bcs if we use good or correct strategies for learners, we can get good results or goals.	New idea Elaborating ideas	L21 shares a new idea and explains
L19	I personally think they are related. First, the teaching strategies depend on instructional goals.	New idea Paraphrase/appraisal/claim Refine/build on ideas	L21 share new idea which is a claim and refined idea like L21’s.
L21	yes good	Agreement Paraphrase/appraisal/claim	L21 agrees with L19’s response with appraisal.
L20	We use the instructional strategies as techniques to achieve the goals and objectives.	Refine/build on ideas New idea	L20 refine and build on L21 and L19’s ideas to generate new idea.
L19	Please, could we cite any specific examples of instructional strategies?	Question Regulation	L19 suggested citing specific examples relating to shared ideas.

From Episode 2 illustrated in Table 5, L19 posed a question to seek for ideas on whether there is a difference between two topics discussed previously. The question encouraged L20 and L21 to brainstorm by comparing which resulted to them generating new ideas based on their previous contributions. L19 then refined the ideas shared by L20 and L21. An interesting aspect of the episode was that group members encouraged each other by making positive remarks such as “great ideas.” Accordingly, Figure 1 shows a connection between regulation, new ideas, paraphrase/appraisal/claim, and questions.

RQ2: Do High- and Low-Performance Groups Differ in Their Idea Generation and Improvement Process in Cross-Cultural Online Collaborative Learning?

Analysis of group performance

The top four scores for group product and post-test quiz were selected as the high-performance group (HPG) and least four scores were considered as the low performance group (LPG). The results show that there were four groups being HPG1, HPG2, HPG3, and HPG4 achieved high scores whiles four groups being LPG1, LPG2, LPG3, and LPG4 achieved scores less than the mean score of the high-performance groups as presented in Table 6. According to Table 6, high-performance groups surpassed the low-performance groups in group product (M_h = 92.50, M_l = 80.25) with a slight difference. The post-test quiz scores also indicated the HPGs outperformed the LPGs based on the mean score (M_h = 75, M_l = 58.13). Furthermore, an independent sample t-test confirmed the statistically significant difference in group product scores between the high-performance and low-performance groups (t = 5.68, p = .001). There was a significant difference in post-test scores (t = 3.63, p = .011).

Table 6.

The Results of High and Low-Performance Groups.

	High-performance groups					Low-performance groups
	HPG1	HPG2	HPG3	HPG4	M	LPG1	LPG2	LPG3	LPG4	M
Group product score	90	93	95	92	92.5	79	81	76	85	80.25
Average score of post-test	64.33	76.67	81	78	75	61.33	62.33	59	49.83	58.13

Differences in categories for group discussions

A comparison network of the different categories of idea generation and improvement in group conversations was conducted to illustrate whether there is a significant difference. The network of each discourse in ENA can be represented as a point in two-dimensional area over X-axis and Y-axis. Figure 2 shows a plotted point graph for HPGs (blue dots) located on the left and LPGs (red dots) located on the right and each dot represents a group. Accordingly, Figure 2 shows that the high performance groups (blue dots) were strongly connected among few social communication (e.g., elaborating ideas and refining/building on ideas) categories and high metacognition (regulation) categories, whiles the low performance groups were strongly connected among social communication categories (such as question, agreement, and new idea). The small blue square represents the mean location of the network for all HPG groups whereas small red square represents the mean location of LPGs. In addition, the boxes surrounding the means represent 95% of intervals for the means location. A two-sample t-test revealed a statistically significant difference between the high- and low-performance groups (t = 4.00; p = .01; Cohen’s d = 2.83). In comparison, Figure 2 corresponds to Figure 1 which shows elaborating ideas, refining or building on ideas, and regulation on the left side.

Figure 2.

Plotted graph of group conversations for LPGs (red) on right and HPGs on left (blue).

Comparison of idea generation and improvement processes

The network of conversations displaying the connections between categories in the high-performance groups and that of low performance groups is presented in Figure 3 showing the differences among the groups. The blue lines depicted a strong relationship between two categories in the HPGs whiles the red lines depicted strong relationship between two categories in the LPGs. For the HPGs, there were strong connections between six categories (new idea, regulation, elaborating ideas, question, refine or build on ideas, and paraphrase/appraisal/claims). Similarly, six categories were strongly connected (new idea, regulation, agreement, question, paraphrase/appraisal/claims, and elaborating ideas) within the LPG conversations. The idea generation and improvement strategies among LPGs mostly involved social and metacognitive processes whiles the HPGs mostly involved cognitive, social, and metacognitive processes. First, refining/building on ideas had strong connection with three categories (elaborating ideas, new idea, and paraphrase/appraisal/claim) as shown in Figure 3. Comparably, this pattern was more frequent among the HPGs than the LPGs, indicating that the HPGs had more utterances of cognitive processes than the LPGs. Second, regulation was strongly connected to three categories (paraphrase/appraisal/claims, questions, and refining/building on ideas) within the HPG groups and two other categories (agreement and new idea) in the LPGs. It should be noted that regulation utterances were more among the LPG groups. A cross examination of the utterances indicated that, among LPGs, utterances of regulatory strategies were used to elicit new ideas and upon agreement, they proceeded to the next topic of discussion. In contrast, the HPGs used regulatory strategies to seek for information, give opinions, or judgments about shared ideas and through this process refine and build on shared ideas. Last, question was found to be more connected to agree in the LPGs utterances, whiles it was more connected to regulation among the HPGs utterances. According to the comparison network in Figure 3 and group conversation data, new ideas were mostly elaborated and refined, or built on by the HPGs than the LPGs. This finding indicated that the high-performance groups were able to build on their ideas through elaboration and giving appraisal and opinions about individual ideas. Whereas, the low performing groups mostly agreed on shared ideas without questions. Comparing to the discussion transcript, it was found that agreement was mostly related to questions such as “can we move to the next task?” and “what do you think?” especially among the LPGs.

Figure 3.

Comparison network of idea generation and improvement among high (blue) and low-performance (red) groups.

RQ3: Do High- and Low-Performance Groups Differ in Their Socially Shared Regulation Behaviors in Cross-Cultural Online Collaborative Learning?

To determine the sequential differences in socially shared regulation behaviors for low and high performing groups, the adjusted residuals for high-performance and low-performance groups were calculated separately with GSEQ 5.1. First, the frequency distribution of all the identified behaviors at the first level dimension were calculated and presented in Table 7 to give the general behavior patterns of different performance groups. The LSA approach suggests a z-score above 1.96 indicates a sequence has obtained the level of significance (Gottman et al., 1977; Pohl et al., 2016). As presented in Table 7, higher proportion of socially shared behaviors occurred in the low-performance group compared to high-performance groups. However, a breakdown of the SSR behaviors show that the high-performance groups exhibited more regulatory strategies with higher proportions of orienting goals (OG), monitoring and controlling (MC), and adapting metacognition (AM) behaviors. In addition, within the high-performance groups the most occurring behavior was MC, followed by AM, then evaluation and reflection (ER), OG, making plan (MP), and last enacting strategies (ES). Whereas the low-performance groups exhibited higher proportions of enacting strategies (ES), evaluation and reflection (ER), and making plans (MP). Within the low-performance groups, ES was the most occurring behavior, followed by ER, MC, then MP, OG, and last AM. Furthermore, the adjusted residuals of the behavioral sequences for the high-performance groups were presented in Tables 8 and 9 displays that of the low-performance group.

Table 7.

Distribution of SSR Behaviors.

Categories	High PG	Percentage (%)	Low PG	Percentage (%)	Interaction posts
Categories	Frequency	Percentage (%)	Frequency	Percentage (%)	Frequency
OG	15	53.57	13	46.43	28
MP	12	35.29	22	64.71	34
ES	10	2.06	475	97.94	485
MC	360	64.17	201	35.83	561
ER	170	41.87	236	58.13	406
AM	226	94.96	12	5.04	238
Total	793	45.26	959	54.74	1,752

Table 8.

Adjusted Residuals for High Performance Group.

Given	OG	MP	ES	MC	ER	AM
OG	16.60*	4.23*	−2.53	−0.77	−1.91	−0.48
MP	2.21*	11.05*	−2.91	0.66	−0.62	−0.39
ES	−3.18	−2.90	9.18*	−4.90	−3.39	−2.60
MC	0.30	−0.89	−1.11	3.28*	−2.27	2.43*
ER	−2.20	0.11	−6.93	2.41*	6.09*	−0.27
AM	−0.43	−0.39	−2.61	0.30	1.64	4.33*

p < .05.

Table 9.

Adjusted Residuals for Low Performance Group.

Given	OG	MP	ES	MC	ER	AM
OG	11.70*	3.02*	−3.21	0.70	−0.89	−0.42
MP	1.41	7.93*	−3.86	0.20	1.31	−0.53
ES	−3.45	−4.26	11.32*	−5.04	−5.58	−1.70
MC	1.05	1.26	−3.40	3.02*	0.66	−1.09
ER	−0.64	0.30	−7.40	2.65*	5.46*	2.74*
AM	0.39	−0.53	−1.14	0.34	0.72	2.22*

p < .05.

The difference in SSR behaviors between the high-performance and low-performance groups were examined through the Chi-square test (α = .05). The results indicated that there were significant differences among the SSR categories between the high performance groups and low performance groups. The proportions of orienting goal behaviors by HPGs and LPGs indicated that they differed significantly (χ² = 28.0, p < .001). HPGs and LPGs significantly differ in ES behaviors (χ² = 187.8, p < .001). There were significant differences among the MC behaviors of between HPGs and LPGs (χ² = 32.6, p < .001). More so, the ER behavior of LPGs significantly differed from HPGs (χ² = 36.9, p < .001). However, the high and low performance groups did not differ in AM behaviors (χ² = 1.4, p = .223). There was a difference among MC behaviors of HPGs and LPGs (χ² = 11.4, p = .001).

Figures 4 and 5 display the visualized significant behavioral transitions of high and low-performance groups. The nodes denote the behavioral categories, the arrows denote the transition direction, the numbers are the Z-value for the sequence, and the thickness depicts the level of significance. The socially shared regulated behavioral paths revealed in Figure 4 indicated 10 statistically significant behavioral paths occurred in among the high-performance groups which were OG → OG, OG → MP, ES → ES, MC → MC, MC → AM, ER → MC, ER → ER, AM → AM, MP → OG, MP → MP. While in Figure 5, nine significant SSR behavioral paths were identified among the low performance group OG → OG, OG → MP, MP → MP, ES → ES, MC → MC, ER → MC, ER → ER, ER → AM, and AM → AM.

Figure 4.

The behavioral transitions of the high-performance group.

Figure 5.

The behavioral transitions of low-performance groups.

Furthermore, it was found that the high-performance groups demonstrated behavioral paths different from the low-achievement groups. The initial behavioral transition in high-performance groups was OG → OG followed by OG → MP while OG → MP was the first path followed by OG → OG in the low-performance groups. This indicated that the high-performance groups discussed and understood the tasks and the group goals before proceeding to make plans to achieve these goals such as assigning roles. Whereas the low-performance groups begun with making plans before discussing the group goals. Next, the high-performance groups could straight away proceed to the main tasks and explaining solutions after making plans ES → ES while the low-performance returned to making plans again MP → MP. A major difference in path between the two groups found was that, the fifth path (MC → AM) in the high-performance groups did not occur in the low-performance groups. This instance suggested that after explaining solutions the high-performance groups could monitor and control their group’s process, claim partial understanding, and detect errors before adapting metacognition while the low performance group did not. Lastly, after applying metacognition, the paths revealed that the high-performance groups evaluated and reflected on their solution to the tasks, applied metacognition and again check their progress in terms of plan and goals (MP → OG, MP → MP) toward the end of the discussion. On the other hand, after the low-performance groups evaluated and reflected, they proceeded to apply metacognition (ER → AM, AM → AM) toward the end of the discussion.

Discussion

This study investigated how collaborative groups generated and improved ideas in a cross-cultural setting. The findings in this study indicated that cross-cultural collaborative learning promoted idea generation and improvement. Furthermore, cross-cultural groups exhibited different patterns of regulation behaviors and conversations patterns to generate ideas and improve their ideas.

Concerning the first research question on how cross-cultural groups generated and improved ideas, different cognitive, metacognitive, and social processes were used by groups during collaborative learning. It was discovered that the discussions pivoted on metacognitive processes which engaged group members to share ideas. To regulate the discussion, group members who assumed the role of group leader often used questions to seek for ideas and also to prompt idea generators to elaborate their ideas. Idea improvement occurred when ideas generated by individuals within a group was further examined or synthesized by group members collaboratively. This created the opportunity for team members to seek consensus or build on their knowledge (W. Chen & Tan, 2021). Tolerance and mutual recognition of ideas shared by giving positive evaluations were behaviors that was discovered to also have a direct connection with idea elaboration and improvement. This agrees with previous studies that although diversity could hinder interaction, perspective taking without negative judgements, but an atmosphere of mutual understanding and tolerance fosters idea generation (Gonçalves et al., 2020; Hawlina et al., 2017; Tlili et al., 2021). This research finding sheds more light and contributes to scholarly studies based on the connectivism theory revealing that sequential discussions with the most connects online collaborative platform has positive impact on collaborative activity (Yousef et al., 2020). Expressing agreement to ideas was also found to relate new ideas, elaborating ideas, and building on ideas. This revealed that during a process of knowledge building discourse, cross-cultural groups in this study with or without consensus seeking stated their agreement with ideas shared, which sometimes lead to building or modifying the shared idea. The reason for the weak connection between agreement and idea improvement might be that most of the agreement expressions shown were direct response to regulation. This finding builds on van Aalst (2012) on the view that the agreement on ideas are not modified as it revealed that agreement without prior perspective taking and appraisal on ideas hinders idea improvement.

Concerning the difference in idea generation and improvement transitions between HPGs and LPGs, the findings showed that there was statistical difference in idea generation and improvement processes. First, utterances of agreement on shared ideas without attempts to refine or improve generated ideas was frequent among LPGs. A reason could be social factors such as friendship (Cavagnetto et al., 2022) or avoiding to offend the idea generator which might have prevented new insights. Next, utterances of paraphrasing, appraisal and opinions about shared ideas, elaborating ideas, and building on ideas were more frequent in the HPGs. This could be the reason for high idea elaboration and improvement among HPGs. This supports the claim that giving attention to other team members viewpoints improve creativity and performance (Hoever et al., 2012). Perspective taking behaviors such as frequent questioning without mutual understanding yielded low performance. The findings indicated that regulation, questioning, elaborating ideas, refining or building on ideas with evaluation, and perspective-taking corresponded to idea improvement. Lastly, the analysis indicated that high frequency of idea sharing and regulation did not correspond to idea improvement and knowledge building in the context of cross-cultural collaborative learning context. On the other hand, higher engagement of elaborating ideas, refining, and building on ideas with perspective taking promoted learning. This indicated that both cognitive strategies (new idea and refine/build on idea), metacognitive strategies (such as monitoring and controlling, and adapting metacognition) strategies and high socio-communication strategies such as elaborating ideas, paraphrase/appraisal/claim, and question were critical for learning performance.

Concerning the third research question on whether there was a difference in SSR behaviors between high and low-performance groups in cross-cultural collaborative learning. Establishing goals and planning were found as initial stage for all the groups and however, the end stage according to the findings differed. All groups elaborated more on goal setting at the beginning of their discussion. This result was consistent with Hoek et al. (2018) that clear elaboration of group goals helps to disperse uncertainty about team goals and increase shared understanding which concerns the expected outcome. The high-performance groups from the middle to the end exhibited monitoring and evaluation of goals and plan, which is confirmed by Quackenbush and Bol (2020). Whereas the low-performance groups mostly ended with adapting metacognition. HPGs kept an eye on their goals and plans (Hogenkamp et al., 2021) at the end while low-performance groups did not. This could be as a result of unclear goals and uncertainty among the (Zhu et al., 2019). A reason could be that the LPGs paid more attention to completing the task and regulation of the collaborative process. However, in collaboration, both task performance and regulation of collaborative process are significant for achieving goals (Saab, 2012). Furthermore, a unique SSR behavior pattern that was found within the HPGs was monitoring and controlling to adapting metacognition in the middle stage of collaboration process. This denoted that halfway into their discussions, HPGs established behaviors of scrutinizing and re-strategized their plans toward achieving group goals. This could be a significant SSR behavioral transition for cross-cultural collaborative groups to improve and apply knowledge gained. These findings serve as up-to-date evidence that supports literature on the fact that collaborative groups’ online interaction patterns affect their performance (Huang et al., 2019).

Conclusions

This study discovered socially shared regulation behavioral patterns and epistemic association of utterances that lead to idea improvement through analysis of cross-cultural groups discourse with LSA and ENA. Regulation of discourse with questions encourages idea generation and idea elaboration leads to idea improvement by refining and building other members ideas. Most significantly, the findings discovered that sharing more ideas without perspective taking on shared ideas to refine and improve them could affect learning performance. Monitoring and controlling online collaborative learning and adapting metacognition behaviors in the middle of the discussion align with current research and this study extends socially shared regulation to idea improvement and knowledge-building. This study revealed that idea generation and improvement was related to appraisal. Moreover, frequent contribution of ideas without metacognitive strategies could affect learning performance.

In addition, this study has implications for implementing cross-cultural collaborative learning idea generations. First, cross-cultural groups could thrive amidst challenges that come with working with culturally diverse individuals through SSR, mutual recognition of ideas, idea elaboration, and synthesizing ideas. Second, it was important to identify the idea generation processes and SSR transitions such as transitioning from monitoring and controlling to adapting metacognition in the middle of a discussion could promote knowledge-building. To achieve this, it is significant to visualize and statistically analyze group conversations. Third, groups should be trained on how to respond to ideas, make, and accept criticism during group discussions. Last, advanced technological solution to intervene during idea generation processes would be rewarding. To achieve this, learning analytics approach would be beneficial for collaborative learning support strategies such as feedforwarding and automatic feedback.

This study had some limitations with sample size. As a result, the findings cannot be generalized. Therefore, it is recommended that future studies should examine more groups with intervention on how to improve perspective-taking and mutual understanding among cross-cultural collaborative learning. Moreover, the extent to which agreement and appraisal could influence idea generation and improvement in a cross-cultural collaborative learning setting should be investigated further. Based on the findings of this study, it is recommended that future studies engage more participants to compare differences in idea generation patterns and socially shared regulation behaviors between the high performance and low performance groups. Moreover, investigating how large cross-cultural groups generate and improve ideas in an online collaborative learning setting will shed more light in the area of study. Most importantly, future studies should delve into the extent to which SSR behavioral transitions are likely to affect learning. Last, this study lasted for a short period of time and future studies will engage students for a more extended period of time and investigate whether cultural orientations matter with regards to idea generation and improvement and socially shared regulation behaviors.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study is funded by the International Joint Research Project of Huiyan International College, Faclty of Education, Beijing Normal University (ICER202101).

Ethical Approval

All participants in this study agreed to participate in this study by giving verbal and written consent. Furthermore, all ethical guidelines were kept.

ORCID iD

Lanqin Zheng

References

Acar

O. A.

Tarakci

Van Knippenberg

(2019). Creativity and innovation under constraints: A cross-disciplinary integrative review. Journal of Management, 45(1), 96–121.

Apostolidou

(2022). Digitally situated knowledge: Connectivism, anthropology and epistemological pluralism. International Journal of Educational Research, 115, 102047.

Breen

(2015). Assessing online collaborative discourse. https://digitalcommons.linfield.edu/nursfac_pubs/4

Cavagnetto

A. R.

Premo

Coleman

Juergens

(2022). Accuracy and idea consideration: A study of small-group interaction in biology. CBE—Life Sciences Education, 21(1), 1–11.

Chen

S.-J.

Hsu

Caropreso

(2006). Cross-cultural collaborative online learning: When the west meets the east. International Journal of Technology in Teaching and Learning, 2(1), 17–35.

Chen

Tan

J. S. H.

(2021). The spiral model of collaborative knowledge improvement: An exploratory study of a networked collaborative classroom. International Journal of Computer-Supported Collaborative Learning, 16(1), 7–35.

Child

Shaw

(2016). Collaboration in the 21st century: Implications for assessment. Research Matters, 22, 17–22

Collis

(2004). Applying activity theory to computer-supported collaborative learning and work-based activities in corporate settings. Educational Technology Research and Development, 52(4), 38–52.

Dai

(2019). Situating videoconferencing in a connected class for intercultural knowledge development: A comparative reflection approach. The Internet and Higher Education, 41, 1–10. https://doi.org/10.1016/j.iheduc.2018.11.001

10.

Downes

(2019). Recent work in connectivism. European Journal of Open, Distance and E-Learning, 22(2), 112–131.

11.

Ellamil

Fox

K. C. R.

Dixon

M. L.

Pritchard

Todd

R. M.

Thompson

Christoff

(2020). Dynamics of neural recruitment surrounding the spontaneous arising of thoughts in experienced mindfulness practitioners. Neuroimage, 136, 186–196.

12.

Escandon-Quintanilla

M.-L.

Jimenez-Narvaez

L.-M.

Gardoni

(2015). Strategies to employ social networks in early design phases (idea generation). Proceedings of the International Conference on Engineering Design, ICED, 8(DS 80-08), 329–338.

13.

Glăveanu

V. P.

(2018). Creativity in perspective: A sociocultural and critical account creativity in perspective: A sociocultural and critical account. Journal of Constructivist Psychology, 31(2), 118–129.

14.

Goldie

J. G. S.

(2016). Connectivism: A knowledge learning theory for the digital age? Medical Teacher, 38(10), 1064–1069.

15.

Gonçalves

Sousa

Arasaratnam-Smith

L. A.

Rodrigues

Carvalheiro

(2020). Intercultural communication competence scale: Invariance and construct validation in Portugal. Journal of Intercultural Communication Research, 49(3), 242–262.

16.

Gottman

Markman

Notarius

(1977). The topography of marital conflict: A sequential analysis of verbal and nonverbal behavior. Journal of Marriage and the Family, 39(3), 461–477.

17.

Gyasi

J. F.

Zheng

Long

(2021). Reflecting on the past to shape the future: A systematic review on cross-cultural collaborative learning from 2011 to 2020. Sustainability (Switzerland), 13(24), 13890. https://doi.org/10.3390/su132413890

18.

Hadwin

Järvelä

Miller

(2019). Self-regulation, co-regulation, and shared regulation in collaborative learning environments. In Schunk

D. H.

Greene

J. A.

(Eds.), Handbook of self-regulation of learning and performance (pp. 83–106). Routledge.

19.

Hawlina

Gillespie

Zittoun

(2017). Difficult differences: A socio-cultural analysis of how diversity can enable and inhibit creativity. Journal of Creative Behavior, 53, 133–144.

20.

Herlache

Renkema

Cummins

Scovotti

(2018). A cross-cultural negotiation role-play for sales classes. Journal for Advancement of Marketing Education, 26, 1–10.

21.

Hoek

van der Groeneveld

Kuipers

(2018). Goal setting in teams: Goal clarity and team performance in the public sector. Review of Public Personnel Administration, 38(4), 472–493. https://doi.org/10.1177/0734371X16682815

22.

Hoever

I. J.

Van Knippenberg

Van Ginkel

W. P.

Barkema

H. G.

(2012). Fostering team creativity: Perspective taking as key to unlocking diversity’ s potential. Journal of Applied Psychology, 97(5), 982–996.

23.

Hofstede

(1986). Cultural differences in teaching and learning. International Journal of Intercultural Relations, 10, 301–320.

24.

Hogenkamp

Van Dijk

A. M.

Eysink

T. H. S.

(2021). Analyzing socially shared regulation of learning during cooperative learning and the role of equal contribution: A grounded theory approach. Education Sciences, 11(512), 1–26.

25.

Huang

C. Q.

Han

Z. M.

M. X.

Jong

M. S.

Tsai

C. C.

(2019). Investigating students’ interaction patterns and dynamic learning sentiments in online discussions. Computers and Education, 140, 103589.

26.

Iiskala

Volet

Lehtinen

Vauras

(2015). Socially shared metacognitive regulation in asynchronous cscl in science: functions, evolution and participation. Frontline Learning Research, 3(1), 78–111.

27.

Isohätälä

Järvenoja

Järvelä

(2017). Socially shared regulation of learning and participation in social interaction in collaborative learning. International Journal of Educational Research, 81, 11–24.

28.

Janssen

J. J. H. M.

Erkens

Jaspers

J. G. M.

Broeken

(2005). Effects of visualizing participation in computer-supported collaborative learning [Paper presentation]. Proceedings of the 11th Biennial conference of the European Association for Research in Learning and Instruction. Cyprus.

29.

Järvelä

Kirschner

P. A.

Panadero

Malmberg

Phielix

Jaspers

Koivuniemi

Järvenoja

(2015). Enhancing socially shared regulation in collaborative learning groups: Designing for CSCL regulation tools. Educational Technology Research and Development, 63, 125–142.

30.

Jeong

Hmelo-Silver

C. E.

(2016). Seven affordances of computer-supported collaborative learning: How to support collaborative learning? How can technologies help?. Educational Psychologist, 51(2), 247–265.

31.

Jeannin

(2013). Students’ perception of diversity in an international classroom. Higher Learning Research Communications, 3(4), 6–23. https://doi.org/10.18870/hlrc.v3i4.165

32.

Jeong

Hmelo-Silver

C. E.

(2019). Ten years of computer-supported collaborative learning: A meta-analysis of CSCL in STEM education during 2005–2014. Educational Research Review, 28, 100284.

33.

Jia

(2019). The role of metacognitive components in creative thinking. Frontiers in Psychology, 10, 2404.

34.

Kawakubo

A. J.

Oshima

(2021). Idea improvement processes leading to high learning outcomes and the development of regulation in collaboration [Conference session]. Proceedings of the 14th International Conference on Computer-Supported Collaborative Learning-CSCL 2021 (pp. 169–172).

35.

Kay Siu

(2012). Collaborative learning: A means to creative thinking in design. International Journal of Education And Information Technologies, 6(1), 33–43.

36.

Kelley

T. R.

(2008). Cognitive processes of students participating in engineering-focused design instruction. Journal of Technology Education, 19(2), 50–64.

37.

Khosa

D. K.

Volet

S. E.

(2014). Productive group engagement in cognitive activity and metacognitive regulation during collaborative learning: Can it explain differences in students’ conceptual understanding? Metacognition and Learning, 9(3), 287–307.

38.

King

(2002). Structuring peer interaction to promote high-level cognitive processing. Theory into Practice, 41(1), 33–39.

39.

Kirchner

Razmerita

(2015). Collaborative learning in the cloud: A cross-cultural perspective of collaboration. In Yesilada

Farzan

Houben

G.-J.

(Eds.), Proceedings of the 26th acm conference on hypertext and social media (pp. 333–336). Association for Computing Machinery. https://doi.org/10.1145/2700171.2804452

40.

Kuhn

(2015). Thinking together and alone. Educational Researcher, 44(1), 46–53.

41.

Kumi-Yeboah

(2018). Designing a cross-cultural collaborative online learning framework for online instructors. Online Learning Journal, 22(4), 181–201.

42.

Laal

Khattami

(2012). 21st century learning; learning in collaboration. Procedia - Social and Behavioral Sciences, 47, 1696–1701.

43.

Lam

(2019). What students do when encountering failure in collaborative tasks. NPL Science of Learning, 4(1), 1–11.

44.

Larson

L. C.

Miller

T. N.

(2011). 21st century skills: Prepare students for the future. Kappa Delta Pi Record, 47(3), 121–123.

45.

Lau

Vuthaluru

Mui

Kerrigan

Kwong

Law

Wong

E. Y. W.

Gibson

(2021). How online teams with diverse backgrounds worked to excel: Findings from an international etournament. Frontiers in Education, 6, 624438.

46.

Lee

Markey

(2014). A study of learners’ perceptions of online intercultural exchange through Web 2.0 technologies. ReCALL, 26(3), 281–297.

47.

Lin

Gao

(2020). Students’ sense of community and perspectives of taking synchronous and asynchronous online courses. Asian Journal of Distance Education, 15(1), 169–179.

48.

Liu

C. Y.

Chen

H. L.

(2020). Effects of peer learning on learning performance, motivation, and attitude. International Journal of Education Economics and Development, 11(4), 420–443. https://doi.org/10.1504/ijeed.2020.10030596

49.

Liu

(2007). Designing quality online education to promote cross-cultural understanding. In Edmundson

(Ed.), Globalized E-learning cultural challenges (pp. 35–59). IGI Global. https://doi.org/10.4018/978-1-59904-301-2.ch003

50.

MacLeod

Yang

H. H.

Xiang

(2017). Understanding College Students’ Intrinsic Motivation and Social Interdependence in Intercultural Computer-Supported Collaborative Learning Between USA and China. Asia-Pacific Education Researcher, 26(3–4), 205–217.

51.

Malmberg

Järvelä

Järvenoja

Panadero

(2015). Promoting socially shared regulation of learning in CSCL: Progress of socially shared regulation among high- and low-performing groups. Computers in Human Behavior, 52, 562–572.

52.

Marambe

K. N.

Vermunt

J. D.

Boshuizen

H. P. A.

(2012). A cross-cultural comparison of student learning patterns in higher education. Higher Education, 64(3), 299–316.

53.

Marquart

C. L.

Hinojosa

Swiecki

Eagan

Shaffer

D. W.

(2018). Epistemic network analysis (version 1.7.0) [Software]. http://app.epistemicnetwork.org

54.

Miller

Hadwin

(2015). Scripting and awareness tools for regulating collaborative learning: Changing the landscape of support in CSCL. Computers in Human Behavior, 52, 573–588.

55.

Nelson

Narens

(1990). Metamemory: a theoretical framework and new findings. The Psychology of Learning and Motivation, 26, 125–173.

56.

P. M. L.

Chan

J. K. Y.

Kong

(2022). Student learning performance in online collaborative learning. Education and Information Technologies, 27, 8129–8145.

57.

Nijstad

B. A.

Stroebe

(2006). How the group affects the mind: A cognitive model of idea generation in groups. Personality and Social Psychology Review, 10(3), 186–213.

58.

O’Donnell

A. M.

King

(1999). Cognitive perspectives on peer learning. Routledge.

59.

Paulus

P. B.

Brown

V. R.

(2007). Toward more creative and innovative group idea generation: A cognitive-social-motivational perspective of brainstorming. Social and Personality Psychology Compass, 1, 248–265.

60.

Paulus

P. B.

Levine

D. S.

Brown

Minai

A. A.

Doboli

(2010). Modeling ideational creativity in groups: Connecting cognitive, neural, and computational approaches. Small Group Research, 41(6), 688–724.

61.

Pohl

Wallner

Kriglstein

(2016). Using lag-sequential analysis for understanding interaction sequences in visualizations. International Journal of Human Computer Studies, 96, 54–66.

62.

Popov

Biemans

H. J. A.

Fortuin

K. P. J.

van Vliet

A. J. H.

Erkens

Mulder

Jaspers

(2019). Effects of an interculturally enriched collaboration script on student attitudes, behavior, and learning performance in a CSCL environment. Learning, Culture and Social Interaction, 21, 100–123.

63.

Popov

Biemans

H. J. A.

Kuznetsov

A. N.

Mulder

(2014). Use of an interculturally enriched collaboration script in computer-supported collaborative learning in higher education. Technology, Pedagogy and Education, 23(3), 349–374. https://doi.org/10.1080/1475939X.2014.945474

64.

Puteh

Bakar

S. A.

Mohamad

A. R.

Ramly

L. Z.

Sazalli

N. A. H.

Zainon

Sjarif

N. N. A.

Hamid

N. S. A.

Law-Chong Lai

Bannister

(2020). Online collaborative learning via astronomy online lab: A cross-cultural communicative experience for malaysian and uk students. Universal Journal of Educational Research, 8(11), 45–54.

65.

Quackenbush

Bol

(2020). Teacher support of co- and socially-shared regulation of learning in middle school mathematics classrooms. Frontiers in Education, 5, 580543.

66.

Saab

(2012). Team regulation, regulation of social activities or co-regulation: Different labels for effective regulation of learning in CSCL. Metacognition and Learning, 7(1), 1–6.

67.

Saadatmand

Buus

HomasRyberg

Georgsen

(2012). Emerging technologies and new learning ecologies: Learner perceptions of learning in open networked environments. In Hodgson

Jones

M. D.

Laat

Ryberg

Sloep

(Eds.), Proceedings of the 8th international conference on networked learning (pp. 266–275). Lancaster University.

68.

Sampson

Clark

D. B.

(2011). A comparison of the collaborative scientific argumentation practices of two high and two low performing groups. Research in Science Education, 41(1), 63–97. https://doi.org/10.1007/s11165-009-9146-9

69.

Scardamalia

(2002). Collective cognitive responsibility for the advancement of knowledge. In Smith

(Ed.), Liberal education in a knowledge society (pp. 76–98). Open Court.

70.

Shaffer

D. W.

(2016). A tutorial on epistemic network analysis. Journal of Learning Analytics, 3(3), 9–45.

71.

Sharma

Papavlasopoulou

Giannakos

(2021). Faces don’t lie: Analysis of children’s facial expressions during collaborative coding. FabLearn Europe/MakeEd 2021: FabLearn Europe/MakeEd 2021 – An International Conference on Computing, Design and Making in Education, St. Gallen

72.

Shi

Frederiksen

C. H.

Muis

K. R.

(2013). A cross-cultural study of self-regulated learning in a computer-supported collaborative learning environment. Learning and Instruction, 23(1), 52–59.

73.

Siemens

(2004). Connectivism: A learning theory for the digital age. https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=f87c61b964e32786e06c969fd24f5a7d9426f3b4.

74.

Spires

H. A.

Kerkhoff

S. N.

Fortune

(2019). Educational cosmopolitanism and collaborative inquiry with Chinese and US teachers. Teaching Education, 30(4), 437–454.

75.

Spires

H. A.

MedlockPaul

Himes

Yuan

(2018). Cross-cultural collaborative inquiry: A collective case study with students from China and the US. International Journal of Educational Research, 91, 28–40. https://doi.org/10.1016/j.ijer.2018.07.002

76.

Y. C.

(2008). Promoting cross-cultural awareness and understanding: Incorporating ethnographic interviews in college EFL classes in Taiwan. Educational Studies, 34(4), 377–398.

77.

Sun

Wang

Wegerif

Peng

(2022). How do students generate ideas together in scientific creativity tasks through computer-based mind mapping? Computers and Education, 176, 104359.

78.

Swiecki

Shaffer

D. W.

(2020). ISENS: An integrated approach to combining epistemic and social network analyses [Conference session]. ACM International Conference Proceeding Series (pp. 305–313).

79.

Syzenko

Diachkova

(2020). Building cross-cultural competence in a Foreign language through technology-enhanced project-based Learning. Revista Amazonia Investiga, 9(27), 411–418.

80.

Tham

Duin

A. H.

Veeramoothoo

S. (Chakrika)

Fuglsby

B. J.

(2021). Connectivism for writing pedagogy: Strategic networked approaches to promote international collaborations and intercultural learning. Computers and Composition, 60, 102643.

81.

Tlili

Wang

Gao

Shi

Zhiying

Looi

C. K.

Huang

(2021). Impact of cultural diversity on students’ learning behavioral patterns in open and online courses: A lag sequential analysis approach. Interactive Learning Environments. Advance online publication. https://doi.org/10.1080/10494820.2021.1946565

82.

Ucan

Webb

(2015). Social regulation of learning during collaborative inquiry learning in science: How does it emerge and what are its functions? International Journal of Science Education, 37(15), 2503–2532.

83.

van Aalst

(2012). Knowledge building: Rationale, examples, design, and assessment. Computers in New Zealand Schools: Learning, Teaching, Technology, 24(3), 220–238.

84.

von Davier

A. A.

Halpin

P. F.

(2013). Collaborative problem solving and the assessment of cognitive skills: Psychometric considerations. ETS Research Report Series, 2013(2), i–36. https://doi.org/10.1002/j.2333-8504.2013.tb02348.x

85.

Wang

(2019). Towards a taxonomy of idea generation techniques. Foundations of Management, 11(1), 65–80.

86.

Wang

S.-L.

Kuo

M.-X.

(2016). The role of cognitive styles in collaborative annotations and collaborative performance in CSCL. Technology-Enhanced Collaborative Learning (TECL 2016) in Conjunction With CRIWG/CollabTech 2016, Kanazawa, Japan, pp. 22–24.

87.

White

D. R.

(2002). Cross-cultural research: An introduction for students (pp. 1–14). https://escholarship.org/content/qt0h07786w/qt0h07786w.pdf

88.

Yamazaki

Kayes

D. C.

(2004). An experiential approach to cross-cultural learning: A review and integration of competencies for successful expatriate adaptation. Academy of Management Learning & Education, 3(4), 362–379. https://doi.org/10.5465/amle.2004.15112543

89.

Yang

Chen

Huang

(2014). Strategies for smooth and effective cross-cultural online collaborative learning, Educational Technology & Society, 17(3), 208–221.

90.

Yousef

A. M. F.

Salah

R. A.

Makram

E. M.

(2020). Investigating different educational blog characteristics to support collaborative learning based on connectivism learning theory. CSEDU 2020 - Proceedings of the 12th International Conference on Computer Supported Education, 2, 118–129.

91.

Zhang

Gao

Sun

Cai

Tang

Liu

(2022). Understanding student teachers’ collaborative problem solving: Insights from an epistemic network analysis (ENA). Computers and Education, 183, 104485.

92.

Zheng

(2017). Analysis of socially shared regulation in CSCL. In Shi

(Eds.), Knowledge building and regulation in computer-supported collaborative learning, perspectives on rethinking and reforming education (pp. 65–81). Springer Singapore. https://doi.org/10.1007/978-981-10-1972-2_5

93.

Zheng

Huang

(2017). The effect of socially shared regulation approach on learning performance in computer-supported collaborative learning. Educational Technology & Society, 20(4), 35–46.

94.

Zheng

Zhang

Sun

(2019). The effects of group metacognitive scaffolding on group metacognitive behaviors, group performance, and cognitive load in computer-supported collaborative learning. The Internet and Higher Education, 42, 13–24.

95.

Zheng

Niu

Zhong

Gyasi

J. F.

(2021). Knowledge-building and metacognition matter: Detecting differences between high- and low-performance groups in computer-supported collaborative learning. Innovations in Education and Teaching International, 1–11. https://doi.org/10.1080/14703297.2021.1988678

96.

Zhu

Moreno

Valencia

Scardamalia

(2019). Idea improvement patterns in knowledge building: Undergraduate and graduate levels [Conference session]. Proceedings of AERA Annual Meeting, Toronto.

97.

Zhu

Xing

Costa

Scardamalia

Pei

(2019). Exploring emotional and cognitive dynamics of Knowledge Building in grades 1 and 2. User Modeling and User-Adapted Interaction, 29(4), 789–820.