- Email: [email protected]
A social network analysis on elementary student engagement in the networked creation community
Accepted Manuscript A social network analysis on elementary student engagement in the networked creation community Chen-Chung Liu, Yu-Chi Chen, Shu-Ju Daian Tai PII:
S0360-1315(17)30188-4
DOI:
10.1016/j.compedu.2017.08.002
Reference:
CAE 3223
To appear in:
Computers & Education
Received Date: 4 March 2017 Revised Date:
5 August 2017
Accepted Date: 6 August 2017
Please cite this article as: Liu C.-C., Chen Y.-C. & Daian Tai S.-J., A social network analysis on elementary student engagement in the networked creation community, Computers & Education (2017), doi: 10.1016/j.compedu.2017.08.002. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
A social network analysis on elementary student engagement in the networked creation community Chen-Chung Liu1, Yu-Chi Chen1, Shu-Ju Daian Tai2 1
RI PT
Graduate Institute of Network Learning Technology, National Central University, Taiwan 2 Beijing University of Chemical Technology, School of International Education, Beijing, CN
M AN U
SC
Abstract The recent development of social networking tools has enabled students to collaborate to learn in the social constructivists’ approach. Yet, the learning process in the networked creation community involves complex dynamics of social networking activities between students to produce online artifacts. This study thus investigated how elementary students teamed and collaborated with peers to create multimedia stories and analyzed their engagement with social network analysis (SNA). This study
TE D
confirmed that the free teaming between students had a positive impact on their engagement. The SNA also revealed that no particular gender showed higher popularity (in-degree centrality) and activeness (out-degree centrality) in the social network. However, students’ knowledge level had a significant impact on their
EP
structural positions in the social network. Students of lower proficiency were more active in the social network while their knowledge level did not directly influence their popularity. Further triangulating the social network with students’ perceptual engagement indicated that students’ structural positions in the social network
AC C
significantly influenced their flow perception and motivation in the networked creation activity. The students who occupied the central position in the out-degree centrality perceived higher level perceptual engagement, while those who occupied the central positions in the in-degree centrality perceived lower level perceptual engagement. The implications of the educational practice are discussed and the direction for future studies is also addressed. Keywords: Learning communities, cooperative/collaborative learning, multimedia/hypermedia systems
evaluation teaching/learning
Correspondence: Chen-Chung Liu Graduate Institute of Network Learning Technology
methodologies, strategies,
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
National Central University No.300, Jhongda Rd., Jhongli City, Taoyuan County 32001, Taiwan (R.O.C.) Telephone: (+886) 3-4227151 ext. 35412 Departmental fax: (+886) 3-4221931 Email: [email protected]
ACCEPTED MANUSCRIPT
RI PT
A social network analysis on elementary student engagement in the networked creation community
Abstract
SC
The recent development of social networking tools has enabled students to collaborate to learn in the social constructivists’ approach. Yet, the learning process in the networked creation
M AN U
community involves complex dynamics of social networking activities among students to produce online artifacts. This study thus investigated how elementary students teamed and collaborated with peers to create multimedia stories and analyzed their engagement with social network analysis (SNA). This study confirmed that the free teaming among students had a positive impact on their engagement. The SNA also revealed that no particular gender showed
TE D
higher popularity (in-degree centrality) and activeness (out-degree centrality) in the social network. However, students’ knowledge level had a significant impact on their structural positions in the social network. Students of lower proficiency were more active in the social
EP
network while their knowledge level did not directly influence their popularity. Further
AC C
triangulating the social network with students’ perceptual engagement indicated that students’ structural positions in the social network significantly influenced their flow perception and motivation in the networked creation activity. The students who occupied the central position in the out-degree centrality perceived higher level perceptual engagement, while those who occupied the central positions in the in-degree centrality perceived lower level perceptual engagement. The implications of the educational practice are discussed and the direction for future studies is also addressed. 1
ACCEPTED MANUSCRIPT
Keywords: Learning communities, evaluation methodologies, cooperative/collaborative learning, teaching/learning strategies, multimedia/hypermedia systems
RI PT
1. Introduction The recent development of social networking tools such as wiki and blog has afforded a new opportunity to implement networked learning where students collaborate to learn in the social
SC
constructivists’ approach. With the aid of these tools, students act as active participants to produce the target content and learning becomes a participatory social process to achieve
M AN U
personal life goals and needs (McLoughlin & Lee, 2007). It was found that while participating in the networked creation activities, students’ application of self-reflection, interaction, and collaboration increased (Ducate & Lomica, 2008; Murray Hourigan & Jeanneau, 2007; Rivens Mompean, 2010). Furthermore, while engaging in the process of producing the target content,
TE D
students were able to develop a deeper level understanding of the content (Liu, Wang & Tai, 2016). Yet, the learning process in the networked creation community involves complex dynamics of social networking activities among students to produce online content. For instance,
EP
in some networked learning communities such as blogs, the social networking activity involves idea exchange in online discussion forums (Lee & Bonk, 2016). Networked creation such as wiki
AC C
editing or open source development is yet another type of learning community where creation of artifacts is the main objective of the community. In such networked creation communities, groups of users frequently align to create the target products (Bergquist & Ljungberg, 2001; Liu, Lin, Deng & Tsai, 2014). The social networking activities may significantly impact the effect of the networked learning activities (Joksimović, Manataki, Gašević, Dawson, Kovanović & de Kereki, 2016). It is thus necessary to understand how students engaged in the networked creation activities to better orchestrate the leaning activities. 2
ACCEPTED MANUSCRIPT
Extensive studies have investigated how students engaged in the networked learning community supported by social networking tools. Most of these studies focused on the perceptual engagement in the networked learning (e.g., Asoodar, Atai, Vaezi, & Marandi, 2014;
RI PT
Ducate et al., 2011; Terumi & Anderson, 2010; Woo, Chu, Ho & Li, 2011). Perceptual
engagement such as sense of enjoyment and the perceived usefulness, satisfaction, confidence, and motivation were analyzed at a general level. Such a body of studies into the networked
SC
learning revealed that networked learning had a positive effect on students’ engagement in terms of their perception (Woo et al., 2011; Hsu Wang & Comac, 2008) and increased students’
M AN U
motivation to learn (Shih, 2011). However, while focusing on general perceptions, it was not well studied regarding how students team and collaborate with peers to participate in the networked creation activity and further how such alignment and collaboration activities may impact student engagement.
TE D
Recently, social network analysis (SNA) (Freeman, 2004) has been widely used to uncover the structural patterns of the linkage between actors in social systems. SNA has been applied increasingly by educational researchers to understand relationship among students in
EP
learning communities in both conventional classrooms (Rienties, Héliot, & Jindal-Snape, 2013)
AC C
or networked learning settings such as MOOCs (Joksimović et al., 2016), blended learning (Lee & Bonk, 2016), online forums (López, Aliaño & Gómez, 2014) or blogs (Jimoyiannis & Angelaina, 2012). Moreover, SNA was found to be helpful to enhance our understanding of students’ engagement in the networked learning as it provides a mathematical manner to measure students’ engagement through their positions in a social network. However, the focus of the current studies was mainly on the social network analysis of networked learning in higher education settings. How elementary students team and collaborate with each other to participate 3
ACCEPTED MANUSCRIPT
in networked creation communities to create multimedia contents and how such social networking activities may impact their perceptual engagement are still not clear in the literature.
RI PT
In this vein, this study attempted to investigate how elementary students teamed and collaborated with peers to create multimedia stories in a networked creation community and analyzes their engagement from both social network analysis and perceptual engagement. At a
SC
macro level, this study investigated whether the free social networking activities, that is, the students have the chance to choose their collaborators to participate in the networked creation
M AN U
activities, would enhance their engagement. To achieve these goals, flow (Csikszentmihalyi, 2000) and strategic motivation framework (Pintrich, Smith, Garcia & McKeachie, 1991) were adopted in this study to obtain a clear picture of students’ perceptual engagement, as they provide sound theoretical frameworks to investigate engagement attributes and motivation. At the micro level, this study explored how elementary students team and collaborate with peer in
TE D
the networked creation activities. SNA was used to depict student social network which was analyzed in relation to students’ gender, knowledge proficiency in the creation domain and their perceptual engagement. By gathering data from 24 elementary students participating in a 20-
EP
week digital storytelling activity on a social network platform, this study aimed to achieve a
AC C
better understanding about elementary students’ engagement in the networked creation community.
2. Related works
2.1 Networked creation community
4
ACCEPTED MANUSCRIPT
The development of the social networking tools, such as wiki and blogs has made the Internet a networked creation community where individuals or groups could share self-created artifacts and interact with audiences. Researchers applied the social networking tools in various approaches to
RI PT
fulfill diverse pedagogical merits, such as reflective learning (Murray et al., 2007) and selfdirected learning (Liu et al., 2016). One of the approaches is to use the networked creation community as a self-presentation tool. In such a line of research, students learned on the
SC
networked learning communities by creating diverse artifacts, such as blogs (Asoodar et al., 2014; Sun, 2010; Ducate & Lomicka, 2008; Murray et al., 2007), voice creations (Hsu, et al.,
M AN U
2008), and personal stories (Ducate, Anderson, & Moreno, 2011) on an individual basis. The networked creation communities become a learning platform where students could publish, reflect upon, and rework on their creations. Another track of studies (e.g. Liu et al., 2016; Woo et al., 2011) incorporated group learning with the network creation activity in which a fixed group
English.
TE D
of students worked together to create artifacts on the networked creation community to learn
However, individuals’ engagement in a social network is a complex process influenced by
EP
multiple factors. Researchers investigated the process of students’ engagement in such a
AC C
community of practice from the perspectives of social cognition or social capital theories (Liu, Lin, Chang, & Chao, 2014). Social cognitive theory asserts that individuals’ cognition and behaviors are shaped under the influence of the social environment (Bandura, 1989). Subjective perceptions including the self-efficacy and the expected value of the participation in the social network significantly influence students’ engagement in the social network (Hsu, Ju, Yen, & Chang, 2007). In this regard, a great body of studies investigated students’ perceptual engagement in the networked creation activity. This track of studies focused on students’ 5
ACCEPTED MANUSCRIPT
subjective perceptions, such as perception of enjoyment (Ducate et al., 2011), usefulness (Terumi & Anderson, 2010), and satisfaction (Sun, 2010). These studies confirmed that at a general level students perceived a positive perception during the participation in the networked creation
RI PT
activity.
To obtain a thorough and deep understanding of students’ perceptual engagement in the creation
SC
activity, researchers also integrated motivation and flow theories to investigate students’ flow perceptions and motivations (Huang, Liu, Wang, Tsai & Lin, 2017; Liu et al., 2016). Their
M AN U
studies found that students’ perceptual engagement is a dynamic process that was significantly influenced by multiple factors, such as students’ knowledge proficiency of the creation domain and the grouping or pairing compositions. Based on social cognitive perspective, these studies investigated students’ perceptions of task value and self-efficacy to understand their perceptual engagement in the social network. Such a track of studies typically relied on data acquired
TE D
through self-reported questionnaires and interviews rather than how students took actions to participate in the activities (Lee & Bonk, 2016). However, these studies did not answer the question of how inter-personal relationships influences learning or perceptual engagement in the
EP
social networking activities.
AC C
To address the above limitation, another track of studies analyzed students’ engagement in social networks from the perspective of social capital theory. Social capital theory claims that the resources which an individual owns do not only exist in the personal properties he/she has, but also in the links through which he/she connects to others (Nahapiet & Ghoshal, 1998). Such an argument suggests that the inter-personal relationship plays an important role in their engagement in social works. The identities of students and the sense of trust among students in a social network, for instance, may significantly impact students’ actions to participate in the 6
ACCEPTED MANUSCRIPT
social network (Liu, Lin, Chang et al., 2014). In particular, the study by Huang et al. (2017) confirmed that the fixed relations during a networked creation activity impeded students’ long term engagement in the activity. However, how students teamed with peers to form pairs or
RI PT
groups in the learning community and how the inter-personal relationships might impact
perceptual engagement in a social network were not clearly addressed in the literature. It is thus worthwhile to investigate students’ social networking activities to better understand their
SC
engagement in the networked creation community.
M AN U
2.2 Social network analysis
Social network analysis (SNA) techniques were reported to be helpful to a better understanding about students’ engagement in networked creation communities. Educators believe that interactive relations among students impact the process and quality of learning (López et al.,
TE D
2014). Understanding the social network among students sheds light on the implementation of learning activities, and, therefore, there is an increasing trend in applying SNA in analyzing student engagement in networked learning activities (Rabbany, Takaffoli, & Zaïane, 2012). SNA
EP
provides both the methods and theory to display, discover and explain the structural patterns of social relationship among students (Giri, Manongga, & Iriani, 2014). It can be integrated with
AC C
quantitative data analysis to achieve a thorough understanding of the learning process (Lee & Bonk, 2016; López et al., 2014). Grounded in the graph theory (Gross & Yellen, 2005), SNA provides statistical methods
to analyze complex interrelationship among students. Several measures were applied to evaluate the social properties of the entire network or entities in the social network. For instance, network density, which displays the ratio of the actual number of ties to the maximum ties between
7
ACCEPTED MANUSCRIPT
entities in a network, can be influenced by pedagogical arrangements such as the teacher’s presence and activity stages (De Laat, Lally, Lipponen & Simons, 2007). Centrality is also a prominent indicator to show the activeness or the prestige of entities in a network as it indicates
RI PT
the extent to which an individual interacts with other members (Wasserman & Faust, 1994). Such an index may reflect a student’s social status in a network. For instance, the in-degree and out-degree centrality which represents the number of ties directed to and that initiated from an
SC
entity represents the prestige and influence of this entity in the network (Rabbany et al., 2012). A previous study found an inverse relationship between the in-degree centralization and the out-
association (López et al., 2014).
M AN U
degree centralization, meaning that students’ prestige and influence has a certain level of
Several studies have applied SNA to understand student interactions in various educational settings and identified several factors, which may be related with students’
TE D
engagement in the social network. For one, students’ gender plays an important role in the social network. The study by Giri et al. (2014) found that female students were more likely to be in the center of the social network than male students as they frequently gave and received messages
EP
while male students tended to play the recipient role. Furthermore, a close relation between
AC C
students’ knowledge proficiency levels and the social centrality was identified by several studies (Russo & Koesten, 2005; Cho, Gay, Davidson & Ingraffea, 2007; Gašević, Zouaq & Janzen, 2013). It was due to the fact that those students who had more sufficient domain knowledge might be more skillful to interact within the network, which in turn, brought benefits to these students. However, a different perspective on the social network suggested that social centrality did not always imply benefits nor represent any superior knowledge level (Krackhardt, 1999). Joksimović et al. (2016) found that the students who located at the positions between two 8
ACCEPTED MANUSCRIPT
strongly connected groups did not significantly benefit from their position. In other words, the relation between the social centrality and students’ knowledge level is not conclusive based on
RI PT
the literature. Furthermore, the findings above obtained from the observation of student interactions in learning environment where the interactions between them were to exchange ideas, not to create
SC
artifacts together. These findings may not be generalized to the context of the networked creation community. This is because the social networks in the above studies addressed the relationship
M AN U
of peer communication with relatively less information on co-work relationship during the creation process. The networked creation community involved dynamic pairing and grouping for learning and working together to achieve a shared goal. Thus, how students team up and collaborate with peers to participate in the networked learning activity require further
TE D
investigation.
Based on the above literature review, one of the goals of this study was to understand the impact of the social networking activities on elementary students’ engagement in the networked
EP
creation community. At a general level, this study intended to compare students’ engagement in two settings based on the flow and motivation theory. One of the settings afforded students to
AC C
freely team up with peers (dynamic teaming) and the other allowed them only to work with a fixed peer (fixed relation). Comparing students’ perceptual engagement in the two settings may shed light on the effect of free teaming in collaborative learning as perceptions and beliefs are influenced through the structural mechanisms of social networks (Jostad, Sibthorp & Paisley, 2013). At a detail level, this study analyzed students’ social network to examine what factors may affect students’ teaming and collaboration patterns. More specifically, students’ knowledge proficiencies and their genders would be analyzed in relation to their social network properties to 9
ACCEPTED MANUSCRIPT
uncover the role that gender and knowledge proficiencies may play in the social network. Last but not least, this study triangulated how students’ perceptual engagement in the social network may be related to their social network properties from both of the social cognitive and capital
RI PT
perspectives. More specifically, the inter-personal relationship and students’ perceptual
engagement were analyzed to uncover the relationship between the two constructs by using the flow, motivation questionnaires and SNA. In sum, this present intended to answer the following
SC
three research questions:
networked creation activity?
M AN U
RQ1: Do the dynamic teaming activities impact the students’ flow and motivation in the
RQ2: How do the students team and collaborate with peers to participate in the networked creation activity?
TE D
RQ3: Do the students’ properties in the social network have an impact on their flow and
3. Method
AC C
3.1 Participants
EP
motivation in the networked creation activity?
This study is part of a year-long program aiming at helping students develop their proficiency in English, computer use, and collaboration through collaborative digital storytelling. The participants were 26 third grade students in an elementary school in northern Taiwan, aged between 8 and 9. They participated in the networked creation activity to create multimedia stories collaboratively with peers. The students started learning English as foreign language from grade 1. Therefore, they were still in the initial stage of learning English with the focus on 10
ACCEPTED MANUSCRIPT
vocabulary and sentence patterns. Before the program, the students did not have prior experience in using computers to create multimedia stories with peers.
RI PT
3.2 Instruments 3.2.1 Storytelling and social networking
To provide participants the opportunity to create multimedia stories on the Internet, Story &
SC
Painting House (Liu et al., 2016), a social networking application on iPads connecting to an
M AN U
online platform, was developed to facilitate the process of the networked creation activity. Story & Painting House supports multimedia authoring functions, including easy-to-use drawing tools, text, and the ability/option to record oral narrations (See Figure 1).
The application also supports social networking facilities by which students could invite
TE D
peers to collaborate synchronously to create a shared multimedia story through iPads. The application depicts the social network showing the collaboration relationship between participants in the community (the up-right of Figure 1). Story & Painting House also allows
EP
participants to share their stories on the online platform. As shown in Figure 1, the platform
AC C
provides a repository to show the multimedia stories in progress and the published ones. ----------------------------------Insert Figure 1 here -----------------------------------
Moreover, participants could freely invite their peers to collaborate during the phase of the dynamic teaming activity. All teaming up activities were logged on the Story & Painting
11
ACCEPTED MANUSCRIPT
House platform. Once an invitation was sent, it would be instantly logged by the system and a link from inviter to invitee would be presented in the SNA tool for further analysis.
RI PT
3.2.2 The flow perception survey A flow perception survey was administrated to probe students’ perceptual engagement associated with the networked creation activity since engagement and flow shared several psychological
SC
attributes (Trevino & Webster, 1992). The survey has been extensively applied and showed adequate reliability in previous literature (Huang et al., 2017; Liu et al., 2016) indicating that this
M AN U
survey is a reliable instrument for assessing the students’ perceptual engagement. Guided by the Trevino and Webster’s flow framework, this survey probes student engagement in four basic flow components including the level of control, attention, curiosity and intrinsic interest that participants perceived when they participated in the networked creation activity. Thus, this
TE D
survey consists of only four 5-point Likert items (with 5 as the highest level and 1 as the lowest) and was administrated right at the end of each activity section where there was no interruptions to participants’ learning activity.
EP
3.2.3 Motivated Strategies for Learning Questionnaire (MSLQ)
AC C
While the flow survey was conducted to obtain students’ short term engagement right after each activity section, the Motivated Strategies for Learning Questionnaire (MSLQ) (Pintrich et al., 1991) was conducted to uncover the long-term impact of the networked creation activity on participants’ motivation associated with the fixed relation and dynamic teaming activities. The MSLQ was adapted according to the context of the activity. The questionnaire includes 25 items asking students’ motivation in the dimensions of self-efficacy, extrinsic/intrinsic goal orientation, task value and peer learning, as these components were directly related to the context of the 12
ACCEPTED MANUSCRIPT
study. The questionnaire was found to be adequately reliable as the Cronbach reliability (alpha) for the extrinsic/intrinsic goal orientation, self-efficacy, task value and peer learning are .68, .78,
RI PT
.81, .83 and .61 respectively. 3.3 Procedure
The networked creation activity was implemented during a semester, for an 80-minute session
SC
per week. Participants formed pairs to create multimedia stories using Story & Painting House. Each student was provided one iPad with an e-book containing 15 stories, which included texts
M AN U
with oral narrations so that students can learn the stories by themselves. Clear instruction was provided for each pair to participate in the activity: (1) Starting reading any story of their preference; (2) Working collaboratively using the storytelling application to retell the story in a multimedia picture book format, including drawings and oral narrations; and (3) Publishing and
TE D
sharing the picture books on the online platform where all participants could view the picture books. Upon finishing a story, they were instructed to continue working on re-telling and publishing the rest of the multimedia stories at their own pace.
EP
Participants experienced the networked creation activity in two different formats during
AC C
the semester. In the first 5 weeks, students were asked to participate in the activity without the dynamic teaming mechanism, hence, the fixed relation, meaning that they formed pairs by themselves and only collaborated with that partner for the multimedia stories activity. The homeroom teacher helped those who needed assistance finding a partner. Starting from week 6, students were instructed that they could freely team up with any peer to create new multimedia stories, hence in the dynamic teaming activity, which lasted for 8 weeks (week 6 to 13).
13
ACCEPTED MANUSCRIPT
The students went through a total of 13 sections. After each activity session, participants answered the flow perception survey. Therefore, 13 sets of flow survey results were obtained. Furthermore, the MSLQ was administered to assess participants’ motivation in week 5 and 13
RI PT
right after they completed the fixed relation and the dynamic teaming activity respectively. The participants’ feedback to the flow perception survey and MSLQ were analyzed to understand
SC
their perceptual engagement associated with the two networked creation activities. 3.4 Data analysis
M AN U
To answer the research question 1, participants’ flow perceptions and motivations associated with the fixed relation and dynamic teaming activities were compared. As the flow survey was administrated right after each activity section each week, participants’ flow perceptions in the first 5 weeks were averaged to obtain an overall flow perception about the fixed relation activity.
TE D
The same procedure was applied to their flow perceptions in dynamic teaming activity. Participants’ motivation and flow perception in the two settings were also analyzed with dependent t-tests.
EP
Participants’ invitation activities on the social network were analyzed with the UCINET
AC C
SNA tool (Borgatti, Everett & Freeman, 2002) to answer RQ2. The out- and in-degree centrality of each participant reflecting how an individual invited peers or received invitations from peers was computed to represent their participation in the social network. The degree centrality was analyzed with respect to genders and English proficiencies to understand how participants with different background team up with peers in the multimedia story creation activity. Regarding research question 3, the impact of participants’ participation in the social network (i.e. out-degree and in-degree centrality) on their perceptual engagement (i.e. flow 14
ACCEPTED MANUSCRIPT
perceptions and motivation) was analyzed. To achieve this goal, their perceptual engagement was also compared based on their degree centrality with t-test, thus, the comparative result can be shown from the integrative analysis of the social network and perceptual engagement that
RI PT
possibly reveals the impact of social network activity on learning experience. 4. Results
SC
4.1 Perceptual engagement in fixed relation and dynamic alignment activities
M AN U
Figure 2 shows participants’ weekly flow perceptions during the 13 weeks. It is shown that participants perceived relatively lower level of flow in the first 5 weeks when the fixed relation activity was implemented and the flow perception was at a higher level during week 6 – 13 when the dynamic teaming activity was administered. As shown in Table 1, the t-test results of participants’ average flow perception indicated that it was significantly associated with the two
TE D
different types of creation activities. When participating in the dynamic teaming activity, participants’ average flow perception was 4.52, which is significantly higher than that in fixed relation activity (4.12) (t=-3.22, p<.01). A closer analysis of the flow dimensions found that
EP
participants perceived a higher level of control (t=-3.46, p<.01), attention (t=-3.56, p<.01) and
AC C
curiosity (t=-2.26, p<.05) associated with the dynamic teaming activity than with the fixed relation activity.
----------------------------------Insert Figure 2 here --------------------------------------------------------------------Insert Table 1 here
15
ACCEPTED MANUSCRIPT
----------------------------------Table 2 lists the t-test results of the students’ motivation during the fixed relation and
RI PT
dynamic teaming activities. Participants perceived a significantly higher level of motivation in the dynamic teaming activity than they did in the fixed-relation activity in two dimensions: intrinsic motivation (t=-3.08, p <.01) and task value (t=-2.51, p<.05). It indicated that
SC
participants showed an enhanced perceptual engagement when they were able to dynamically select their own collaborating peers in the networked creation activity.
M AN U
----------------------------------Insert Table 2 here
----------------------------------4.2 Social network analysis
TE D
UCINET was applied to analyze the formation of participants’ invitation activities in the social network in relation to participants’ genders (F and M) and English proficiency levels (High, Middle and Low) (See Figure 3). The network involved 26 participants with 46 ties showing a
EP
network density of .07. By using the subgrouping analysis with factions, the social network
AC C
could be divided into two subgroups: the girls group (left) and the boys group (right). As shown in Figure 3 only 4 ties crosslinked the participants in the two different subgroups, which indicated that elementary students tended to team up with students of the same gender when participating in the networked creation activity. Figure 3 also showed the in-degree centrality of the nodes in the social network ranging from 0 to 7. Among the 12 students who received more invitations to form pairs for the networked creation activity and displayed higher in-degree centrality (centrality ≥2), 6 were girls 16
ACCEPTED MANUSCRIPT
and 6 boys (in the circle) suggesting that no particular gender contributed to higher popularity. It was also found that the 12 students showed diverse English proficiency levels, (4 at high-level, 5
influence students’ popularity during the creation activity. ----------------------------------Insert Figure 3 here
SC
-----------------------------------
RI PT
at mid-level, and 3 at low-level), indicating that the students’ content knowledge did not directly
M AN U
Furthermore, participants’ out-degree centrality was also analyzed to show their activeness while working on the activity. As seen in Figure 4, participants’ out-degree centrality ranged from 0 to 7 and 9 of them (located in the circle) showed higher out-degree centrality (centrality ≥2). These students actively teamed up with peers for the networked creation activity. It was found that participants’ English proficiency levels were related participants’ activeness in
TE D
their teaming up with peers. The 9 students in the high out-degree centrality were from the mid to low level proficiency groups (6 from low-level and 3 from mid-level), indicating that the
EP
lower-proficiency students were more active in teaming up with peers in the activity. However, genders were found not related because the high out-degree centrality group contained almost
AC C
equal number of students from the two genders, four girls and five boys. ----------------------------------Insert Figure 4 here -----------------------------------
17
ACCEPTED MANUSCRIPT
4.3 Impact of social network on perceptual engagement Participants’ perceptual engagement was analyzed in relation to their in- and out-degree
RI PT
centrality with t-test. Regarding the in-degree centrality, the 12 students who showed higher indegree centrality (centrality ≥2) perceived a significantly lower perceptual engagement than the other 14 who showed lower in-degree centrality (centrality <2). Table 3 displays the results of
SC
the t-test analysis of the two groups. Such significant difference was found in the level of overall flow perception (t=2.31, p<.05), control (t=2.87, p<.05), intrinsic (t=2.25, p<.05), extrinsic
M AN U
motivation (t=2.15, p<.05), and the task value (t=2.32; p<.05).
----------------------------------Insert Table 3 here
-----------------------------------
TE D
The out-degree centrality also had a significant influence on the perceptual engagement. As shown in Figure 4, the 9 students who showed higher out-degree centrality (centrality ≥2) perceived higher level of perceptual engagement than the other 17 students in the lower out-
EP
degree centrality (centrality <2) (See Table 4). It was shown that the high out-degree centrality group perceived significantly higher level of overall flow perception (t=3.25, p<.01), control
AC C
(t=2.55, p<.05), curiosity (t=3.71, p<.01) interest (t=2.37, p<.05) and significantly higher level of task value (t=2.52, p<.05).
----------------------------------Insert Table 4 here -----------------------------------
18
ACCEPTED MANUSCRIPT
5. Conclusions and discussion Examining students’ perceptual engagement and their participation in the social network for the
RI PT
13-week networked creation activities, this study found that the students demonstrated improved engagement as they could freely team up with peers to create multimedia stories. Both their flow perceptions and motivations increased as well. The social network analysis revealed that
SC
students’ prominent features including gender and proficiency levels of the target content had an impact on their participation in the social networking activities. Moreover, students’
M AN U
participatory features including in-degree and out-degree centrality in the social network showed a significant impact on their flow perceptions and motivations. The results of this study indicated that triangulating data from flow, motivation, and the social network were useful to help researchers understand student engagement in the networked creation activities from different
TE D
perspectives.
Many studies (Huang et al., 2017; Azmitia, 1988; Liu & Tsai, 2008; Rogoff, 1990) have confirmed pairing or grouping schemes impacted students’ engagement in collaborative learning
EP
activities. The study by Huang et al. (2017) further confirmed a student who had a long-term partnership with one fixed student had the tendency to cause uneven relationship between two
AC C
students in which one student played the guiding role to help the other student. The uneven roles largely decreased students’ engagement in the collaborative learning activity. Huang et al. suggested that a mechanism that breaks down such an uneven role distribution is necessary to restore students’ engagement. This study found that the students perceived an enhanced flow perception and motivation when they were able to dynamically team up with peers. With the opportunity for freely teaming up, students had full control during the creation process and avoided the risk of long-term uneven role. This explained why students perceived higher level of 19
ACCEPTED MANUSCRIPT
perceptual engagement than they did in the fixed relation activity. In other words, educators could take advantage of the dynamic pairing or grouping mechanism to sustain students’
RI PT
engagement in other networked learning activities. The SNA of this study indicated that no particular gender showed higher popularity (indegree centrality) and activeness (out-degree centrality) in the social network and boys and girls
SC
mainly interacted with peers of the same gender. Such social network structure is different from what Giri et al. (2014) identified where female students were reported to more likely be the
M AN U
center of social networks and more actively participated in the social network than male students. Several factors may explain the different findings between this study and Giri et al. (2014). For instance, students of this study were elementary students while participants in Giri et al. (2014) study were university students. Furthermore, the features of the social network activities of the two studies were also significantly different. While the students of this study had to work closely
TE D
with peers, the students in Giri et al. (2014) study solely exchanged ideas on Internet without the need to work with peers. In other words, the findings of the present study contributed to draw attention of educators and researchers to take age and the types of social networking activities
EP
into account while analyzing social networks in other educational contexts.
AC C
This study found that students’ proficiency level in the target content was not related to
their popularity (in-degree centrality) when working on the activities. In other words, students’ content knowledge level required in the activity did not directly influence their popularity in the network. However, the finding of low proficiency level students showing higher out-degree centrality in the social network revealed that the low proficiency level students were more active in teaming up with peers than the high proficient students, which conflicted the results of previous studies (Cho et al., 2007; Gašević et al., 2013; Russo & Koesten, 2005) where students 20
ACCEPTED MANUSCRIPT
of higher knowledge level occupied the central positions in social networks. This may be due to the particular nature of the social networking activity where students had to form pairs to create shared multimedia stories. In other words, students of low proficiency level needed to actively
RI PT
seek help from others so that they could accomplish the shared goal with peers, which is
different from the information sharing activities in previous studies, such as online discussion
SC
where students were not bound to achieve a shared goal.
The analyses on students’ flow perception, motivations and the social network revealed
M AN U
an interesting relationship between students’ perceptual engagement and their social networks. The students’ out-degree centrality had a positive impact on students’ flow perception and motivation. In other words, the students who occupied the central position in the out-degree centrality perceived higher level perceptual engagement. However, the high in-degree centrality had a negative impact on flow perceptions and motivation. The students who occupied the
TE D
central positions in the in-degree centrality perceived lower level perceptual engagement. As the in-degree centrality reflected an entity’s popularity (Rabbany et al., 2012), the finding suggested that popularity did not imply high level of engagement in the networked learning activity. On the
EP
contrary, the out-degree centrality may appropriately represent the students’ engagement in the
AC C
activity as it had a positive relationship with students’ flow and motivation. Educators could use the two indicators to monitor students’ engagement in other types of networked learning activities.
The results of this study contributed to a sound understanding of students’ engagement in
networked creation community and suggested that free pairing strategy among students had a positive impact on their engagement. It was concluded that knowledge level had an impact on students’ engagement in the social network. Lower-proficiency students were more active in 21
ACCEPTED MANUSCRIPT
teaming up with peers in the activity. Furthermore, the individual student’s positions in the social network had a significant impact on their flow and motivation in the networked creation activity. However, this study is an exploratory study where students’ engagement was not compared with
RI PT
that of a control group. Further comparative investigation with experiments is still needed with a wider sample to provide additional evidences. Furthermore, it would be interesting to see how the patterns of students’ social network may change over time. The timeline analysis of social
SC
networks may uncover how the interactions among students may evolve in a long-term
networked learning activity. It would also be worthwhile to extend the investigation to the other
M AN U
types of social networking activities, such as knowledge construction through online discussion. Gathering information on these issues through further studies can help obtain a more thorough understanding of student engagement in a networked learning community, which would help
AC C
EP
TE D
educators and teachers implement effective networked learning activities.
22
ACCEPTED MANUSCRIPT
References Asoodar, M., Atai, M. R., Vaezi, S. & Marandi, S. S. (2014). Examining effectiveness of
RI PT
communities of practice in online English for academic purposes (EAP) assessment in virtual classes. Computers & Education, 70, 291-300.
Azmitia, M. (1988). Peer interaction and problem solving: When are two heads better than one?
SC
Child Development, 59, 87–96.
M AN U
Bandura, A. (1989). Social cognitive theory. In R. Vasta (Ed.), Annals of Child Development (pp. 1-60). Greenwich, CT: Jai Press LTD.
Bergquist, M. & Ljungberg, J. (2001). The power of gifts: organizing social relationships in open
TE D
source communities. Information Systems Journal, 11(4), 305-320. Borgatti, S.P., Everett, M.G., & Freeman, L.C. (2002). Ucinet for Windows: Software for Social Network Analysis. Harvard, MA: Analytic Technologies.
EP
Cho, H., Gay, G. , Davidson, B., & Ingraffea, A. (2007). Social networks, communication styles,
AC C
and learning performance in a CSCL community, Computers & Education, 49(2), 309–329. Csikszentmihalyi, M. (2000). Beyond boredom and anxiety. Jossey-Bass. De Laat, M., Lally, V., Lipponen, L., & Simons, R.-J. (2007). Investigating patterns of interaction in networked learning and computer-supported collaborative learning: a role for Social Network Analysis. International Journal of Computer-Supported Collaborative Learning, 2(1), 87-103.
23
ACCEPTED MANUSCRIPT
Ducate, L. C., Anderson, L. L. & Moreno, N. (2011) Wading through the world of wikis: An analysis of three wiki projects. Foreign Language Annals, 44(3): 495-524.
RI PT
Ducate, L. C. & Lomicka, L. L. (2008). Adventures in the blogosphere: From blog readers to blog writers. Computer Assisted Language Learning, 21(1), 9-28.
Freeman, L. C. (2004). The development of social network analysis: a study in the sociology of
SC
science. Vancouver, B. C.: Empirical Press.
M AN U
Gašević, D., Zouaq, A., & Janzen R. (2013). Choose your classmates, your GPA is at stake!: the association of cross-class social ties and academic performance, American Behavioral Scientist, 57, 1460-1479.
Gross, J. L., & Yellen, J. (2005). Graph theory and its applications. CRC press.
TE D
Giri, B. E., Manongga, D. & Iriani, A. (2014). Using Social Networking Analysis (SNA) to Analyze Collaboration between Students (Case Study: Students of Open University in
EP
Kupang), International Journal of Computer Applications (0975 – 8887), 85(1), 44 – 49. Hsu, M. H., Ju, T. L., Yen, C. H., & Chang, C. M. (2007). Knowledge sharing behavior in virtual
AC C
communities: The relationship between trust, self-efficacy, and outcome expectations. International Journal of Human-Computer Studies, 65(2), 153-169. Hsu, H.-Y., Wang, S.-K. & Comac, L. (2008). Using audioblogs to assist English-language learning: An investigation into student perception. Computer Assisted Language Learning, 21(2), 181-198.
24
ACCEPTED MANUSCRIPT
Huang, Y. Y., Liu, C. C., Wang, Y., Tsai, C. C. & Lin H. M. (2017). Student engagement in long-term collaborative EFL storytelling activities: An analysis of learners with English
RI PT
proficiency differences. Educational Technology & Society, 20 (3), 95–109. Jimoyiannis, A. and Angelaina, S. (2012). Towards an analysis framework for investigating students’ engagement and learning in educational blogs. Journal of Computer Assisted
SC
Learning, 28(3), 222-234.
Joksimović, S., Manataki, A., Gašević, D., Dawson, S., Kovanović, V., & de Kereki, I. F. (2016).
M AN U
Translating Network Position into Performance: Importance of Centrality in Different Network Configurations. In Proceedings of the Sixth International Conference on Learning Analytics & Knowledge (pp. 314–323). New York, NY, USA: ACM. Jostad J., Sibthorp, J. & Paisley, K. (2013). Understanding groups in outdoor adventure
17-31, 2013.
TE D
education through social network analysis. Australian Journal of Outdoor Education, 17(1),
EP
Krackhardt, D. (1999). The ties that torture: Simmelian tie analysis in organizations. Research in the Sociology of Organizations, 16, 83-210.
AC C
Lee, J. & Bonk, J. C. (2016). Social network analysis of peer relationships and online interactions in a blended class using blogs. Internet and Higher Education, 28, 35-44. Liu, C. C., Lin, C. C., Chang, C. Y., & Chao, P. Y. (2014). Knowledge Sharing among University Students Facilitated with a Creative Commons Licensing Mechanism: A Case Study in a Programming Course. Journal of Educational Technology & Society, 17(3), 154-167. 25
ACCEPTED MANUSCRIPT
Liu, C.-C., Lin, C.-C., Deng, G.-Y., & Tsai, C.-C. (2014). Online knowledge sharing experience with Creative Commons. Online Information Review, 38(5), 680-696.
RI PT
Liu, C. C. & Tsai, C. C. (2008). An analysis of peer interaction patterns as discoursed by on-line small group problem-solving activity. Computers & Education, 50(3), 627-639.
Liu, C. C., Wang, P. C., & Tai, D. S. J. (2016). An analysis of student engagement patterns in
SC
language learning facilitated by Web 2.0 technologies. ReCALL, 28(02), 104-122.
M AN U
López, P. M., Aliaño, A. M. & Gómez, J. I. A. (2014). Social network analysis of a blended learning experience in higher education, Research on Education and Media, 6(2), 69-78. McLoughlin, C., & Lee, M. J. W. (2007). Social software and participatory learning: pedagogical choices with technology affordances in the Web 2.0 era. Paper presented at the Ascilite,
TE D
Singapore.
Murray, L., Hourigan, T. & Jeanneau, C. (2007). Blog writing integration for academic language
EP
learning purposes: towards an assessment framework. IBÉRICA, 14, 9-32. Nahapiet, J., & Ghoshal, S. (1998). Social capital, intellectual capital, and the organizational
AC C
advantage. The Academy of Management Review, 23(2), 242-266. Pintrich, P. R., Smith, D., Garcia, T. & McKeachie, W. (1991). A manual for the use of the motivated strategies for learning questionnaire (MSLQ). Ann Arbor Michigan: The University of Michigan.
26
ACCEPTED MANUSCRIPT
Rabbany, R., Takaffoli, M., & Zaïane, O. R. (2012). Social network analysis and mining to support the assessment of on-line student participation. ACM SIGKDD Explorations
RI PT
Newsletter, 13(2), 20–29. Rienties, B., Héliot, Y., & Jindal-Snape, D. (2013). Understanding social learning relations of international students in a large classroom using social network analysis. Higher
SC
Education,66(4), 489–504.
Rivens Mompean, A. (2010). The development of meaningful interactions on a blog used for the
M AN U
learning of English as a Foreign Language. ReCALL, 22(3), 376-395.
Rogoff, B. (1990). Apprenticeship in Thinking: Cognitive Development in Social Context. Oxford: Oxford University Press.
TE D
Russo, T. C. & Koesten, J. (2005). Prestige, centrality, and learning: A social network analysis of an online class, Communication Education, 54(3), 254–261. Shih, R. C. (2011). Can Web 2.0 technology assist college students in learning English writing?
EP
Integrating Facebook and peer assessment with blended learning. Australasian Journal of
AC C
Educational Technology, 27(5), 829-845. Sun, Y. C. (2010) Extensive writing in foreign‐language classrooms: A blogging approach. Innovations in Education and Teaching International, 47(3), 327-339. Terumi, M. T. & Anderson, T. (2010). Learning outcomes and students’ perceptions of online writing: Simultaneous implementation of a forum, blog, and wiki in an EFL blended learning setting. System, 38(2): 185-199.
27
ACCEPTED MANUSCRIPT
Trevino, L. K. & Webster, J. (1992) Flow in computer-mediated communication: electronic mail and voice mail evaluation and impacts. Communication Research, 19(5), 539-573.
RI PT
Wasserman, S., & Faust, K. (1994). Social network analysis: Methods and applications. Cambridge university press.
Woo, M., Chu, S., Ho, A. & Li, X. (2011). Using a wiki to scaffold primary-school students'
AC C
EP
TE D
M AN U
SC
collaborative writing. Educational Technology and Society, 14(1): 43-54.
28
ACCEPTED MANUSCRIPT Acknowledgements This research was partially funded by the Ministry of Science and Technology under
AC C
EP
TE D
M AN U
SC
RI PT
contract number 103-2511-S-008-014-MY3.
ACCEPTED MANUSCRIPT List of Tables Table 1. The t-test results of the students’ flow perceptions during the two networked creation activities (N=26) Table 2. The t-test results of the students’ motivation during the two networked learning activities (N=26)
AC C
EP
TE D
M AN U
SC
RI PT
Table 3. The t-test results of flow perceptions and motivation for different levels of in-degree centrality Table 4. The t-test results of flow perceptions and motivation for different levels of out-degree centrality
ACCEPTED MANUSCRIPT Table 1. The t-test results of the students’ flow perceptions during the two networked creation activities (N=26)
Curiosity Interest
t
p
Fixed Dynamic Fixed
4.12 4.52 4.28
.67 .43 .64
-3.22**
<.01
Dynamic Fixed Dynamic
4.63 3.82 4.31
.49 .94 .63
Fixed Dynamic Fixed Dynamic
4.13 4.52 4.26 4.60
.85 .55 .86 .43
AC C
EP
TE D
**p<.01；*p<.05
-3.46** -3.56** -2.26*
RI PT
Attention
SD
<.01 <.01
.03
SC
Control
M
-2.00
M AN U
Overall
Period
.06
ACCEPTED MANUSCRIPT Table 2. The t-test results of the students’ motivation during the two networked learning activities (N=26) M
SD
T
p
Intrinsic
Fixed Dynamic
4.14 4.53
.83 .63
-3.08**
<.01
Extrinsic
Fixed Dynamic
3.91 3.83
.87 .92
.46
.65
Task value
Fixed
4.16
.83
-2.51*
Dynamic Fixed
4.56 3.59
.63 1.02
Dynamic
3.97
1.01
Fixed Dynamic
3.95 4.28
.70 .81
AC C
EP
TE D
**p<.01；*p<.05
.02
.19
SC
Self-efficacy
-1.36
-1.96
M AN U
Peer learning
RI PT
Period
.06
ACCEPTED MANUSCRIPT Table 3. The t-test results of flow perceptions and motivation for different levels of in-degree centrality Centrality
N
M
SD
t
p
High
12
4.32
.47
2.31*
.03
Low
14
4.68
High
12
4.35
Low
14
4.87
High
12
Low
14
Control
Curiosity
4.06
.73
4.53
.46
12
4.37
.56
Low
14
4.66
.52
High
12
4.51
.45
Low
14
4.68
.40
High
12
4.25
.64
Low
14
4.77
.53
Extrinsic
High
12
3.42
1.01
Low
14
4.16
.69
High
12
4.26
.80
Low
14
4.82
.26
High
12
3.64
.74
Low
14
4.26
1.13
High
12
4.11
.70
Low
14
4.42
.89
TE D
Intrinsic
AC C
Task Value
Peer learning
Self-efficacy
**p<.01；*p<.05
2.87*
.01
2.01
.06
1.34
.19
1.04
.31
2.25*
.03
2.15*
.04
2.32*
.04
1.63
.12
.96
.34
.18
High
Interest
Motivation
.59
EP
Attention
.31
SC
Overall
M AN U
Flow
RI PT
(in-degree)
ACCEPTED MANUSCRIPT Table 4. The t-test results of flow perceptions and motivation for different levels of out-degree centrality Dimension
Centrality
N
M
SD
High
9
4.78 .18
Low
17
4.38 .46
High
9
4.88 .18
Low
17
4.50 .55
High
9
Low
17
t
p
(out-degree)
Attention
Curiosity
Interest
Intrinsic
EP
Extrinsic
AC C
Task Value
Peer learning
Self-efficacy
**p<.01；*p<.05
4.57 .51
2.55** .02
1.54
.14
4.18 .66
High
9
4.89 .13
Low
17
4.33 .59
High
9
4.81 .18
Low
17
4.50 .48
High
9
4.72 .65
Low
17
4.43 .61
High
9
4.16 .67
Low
17
3.65 .99
High
9
4.87 .18
Low
17
4.40 .72
High
9
4.11 1.34 .50
Low
17
3.90 .81
High
9
4.26 1.08 .07
Low
17
4.27 .66
TE D
Motivation
3.25** <.01
RI PT
Control
SC
Overall
M AN U
Flow
3.71** <.01
2.37*
.03
1.15
.26
1.40
.17
2.52*
.02
.62
.95
ACCEPTED MANUSCRIPT List of Figures Figure 1. The social network and the online repository of multimedia stories supported by Story & Painting House Figure 2. The weekly flow perceptions of the students during the 13 activity weeks Figure 3. The social network with the high in-degree centrality at the center
AC C
EP
TE D
M AN U
SC
RI PT
Figure 4. The social network with the high out-degree centrality at the center
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
Figure 1. The social network and the online repository of multimedia stories supported by Story & Painting House
ACCEPTED MANUSCRIPT 4.8
4.6
RI PT
4.4
4.2
SC
4
3.6 1
2
3
4
5
M AN U
3.8
6
7
8
9
10
11
12
13
AC C
EP
TE D
Figure 2. The weekly flow perceptions of the students during the 13 activity weeks
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Figure 3. The social network with the high in-degree centrality at the center
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Figure 4. The social network with the high out-degree centrality at the center
ACCEPTED MANUSCRIPT Highlights Networked creation activities had a positive effect on students’ perceptual engagement. This study analyzed students’ engagement in social network to learn through creation. Dynamic pairing or grouping mechanism is helpful to sustain students’
AC C
EP
TE D
M AN U
SC
RI PT
engagement. Knowledge level had an impact on students’ position in the social network. In- and out-degree centrality showed significant influence on perceptual engagement.
S0360-1315(17)30188-4
DOI:
10.1016/j.compedu.2017.08.002
Reference:
CAE 3223
To appear in:
Computers & Education
Received Date: 4 March 2017 Revised Date:
5 August 2017
Accepted Date: 6 August 2017
Please cite this article as: Liu C.-C., Chen Y.-C. & Daian Tai S.-J., A social network analysis on elementary student engagement in the networked creation community, Computers & Education (2017), doi: 10.1016/j.compedu.2017.08.002. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
A social network analysis on elementary student engagement in the networked creation community Chen-Chung Liu1, Yu-Chi Chen1, Shu-Ju Daian Tai2 1
RI PT
Graduate Institute of Network Learning Technology, National Central University, Taiwan 2 Beijing University of Chemical Technology, School of International Education, Beijing, CN
M AN U
SC
Abstract The recent development of social networking tools has enabled students to collaborate to learn in the social constructivists’ approach. Yet, the learning process in the networked creation community involves complex dynamics of social networking activities between students to produce online artifacts. This study thus investigated how elementary students teamed and collaborated with peers to create multimedia stories and analyzed their engagement with social network analysis (SNA). This study
TE D
confirmed that the free teaming between students had a positive impact on their engagement. The SNA also revealed that no particular gender showed higher popularity (in-degree centrality) and activeness (out-degree centrality) in the social network. However, students’ knowledge level had a significant impact on their
EP
structural positions in the social network. Students of lower proficiency were more active in the social network while their knowledge level did not directly influence their popularity. Further triangulating the social network with students’ perceptual engagement indicated that students’ structural positions in the social network
AC C
significantly influenced their flow perception and motivation in the networked creation activity. The students who occupied the central position in the out-degree centrality perceived higher level perceptual engagement, while those who occupied the central positions in the in-degree centrality perceived lower level perceptual engagement. The implications of the educational practice are discussed and the direction for future studies is also addressed. Keywords: Learning communities, cooperative/collaborative learning, multimedia/hypermedia systems
evaluation teaching/learning
Correspondence: Chen-Chung Liu Graduate Institute of Network Learning Technology
methodologies, strategies,
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
National Central University No.300, Jhongda Rd., Jhongli City, Taoyuan County 32001, Taiwan (R.O.C.) Telephone: (+886) 3-4227151 ext. 35412 Departmental fax: (+886) 3-4221931 Email: [email protected]
ACCEPTED MANUSCRIPT
RI PT
A social network analysis on elementary student engagement in the networked creation community
Abstract
SC
The recent development of social networking tools has enabled students to collaborate to learn in the social constructivists’ approach. Yet, the learning process in the networked creation
M AN U
community involves complex dynamics of social networking activities among students to produce online artifacts. This study thus investigated how elementary students teamed and collaborated with peers to create multimedia stories and analyzed their engagement with social network analysis (SNA). This study confirmed that the free teaming among students had a positive impact on their engagement. The SNA also revealed that no particular gender showed
TE D
higher popularity (in-degree centrality) and activeness (out-degree centrality) in the social network. However, students’ knowledge level had a significant impact on their structural positions in the social network. Students of lower proficiency were more active in the social
EP
network while their knowledge level did not directly influence their popularity. Further
AC C
triangulating the social network with students’ perceptual engagement indicated that students’ structural positions in the social network significantly influenced their flow perception and motivation in the networked creation activity. The students who occupied the central position in the out-degree centrality perceived higher level perceptual engagement, while those who occupied the central positions in the in-degree centrality perceived lower level perceptual engagement. The implications of the educational practice are discussed and the direction for future studies is also addressed. 1
ACCEPTED MANUSCRIPT
Keywords: Learning communities, evaluation methodologies, cooperative/collaborative learning, teaching/learning strategies, multimedia/hypermedia systems
RI PT
1. Introduction The recent development of social networking tools such as wiki and blog has afforded a new opportunity to implement networked learning where students collaborate to learn in the social
SC
constructivists’ approach. With the aid of these tools, students act as active participants to produce the target content and learning becomes a participatory social process to achieve
M AN U
personal life goals and needs (McLoughlin & Lee, 2007). It was found that while participating in the networked creation activities, students’ application of self-reflection, interaction, and collaboration increased (Ducate & Lomica, 2008; Murray Hourigan & Jeanneau, 2007; Rivens Mompean, 2010). Furthermore, while engaging in the process of producing the target content,
TE D
students were able to develop a deeper level understanding of the content (Liu, Wang & Tai, 2016). Yet, the learning process in the networked creation community involves complex dynamics of social networking activities among students to produce online content. For instance,
EP
in some networked learning communities such as blogs, the social networking activity involves idea exchange in online discussion forums (Lee & Bonk, 2016). Networked creation such as wiki
AC C
editing or open source development is yet another type of learning community where creation of artifacts is the main objective of the community. In such networked creation communities, groups of users frequently align to create the target products (Bergquist & Ljungberg, 2001; Liu, Lin, Deng & Tsai, 2014). The social networking activities may significantly impact the effect of the networked learning activities (Joksimović, Manataki, Gašević, Dawson, Kovanović & de Kereki, 2016). It is thus necessary to understand how students engaged in the networked creation activities to better orchestrate the leaning activities. 2
ACCEPTED MANUSCRIPT
Extensive studies have investigated how students engaged in the networked learning community supported by social networking tools. Most of these studies focused on the perceptual engagement in the networked learning (e.g., Asoodar, Atai, Vaezi, & Marandi, 2014;
RI PT
Ducate et al., 2011; Terumi & Anderson, 2010; Woo, Chu, Ho & Li, 2011). Perceptual
engagement such as sense of enjoyment and the perceived usefulness, satisfaction, confidence, and motivation were analyzed at a general level. Such a body of studies into the networked
SC
learning revealed that networked learning had a positive effect on students’ engagement in terms of their perception (Woo et al., 2011; Hsu Wang & Comac, 2008) and increased students’
M AN U
motivation to learn (Shih, 2011). However, while focusing on general perceptions, it was not well studied regarding how students team and collaborate with peers to participate in the networked creation activity and further how such alignment and collaboration activities may impact student engagement.
TE D
Recently, social network analysis (SNA) (Freeman, 2004) has been widely used to uncover the structural patterns of the linkage between actors in social systems. SNA has been applied increasingly by educational researchers to understand relationship among students in
EP
learning communities in both conventional classrooms (Rienties, Héliot, & Jindal-Snape, 2013)
AC C
or networked learning settings such as MOOCs (Joksimović et al., 2016), blended learning (Lee & Bonk, 2016), online forums (López, Aliaño & Gómez, 2014) or blogs (Jimoyiannis & Angelaina, 2012). Moreover, SNA was found to be helpful to enhance our understanding of students’ engagement in the networked learning as it provides a mathematical manner to measure students’ engagement through their positions in a social network. However, the focus of the current studies was mainly on the social network analysis of networked learning in higher education settings. How elementary students team and collaborate with each other to participate 3
ACCEPTED MANUSCRIPT
in networked creation communities to create multimedia contents and how such social networking activities may impact their perceptual engagement are still not clear in the literature.
RI PT
In this vein, this study attempted to investigate how elementary students teamed and collaborated with peers to create multimedia stories in a networked creation community and analyzes their engagement from both social network analysis and perceptual engagement. At a
SC
macro level, this study investigated whether the free social networking activities, that is, the students have the chance to choose their collaborators to participate in the networked creation
M AN U
activities, would enhance their engagement. To achieve these goals, flow (Csikszentmihalyi, 2000) and strategic motivation framework (Pintrich, Smith, Garcia & McKeachie, 1991) were adopted in this study to obtain a clear picture of students’ perceptual engagement, as they provide sound theoretical frameworks to investigate engagement attributes and motivation. At the micro level, this study explored how elementary students team and collaborate with peer in
TE D
the networked creation activities. SNA was used to depict student social network which was analyzed in relation to students’ gender, knowledge proficiency in the creation domain and their perceptual engagement. By gathering data from 24 elementary students participating in a 20-
EP
week digital storytelling activity on a social network platform, this study aimed to achieve a
AC C
better understanding about elementary students’ engagement in the networked creation community.
2. Related works
2.1 Networked creation community
4
ACCEPTED MANUSCRIPT
The development of the social networking tools, such as wiki and blogs has made the Internet a networked creation community where individuals or groups could share self-created artifacts and interact with audiences. Researchers applied the social networking tools in various approaches to
RI PT
fulfill diverse pedagogical merits, such as reflective learning (Murray et al., 2007) and selfdirected learning (Liu et al., 2016). One of the approaches is to use the networked creation community as a self-presentation tool. In such a line of research, students learned on the
SC
networked learning communities by creating diverse artifacts, such as blogs (Asoodar et al., 2014; Sun, 2010; Ducate & Lomicka, 2008; Murray et al., 2007), voice creations (Hsu, et al.,
M AN U
2008), and personal stories (Ducate, Anderson, & Moreno, 2011) on an individual basis. The networked creation communities become a learning platform where students could publish, reflect upon, and rework on their creations. Another track of studies (e.g. Liu et al., 2016; Woo et al., 2011) incorporated group learning with the network creation activity in which a fixed group
English.
TE D
of students worked together to create artifacts on the networked creation community to learn
However, individuals’ engagement in a social network is a complex process influenced by
EP
multiple factors. Researchers investigated the process of students’ engagement in such a
AC C
community of practice from the perspectives of social cognition or social capital theories (Liu, Lin, Chang, & Chao, 2014). Social cognitive theory asserts that individuals’ cognition and behaviors are shaped under the influence of the social environment (Bandura, 1989). Subjective perceptions including the self-efficacy and the expected value of the participation in the social network significantly influence students’ engagement in the social network (Hsu, Ju, Yen, & Chang, 2007). In this regard, a great body of studies investigated students’ perceptual engagement in the networked creation activity. This track of studies focused on students’ 5
ACCEPTED MANUSCRIPT
subjective perceptions, such as perception of enjoyment (Ducate et al., 2011), usefulness (Terumi & Anderson, 2010), and satisfaction (Sun, 2010). These studies confirmed that at a general level students perceived a positive perception during the participation in the networked creation
RI PT
activity.
To obtain a thorough and deep understanding of students’ perceptual engagement in the creation
SC
activity, researchers also integrated motivation and flow theories to investigate students’ flow perceptions and motivations (Huang, Liu, Wang, Tsai & Lin, 2017; Liu et al., 2016). Their
M AN U
studies found that students’ perceptual engagement is a dynamic process that was significantly influenced by multiple factors, such as students’ knowledge proficiency of the creation domain and the grouping or pairing compositions. Based on social cognitive perspective, these studies investigated students’ perceptions of task value and self-efficacy to understand their perceptual engagement in the social network. Such a track of studies typically relied on data acquired
TE D
through self-reported questionnaires and interviews rather than how students took actions to participate in the activities (Lee & Bonk, 2016). However, these studies did not answer the question of how inter-personal relationships influences learning or perceptual engagement in the
EP
social networking activities.
AC C
To address the above limitation, another track of studies analyzed students’ engagement in social networks from the perspective of social capital theory. Social capital theory claims that the resources which an individual owns do not only exist in the personal properties he/she has, but also in the links through which he/she connects to others (Nahapiet & Ghoshal, 1998). Such an argument suggests that the inter-personal relationship plays an important role in their engagement in social works. The identities of students and the sense of trust among students in a social network, for instance, may significantly impact students’ actions to participate in the 6
ACCEPTED MANUSCRIPT
social network (Liu, Lin, Chang et al., 2014). In particular, the study by Huang et al. (2017) confirmed that the fixed relations during a networked creation activity impeded students’ long term engagement in the activity. However, how students teamed with peers to form pairs or
RI PT
groups in the learning community and how the inter-personal relationships might impact
perceptual engagement in a social network were not clearly addressed in the literature. It is thus worthwhile to investigate students’ social networking activities to better understand their
SC
engagement in the networked creation community.
M AN U
2.2 Social network analysis
Social network analysis (SNA) techniques were reported to be helpful to a better understanding about students’ engagement in networked creation communities. Educators believe that interactive relations among students impact the process and quality of learning (López et al.,
TE D
2014). Understanding the social network among students sheds light on the implementation of learning activities, and, therefore, there is an increasing trend in applying SNA in analyzing student engagement in networked learning activities (Rabbany, Takaffoli, & Zaïane, 2012). SNA
EP
provides both the methods and theory to display, discover and explain the structural patterns of social relationship among students (Giri, Manongga, & Iriani, 2014). It can be integrated with
AC C
quantitative data analysis to achieve a thorough understanding of the learning process (Lee & Bonk, 2016; López et al., 2014). Grounded in the graph theory (Gross & Yellen, 2005), SNA provides statistical methods
to analyze complex interrelationship among students. Several measures were applied to evaluate the social properties of the entire network or entities in the social network. For instance, network density, which displays the ratio of the actual number of ties to the maximum ties between
7
ACCEPTED MANUSCRIPT
entities in a network, can be influenced by pedagogical arrangements such as the teacher’s presence and activity stages (De Laat, Lally, Lipponen & Simons, 2007). Centrality is also a prominent indicator to show the activeness or the prestige of entities in a network as it indicates
RI PT
the extent to which an individual interacts with other members (Wasserman & Faust, 1994). Such an index may reflect a student’s social status in a network. For instance, the in-degree and out-degree centrality which represents the number of ties directed to and that initiated from an
SC
entity represents the prestige and influence of this entity in the network (Rabbany et al., 2012). A previous study found an inverse relationship between the in-degree centralization and the out-
association (López et al., 2014).
M AN U
degree centralization, meaning that students’ prestige and influence has a certain level of
Several studies have applied SNA to understand student interactions in various educational settings and identified several factors, which may be related with students’
TE D
engagement in the social network. For one, students’ gender plays an important role in the social network. The study by Giri et al. (2014) found that female students were more likely to be in the center of the social network than male students as they frequently gave and received messages
EP
while male students tended to play the recipient role. Furthermore, a close relation between
AC C
students’ knowledge proficiency levels and the social centrality was identified by several studies (Russo & Koesten, 2005; Cho, Gay, Davidson & Ingraffea, 2007; Gašević, Zouaq & Janzen, 2013). It was due to the fact that those students who had more sufficient domain knowledge might be more skillful to interact within the network, which in turn, brought benefits to these students. However, a different perspective on the social network suggested that social centrality did not always imply benefits nor represent any superior knowledge level (Krackhardt, 1999). Joksimović et al. (2016) found that the students who located at the positions between two 8
ACCEPTED MANUSCRIPT
strongly connected groups did not significantly benefit from their position. In other words, the relation between the social centrality and students’ knowledge level is not conclusive based on
RI PT
the literature. Furthermore, the findings above obtained from the observation of student interactions in learning environment where the interactions between them were to exchange ideas, not to create
SC
artifacts together. These findings may not be generalized to the context of the networked creation community. This is because the social networks in the above studies addressed the relationship
M AN U
of peer communication with relatively less information on co-work relationship during the creation process. The networked creation community involved dynamic pairing and grouping for learning and working together to achieve a shared goal. Thus, how students team up and collaborate with peers to participate in the networked learning activity require further
TE D
investigation.
Based on the above literature review, one of the goals of this study was to understand the impact of the social networking activities on elementary students’ engagement in the networked
EP
creation community. At a general level, this study intended to compare students’ engagement in two settings based on the flow and motivation theory. One of the settings afforded students to
AC C
freely team up with peers (dynamic teaming) and the other allowed them only to work with a fixed peer (fixed relation). Comparing students’ perceptual engagement in the two settings may shed light on the effect of free teaming in collaborative learning as perceptions and beliefs are influenced through the structural mechanisms of social networks (Jostad, Sibthorp & Paisley, 2013). At a detail level, this study analyzed students’ social network to examine what factors may affect students’ teaming and collaboration patterns. More specifically, students’ knowledge proficiencies and their genders would be analyzed in relation to their social network properties to 9
ACCEPTED MANUSCRIPT
uncover the role that gender and knowledge proficiencies may play in the social network. Last but not least, this study triangulated how students’ perceptual engagement in the social network may be related to their social network properties from both of the social cognitive and capital
RI PT
perspectives. More specifically, the inter-personal relationship and students’ perceptual
engagement were analyzed to uncover the relationship between the two constructs by using the flow, motivation questionnaires and SNA. In sum, this present intended to answer the following
SC
three research questions:
networked creation activity?
M AN U
RQ1: Do the dynamic teaming activities impact the students’ flow and motivation in the
RQ2: How do the students team and collaborate with peers to participate in the networked creation activity?
TE D
RQ3: Do the students’ properties in the social network have an impact on their flow and
3. Method
AC C
3.1 Participants
EP
motivation in the networked creation activity?
This study is part of a year-long program aiming at helping students develop their proficiency in English, computer use, and collaboration through collaborative digital storytelling. The participants were 26 third grade students in an elementary school in northern Taiwan, aged between 8 and 9. They participated in the networked creation activity to create multimedia stories collaboratively with peers. The students started learning English as foreign language from grade 1. Therefore, they were still in the initial stage of learning English with the focus on 10
ACCEPTED MANUSCRIPT
vocabulary and sentence patterns. Before the program, the students did not have prior experience in using computers to create multimedia stories with peers.
RI PT
3.2 Instruments 3.2.1 Storytelling and social networking
To provide participants the opportunity to create multimedia stories on the Internet, Story &
SC
Painting House (Liu et al., 2016), a social networking application on iPads connecting to an
M AN U
online platform, was developed to facilitate the process of the networked creation activity. Story & Painting House supports multimedia authoring functions, including easy-to-use drawing tools, text, and the ability/option to record oral narrations (See Figure 1).
The application also supports social networking facilities by which students could invite
TE D
peers to collaborate synchronously to create a shared multimedia story through iPads. The application depicts the social network showing the collaboration relationship between participants in the community (the up-right of Figure 1). Story & Painting House also allows
EP
participants to share their stories on the online platform. As shown in Figure 1, the platform
AC C
provides a repository to show the multimedia stories in progress and the published ones. ----------------------------------Insert Figure 1 here -----------------------------------
Moreover, participants could freely invite their peers to collaborate during the phase of the dynamic teaming activity. All teaming up activities were logged on the Story & Painting
11
ACCEPTED MANUSCRIPT
House platform. Once an invitation was sent, it would be instantly logged by the system and a link from inviter to invitee would be presented in the SNA tool for further analysis.
RI PT
3.2.2 The flow perception survey A flow perception survey was administrated to probe students’ perceptual engagement associated with the networked creation activity since engagement and flow shared several psychological
SC
attributes (Trevino & Webster, 1992). The survey has been extensively applied and showed adequate reliability in previous literature (Huang et al., 2017; Liu et al., 2016) indicating that this
M AN U
survey is a reliable instrument for assessing the students’ perceptual engagement. Guided by the Trevino and Webster’s flow framework, this survey probes student engagement in four basic flow components including the level of control, attention, curiosity and intrinsic interest that participants perceived when they participated in the networked creation activity. Thus, this
TE D
survey consists of only four 5-point Likert items (with 5 as the highest level and 1 as the lowest) and was administrated right at the end of each activity section where there was no interruptions to participants’ learning activity.
EP
3.2.3 Motivated Strategies for Learning Questionnaire (MSLQ)
AC C
While the flow survey was conducted to obtain students’ short term engagement right after each activity section, the Motivated Strategies for Learning Questionnaire (MSLQ) (Pintrich et al., 1991) was conducted to uncover the long-term impact of the networked creation activity on participants’ motivation associated with the fixed relation and dynamic teaming activities. The MSLQ was adapted according to the context of the activity. The questionnaire includes 25 items asking students’ motivation in the dimensions of self-efficacy, extrinsic/intrinsic goal orientation, task value and peer learning, as these components were directly related to the context of the 12
ACCEPTED MANUSCRIPT
study. The questionnaire was found to be adequately reliable as the Cronbach reliability (alpha) for the extrinsic/intrinsic goal orientation, self-efficacy, task value and peer learning are .68, .78,
RI PT
.81, .83 and .61 respectively. 3.3 Procedure
The networked creation activity was implemented during a semester, for an 80-minute session
SC
per week. Participants formed pairs to create multimedia stories using Story & Painting House. Each student was provided one iPad with an e-book containing 15 stories, which included texts
M AN U
with oral narrations so that students can learn the stories by themselves. Clear instruction was provided for each pair to participate in the activity: (1) Starting reading any story of their preference; (2) Working collaboratively using the storytelling application to retell the story in a multimedia picture book format, including drawings and oral narrations; and (3) Publishing and
TE D
sharing the picture books on the online platform where all participants could view the picture books. Upon finishing a story, they were instructed to continue working on re-telling and publishing the rest of the multimedia stories at their own pace.
EP
Participants experienced the networked creation activity in two different formats during
AC C
the semester. In the first 5 weeks, students were asked to participate in the activity without the dynamic teaming mechanism, hence, the fixed relation, meaning that they formed pairs by themselves and only collaborated with that partner for the multimedia stories activity. The homeroom teacher helped those who needed assistance finding a partner. Starting from week 6, students were instructed that they could freely team up with any peer to create new multimedia stories, hence in the dynamic teaming activity, which lasted for 8 weeks (week 6 to 13).
13
ACCEPTED MANUSCRIPT
The students went through a total of 13 sections. After each activity session, participants answered the flow perception survey. Therefore, 13 sets of flow survey results were obtained. Furthermore, the MSLQ was administered to assess participants’ motivation in week 5 and 13
RI PT
right after they completed the fixed relation and the dynamic teaming activity respectively. The participants’ feedback to the flow perception survey and MSLQ were analyzed to understand
SC
their perceptual engagement associated with the two networked creation activities. 3.4 Data analysis
M AN U
To answer the research question 1, participants’ flow perceptions and motivations associated with the fixed relation and dynamic teaming activities were compared. As the flow survey was administrated right after each activity section each week, participants’ flow perceptions in the first 5 weeks were averaged to obtain an overall flow perception about the fixed relation activity.
TE D
The same procedure was applied to their flow perceptions in dynamic teaming activity. Participants’ motivation and flow perception in the two settings were also analyzed with dependent t-tests.
EP
Participants’ invitation activities on the social network were analyzed with the UCINET
AC C
SNA tool (Borgatti, Everett & Freeman, 2002) to answer RQ2. The out- and in-degree centrality of each participant reflecting how an individual invited peers or received invitations from peers was computed to represent their participation in the social network. The degree centrality was analyzed with respect to genders and English proficiencies to understand how participants with different background team up with peers in the multimedia story creation activity. Regarding research question 3, the impact of participants’ participation in the social network (i.e. out-degree and in-degree centrality) on their perceptual engagement (i.e. flow 14
ACCEPTED MANUSCRIPT
perceptions and motivation) was analyzed. To achieve this goal, their perceptual engagement was also compared based on their degree centrality with t-test, thus, the comparative result can be shown from the integrative analysis of the social network and perceptual engagement that
RI PT
possibly reveals the impact of social network activity on learning experience. 4. Results
SC
4.1 Perceptual engagement in fixed relation and dynamic alignment activities
M AN U
Figure 2 shows participants’ weekly flow perceptions during the 13 weeks. It is shown that participants perceived relatively lower level of flow in the first 5 weeks when the fixed relation activity was implemented and the flow perception was at a higher level during week 6 – 13 when the dynamic teaming activity was administered. As shown in Table 1, the t-test results of participants’ average flow perception indicated that it was significantly associated with the two
TE D
different types of creation activities. When participating in the dynamic teaming activity, participants’ average flow perception was 4.52, which is significantly higher than that in fixed relation activity (4.12) (t=-3.22, p<.01). A closer analysis of the flow dimensions found that
EP
participants perceived a higher level of control (t=-3.46, p<.01), attention (t=-3.56, p<.01) and
AC C
curiosity (t=-2.26, p<.05) associated with the dynamic teaming activity than with the fixed relation activity.
----------------------------------Insert Figure 2 here --------------------------------------------------------------------Insert Table 1 here
15
ACCEPTED MANUSCRIPT
----------------------------------Table 2 lists the t-test results of the students’ motivation during the fixed relation and
RI PT
dynamic teaming activities. Participants perceived a significantly higher level of motivation in the dynamic teaming activity than they did in the fixed-relation activity in two dimensions: intrinsic motivation (t=-3.08, p <.01) and task value (t=-2.51, p<.05). It indicated that
SC
participants showed an enhanced perceptual engagement when they were able to dynamically select their own collaborating peers in the networked creation activity.
M AN U
----------------------------------Insert Table 2 here
----------------------------------4.2 Social network analysis
TE D
UCINET was applied to analyze the formation of participants’ invitation activities in the social network in relation to participants’ genders (F and M) and English proficiency levels (High, Middle and Low) (See Figure 3). The network involved 26 participants with 46 ties showing a
EP
network density of .07. By using the subgrouping analysis with factions, the social network
AC C
could be divided into two subgroups: the girls group (left) and the boys group (right). As shown in Figure 3 only 4 ties crosslinked the participants in the two different subgroups, which indicated that elementary students tended to team up with students of the same gender when participating in the networked creation activity. Figure 3 also showed the in-degree centrality of the nodes in the social network ranging from 0 to 7. Among the 12 students who received more invitations to form pairs for the networked creation activity and displayed higher in-degree centrality (centrality ≥2), 6 were girls 16
ACCEPTED MANUSCRIPT
and 6 boys (in the circle) suggesting that no particular gender contributed to higher popularity. It was also found that the 12 students showed diverse English proficiency levels, (4 at high-level, 5
influence students’ popularity during the creation activity. ----------------------------------Insert Figure 3 here
SC
-----------------------------------
RI PT
at mid-level, and 3 at low-level), indicating that the students’ content knowledge did not directly
M AN U
Furthermore, participants’ out-degree centrality was also analyzed to show their activeness while working on the activity. As seen in Figure 4, participants’ out-degree centrality ranged from 0 to 7 and 9 of them (located in the circle) showed higher out-degree centrality (centrality ≥2). These students actively teamed up with peers for the networked creation activity. It was found that participants’ English proficiency levels were related participants’ activeness in
TE D
their teaming up with peers. The 9 students in the high out-degree centrality were from the mid to low level proficiency groups (6 from low-level and 3 from mid-level), indicating that the
EP
lower-proficiency students were more active in teaming up with peers in the activity. However, genders were found not related because the high out-degree centrality group contained almost
AC C
equal number of students from the two genders, four girls and five boys. ----------------------------------Insert Figure 4 here -----------------------------------
17
ACCEPTED MANUSCRIPT
4.3 Impact of social network on perceptual engagement Participants’ perceptual engagement was analyzed in relation to their in- and out-degree
RI PT
centrality with t-test. Regarding the in-degree centrality, the 12 students who showed higher indegree centrality (centrality ≥2) perceived a significantly lower perceptual engagement than the other 14 who showed lower in-degree centrality (centrality <2). Table 3 displays the results of
SC
the t-test analysis of the two groups. Such significant difference was found in the level of overall flow perception (t=2.31, p<.05), control (t=2.87, p<.05), intrinsic (t=2.25, p<.05), extrinsic
M AN U
motivation (t=2.15, p<.05), and the task value (t=2.32; p<.05).
----------------------------------Insert Table 3 here
-----------------------------------
TE D
The out-degree centrality also had a significant influence on the perceptual engagement. As shown in Figure 4, the 9 students who showed higher out-degree centrality (centrality ≥2) perceived higher level of perceptual engagement than the other 17 students in the lower out-
EP
degree centrality (centrality <2) (See Table 4). It was shown that the high out-degree centrality group perceived significantly higher level of overall flow perception (t=3.25, p<.01), control
AC C
(t=2.55, p<.05), curiosity (t=3.71, p<.01) interest (t=2.37, p<.05) and significantly higher level of task value (t=2.52, p<.05).
----------------------------------Insert Table 4 here -----------------------------------
18
ACCEPTED MANUSCRIPT
5. Conclusions and discussion Examining students’ perceptual engagement and their participation in the social network for the
RI PT
13-week networked creation activities, this study found that the students demonstrated improved engagement as they could freely team up with peers to create multimedia stories. Both their flow perceptions and motivations increased as well. The social network analysis revealed that
SC
students’ prominent features including gender and proficiency levels of the target content had an impact on their participation in the social networking activities. Moreover, students’
M AN U
participatory features including in-degree and out-degree centrality in the social network showed a significant impact on their flow perceptions and motivations. The results of this study indicated that triangulating data from flow, motivation, and the social network were useful to help researchers understand student engagement in the networked creation activities from different
TE D
perspectives.
Many studies (Huang et al., 2017; Azmitia, 1988; Liu & Tsai, 2008; Rogoff, 1990) have confirmed pairing or grouping schemes impacted students’ engagement in collaborative learning
EP
activities. The study by Huang et al. (2017) further confirmed a student who had a long-term partnership with one fixed student had the tendency to cause uneven relationship between two
AC C
students in which one student played the guiding role to help the other student. The uneven roles largely decreased students’ engagement in the collaborative learning activity. Huang et al. suggested that a mechanism that breaks down such an uneven role distribution is necessary to restore students’ engagement. This study found that the students perceived an enhanced flow perception and motivation when they were able to dynamically team up with peers. With the opportunity for freely teaming up, students had full control during the creation process and avoided the risk of long-term uneven role. This explained why students perceived higher level of 19
ACCEPTED MANUSCRIPT
perceptual engagement than they did in the fixed relation activity. In other words, educators could take advantage of the dynamic pairing or grouping mechanism to sustain students’
RI PT
engagement in other networked learning activities. The SNA of this study indicated that no particular gender showed higher popularity (indegree centrality) and activeness (out-degree centrality) in the social network and boys and girls
SC
mainly interacted with peers of the same gender. Such social network structure is different from what Giri et al. (2014) identified where female students were reported to more likely be the
M AN U
center of social networks and more actively participated in the social network than male students. Several factors may explain the different findings between this study and Giri et al. (2014). For instance, students of this study were elementary students while participants in Giri et al. (2014) study were university students. Furthermore, the features of the social network activities of the two studies were also significantly different. While the students of this study had to work closely
TE D
with peers, the students in Giri et al. (2014) study solely exchanged ideas on Internet without the need to work with peers. In other words, the findings of the present study contributed to draw attention of educators and researchers to take age and the types of social networking activities
EP
into account while analyzing social networks in other educational contexts.
AC C
This study found that students’ proficiency level in the target content was not related to
their popularity (in-degree centrality) when working on the activities. In other words, students’ content knowledge level required in the activity did not directly influence their popularity in the network. However, the finding of low proficiency level students showing higher out-degree centrality in the social network revealed that the low proficiency level students were more active in teaming up with peers than the high proficient students, which conflicted the results of previous studies (Cho et al., 2007; Gašević et al., 2013; Russo & Koesten, 2005) where students 20
ACCEPTED MANUSCRIPT
of higher knowledge level occupied the central positions in social networks. This may be due to the particular nature of the social networking activity where students had to form pairs to create shared multimedia stories. In other words, students of low proficiency level needed to actively
RI PT
seek help from others so that they could accomplish the shared goal with peers, which is
different from the information sharing activities in previous studies, such as online discussion
SC
where students were not bound to achieve a shared goal.
The analyses on students’ flow perception, motivations and the social network revealed
M AN U
an interesting relationship between students’ perceptual engagement and their social networks. The students’ out-degree centrality had a positive impact on students’ flow perception and motivation. In other words, the students who occupied the central position in the out-degree centrality perceived higher level perceptual engagement. However, the high in-degree centrality had a negative impact on flow perceptions and motivation. The students who occupied the
TE D
central positions in the in-degree centrality perceived lower level perceptual engagement. As the in-degree centrality reflected an entity’s popularity (Rabbany et al., 2012), the finding suggested that popularity did not imply high level of engagement in the networked learning activity. On the
EP
contrary, the out-degree centrality may appropriately represent the students’ engagement in the
AC C
activity as it had a positive relationship with students’ flow and motivation. Educators could use the two indicators to monitor students’ engagement in other types of networked learning activities.
The results of this study contributed to a sound understanding of students’ engagement in
networked creation community and suggested that free pairing strategy among students had a positive impact on their engagement. It was concluded that knowledge level had an impact on students’ engagement in the social network. Lower-proficiency students were more active in 21
ACCEPTED MANUSCRIPT
teaming up with peers in the activity. Furthermore, the individual student’s positions in the social network had a significant impact on their flow and motivation in the networked creation activity. However, this study is an exploratory study where students’ engagement was not compared with
RI PT
that of a control group. Further comparative investigation with experiments is still needed with a wider sample to provide additional evidences. Furthermore, it would be interesting to see how the patterns of students’ social network may change over time. The timeline analysis of social
SC
networks may uncover how the interactions among students may evolve in a long-term
networked learning activity. It would also be worthwhile to extend the investigation to the other
M AN U
types of social networking activities, such as knowledge construction through online discussion. Gathering information on these issues through further studies can help obtain a more thorough understanding of student engagement in a networked learning community, which would help
AC C
EP
TE D
educators and teachers implement effective networked learning activities.
22
ACCEPTED MANUSCRIPT
References Asoodar, M., Atai, M. R., Vaezi, S. & Marandi, S. S. (2014). Examining effectiveness of
RI PT
communities of practice in online English for academic purposes (EAP) assessment in virtual classes. Computers & Education, 70, 291-300.
Azmitia, M. (1988). Peer interaction and problem solving: When are two heads better than one?
SC
Child Development, 59, 87–96.
M AN U
Bandura, A. (1989). Social cognitive theory. In R. Vasta (Ed.), Annals of Child Development (pp. 1-60). Greenwich, CT: Jai Press LTD.
Bergquist, M. & Ljungberg, J. (2001). The power of gifts: organizing social relationships in open
TE D
source communities. Information Systems Journal, 11(4), 305-320. Borgatti, S.P., Everett, M.G., & Freeman, L.C. (2002). Ucinet for Windows: Software for Social Network Analysis. Harvard, MA: Analytic Technologies.
EP
Cho, H., Gay, G. , Davidson, B., & Ingraffea, A. (2007). Social networks, communication styles,
AC C
and learning performance in a CSCL community, Computers & Education, 49(2), 309–329. Csikszentmihalyi, M. (2000). Beyond boredom and anxiety. Jossey-Bass. De Laat, M., Lally, V., Lipponen, L., & Simons, R.-J. (2007). Investigating patterns of interaction in networked learning and computer-supported collaborative learning: a role for Social Network Analysis. International Journal of Computer-Supported Collaborative Learning, 2(1), 87-103.
23
ACCEPTED MANUSCRIPT
Ducate, L. C., Anderson, L. L. & Moreno, N. (2011) Wading through the world of wikis: An analysis of three wiki projects. Foreign Language Annals, 44(3): 495-524.
RI PT
Ducate, L. C. & Lomicka, L. L. (2008). Adventures in the blogosphere: From blog readers to blog writers. Computer Assisted Language Learning, 21(1), 9-28.
Freeman, L. C. (2004). The development of social network analysis: a study in the sociology of
SC
science. Vancouver, B. C.: Empirical Press.
M AN U
Gašević, D., Zouaq, A., & Janzen R. (2013). Choose your classmates, your GPA is at stake!: the association of cross-class social ties and academic performance, American Behavioral Scientist, 57, 1460-1479.
Gross, J. L., & Yellen, J. (2005). Graph theory and its applications. CRC press.
TE D
Giri, B. E., Manongga, D. & Iriani, A. (2014). Using Social Networking Analysis (SNA) to Analyze Collaboration between Students (Case Study: Students of Open University in
EP
Kupang), International Journal of Computer Applications (0975 – 8887), 85(1), 44 – 49. Hsu, M. H., Ju, T. L., Yen, C. H., & Chang, C. M. (2007). Knowledge sharing behavior in virtual
AC C
communities: The relationship between trust, self-efficacy, and outcome expectations. International Journal of Human-Computer Studies, 65(2), 153-169. Hsu, H.-Y., Wang, S.-K. & Comac, L. (2008). Using audioblogs to assist English-language learning: An investigation into student perception. Computer Assisted Language Learning, 21(2), 181-198.
24
ACCEPTED MANUSCRIPT
Huang, Y. Y., Liu, C. C., Wang, Y., Tsai, C. C. & Lin H. M. (2017). Student engagement in long-term collaborative EFL storytelling activities: An analysis of learners with English
RI PT
proficiency differences. Educational Technology & Society, 20 (3), 95–109. Jimoyiannis, A. and Angelaina, S. (2012). Towards an analysis framework for investigating students’ engagement and learning in educational blogs. Journal of Computer Assisted
SC
Learning, 28(3), 222-234.
Joksimović, S., Manataki, A., Gašević, D., Dawson, S., Kovanović, V., & de Kereki, I. F. (2016).
M AN U
Translating Network Position into Performance: Importance of Centrality in Different Network Configurations. In Proceedings of the Sixth International Conference on Learning Analytics & Knowledge (pp. 314–323). New York, NY, USA: ACM. Jostad J., Sibthorp, J. & Paisley, K. (2013). Understanding groups in outdoor adventure
17-31, 2013.
TE D
education through social network analysis. Australian Journal of Outdoor Education, 17(1),
EP
Krackhardt, D. (1999). The ties that torture: Simmelian tie analysis in organizations. Research in the Sociology of Organizations, 16, 83-210.
AC C
Lee, J. & Bonk, J. C. (2016). Social network analysis of peer relationships and online interactions in a blended class using blogs. Internet and Higher Education, 28, 35-44. Liu, C. C., Lin, C. C., Chang, C. Y., & Chao, P. Y. (2014). Knowledge Sharing among University Students Facilitated with a Creative Commons Licensing Mechanism: A Case Study in a Programming Course. Journal of Educational Technology & Society, 17(3), 154-167. 25
ACCEPTED MANUSCRIPT
Liu, C.-C., Lin, C.-C., Deng, G.-Y., & Tsai, C.-C. (2014). Online knowledge sharing experience with Creative Commons. Online Information Review, 38(5), 680-696.
RI PT
Liu, C. C. & Tsai, C. C. (2008). An analysis of peer interaction patterns as discoursed by on-line small group problem-solving activity. Computers & Education, 50(3), 627-639.
Liu, C. C., Wang, P. C., & Tai, D. S. J. (2016). An analysis of student engagement patterns in
SC
language learning facilitated by Web 2.0 technologies. ReCALL, 28(02), 104-122.
M AN U
López, P. M., Aliaño, A. M. & Gómez, J. I. A. (2014). Social network analysis of a blended learning experience in higher education, Research on Education and Media, 6(2), 69-78. McLoughlin, C., & Lee, M. J. W. (2007). Social software and participatory learning: pedagogical choices with technology affordances in the Web 2.0 era. Paper presented at the Ascilite,
TE D
Singapore.
Murray, L., Hourigan, T. & Jeanneau, C. (2007). Blog writing integration for academic language
EP
learning purposes: towards an assessment framework. IBÉRICA, 14, 9-32. Nahapiet, J., & Ghoshal, S. (1998). Social capital, intellectual capital, and the organizational
AC C
advantage. The Academy of Management Review, 23(2), 242-266. Pintrich, P. R., Smith, D., Garcia, T. & McKeachie, W. (1991). A manual for the use of the motivated strategies for learning questionnaire (MSLQ). Ann Arbor Michigan: The University of Michigan.
26
ACCEPTED MANUSCRIPT
Rabbany, R., Takaffoli, M., & Zaïane, O. R. (2012). Social network analysis and mining to support the assessment of on-line student participation. ACM SIGKDD Explorations
RI PT
Newsletter, 13(2), 20–29. Rienties, B., Héliot, Y., & Jindal-Snape, D. (2013). Understanding social learning relations of international students in a large classroom using social network analysis. Higher
SC
Education,66(4), 489–504.
Rivens Mompean, A. (2010). The development of meaningful interactions on a blog used for the
M AN U
learning of English as a Foreign Language. ReCALL, 22(3), 376-395.
Rogoff, B. (1990). Apprenticeship in Thinking: Cognitive Development in Social Context. Oxford: Oxford University Press.
TE D
Russo, T. C. & Koesten, J. (2005). Prestige, centrality, and learning: A social network analysis of an online class, Communication Education, 54(3), 254–261. Shih, R. C. (2011). Can Web 2.0 technology assist college students in learning English writing?
EP
Integrating Facebook and peer assessment with blended learning. Australasian Journal of
AC C
Educational Technology, 27(5), 829-845. Sun, Y. C. (2010) Extensive writing in foreign‐language classrooms: A blogging approach. Innovations in Education and Teaching International, 47(3), 327-339. Terumi, M. T. & Anderson, T. (2010). Learning outcomes and students’ perceptions of online writing: Simultaneous implementation of a forum, blog, and wiki in an EFL blended learning setting. System, 38(2): 185-199.
27
ACCEPTED MANUSCRIPT
Trevino, L. K. & Webster, J. (1992) Flow in computer-mediated communication: electronic mail and voice mail evaluation and impacts. Communication Research, 19(5), 539-573.
RI PT
Wasserman, S., & Faust, K. (1994). Social network analysis: Methods and applications. Cambridge university press.
Woo, M., Chu, S., Ho, A. & Li, X. (2011). Using a wiki to scaffold primary-school students'
AC C
EP
TE D
M AN U
SC
collaborative writing. Educational Technology and Society, 14(1): 43-54.
28
ACCEPTED MANUSCRIPT Acknowledgements This research was partially funded by the Ministry of Science and Technology under
AC C
EP
TE D
M AN U
SC
RI PT
contract number 103-2511-S-008-014-MY3.
ACCEPTED MANUSCRIPT List of Tables Table 1. The t-test results of the students’ flow perceptions during the two networked creation activities (N=26) Table 2. The t-test results of the students’ motivation during the two networked learning activities (N=26)
AC C
EP
TE D
M AN U
SC
RI PT
Table 3. The t-test results of flow perceptions and motivation for different levels of in-degree centrality Table 4. The t-test results of flow perceptions and motivation for different levels of out-degree centrality
ACCEPTED MANUSCRIPT Table 1. The t-test results of the students’ flow perceptions during the two networked creation activities (N=26)
Curiosity Interest
t
p
Fixed Dynamic Fixed
4.12 4.52 4.28
.67 .43 .64
-3.22**
<.01
Dynamic Fixed Dynamic
4.63 3.82 4.31
.49 .94 .63
Fixed Dynamic Fixed Dynamic
4.13 4.52 4.26 4.60
.85 .55 .86 .43
AC C
EP
TE D
**p<.01；*p<.05
-3.46** -3.56** -2.26*
RI PT
Attention
SD
<.01 <.01
.03
SC
Control
M
-2.00
M AN U
Overall
Period
.06
ACCEPTED MANUSCRIPT Table 2. The t-test results of the students’ motivation during the two networked learning activities (N=26) M
SD
T
p
Intrinsic
Fixed Dynamic
4.14 4.53
.83 .63
-3.08**
<.01
Extrinsic
Fixed Dynamic
3.91 3.83
.87 .92
.46
.65
Task value
Fixed
4.16
.83
-2.51*
Dynamic Fixed
4.56 3.59
.63 1.02
Dynamic
3.97
1.01
Fixed Dynamic
3.95 4.28
.70 .81
AC C
EP
TE D
**p<.01；*p<.05
.02
.19
SC
Self-efficacy
-1.36
-1.96
M AN U
Peer learning
RI PT
Period
.06
ACCEPTED MANUSCRIPT Table 3. The t-test results of flow perceptions and motivation for different levels of in-degree centrality Centrality
N
M
SD
t
p
High
12
4.32
.47
2.31*
.03
Low
14
4.68
High
12
4.35
Low
14
4.87
High
12
Low
14
Control
Curiosity
4.06
.73
4.53
.46
12
4.37
.56
Low
14
4.66
.52
High
12
4.51
.45
Low
14
4.68
.40
High
12
4.25
.64
Low
14
4.77
.53
Extrinsic
High
12
3.42
1.01
Low
14
4.16
.69
High
12
4.26
.80
Low
14
4.82
.26
High
12
3.64
.74
Low
14
4.26
1.13
High
12
4.11
.70
Low
14
4.42
.89
TE D
Intrinsic
AC C
Task Value
Peer learning
Self-efficacy
**p<.01；*p<.05
2.87*
.01
2.01
.06
1.34
.19
1.04
.31
2.25*
.03
2.15*
.04
2.32*
.04
1.63
.12
.96
.34
.18
High
Interest
Motivation
.59
EP
Attention
.31
SC
Overall
M AN U
Flow
RI PT
(in-degree)
ACCEPTED MANUSCRIPT Table 4. The t-test results of flow perceptions and motivation for different levels of out-degree centrality Dimension
Centrality
N
M
SD
High
9
4.78 .18
Low
17
4.38 .46
High
9
4.88 .18
Low
17
4.50 .55
High
9
Low
17
t
p
(out-degree)
Attention
Curiosity
Interest
Intrinsic
EP
Extrinsic
AC C
Task Value
Peer learning
Self-efficacy
**p<.01；*p<.05
4.57 .51
2.55** .02
1.54
.14
4.18 .66
High
9
4.89 .13
Low
17
4.33 .59
High
9
4.81 .18
Low
17
4.50 .48
High
9
4.72 .65
Low
17
4.43 .61
High
9
4.16 .67
Low
17
3.65 .99
High
9
4.87 .18
Low
17
4.40 .72
High
9
4.11 1.34 .50
Low
17
3.90 .81
High
9
4.26 1.08 .07
Low
17
4.27 .66
TE D
Motivation
3.25** <.01
RI PT
Control
SC
Overall
M AN U
Flow
3.71** <.01
2.37*
.03
1.15
.26
1.40
.17
2.52*
.02
.62
.95
ACCEPTED MANUSCRIPT List of Figures Figure 1. The social network and the online repository of multimedia stories supported by Story & Painting House Figure 2. The weekly flow perceptions of the students during the 13 activity weeks Figure 3. The social network with the high in-degree centrality at the center
AC C
EP
TE D
M AN U
SC
RI PT
Figure 4. The social network with the high out-degree centrality at the center
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
Figure 1. The social network and the online repository of multimedia stories supported by Story & Painting House
ACCEPTED MANUSCRIPT 4.8
4.6
RI PT
4.4
4.2
SC
4
3.6 1
2
3
4
5
M AN U
3.8
6
7
8
9
10
11
12
13
AC C
EP
TE D
Figure 2. The weekly flow perceptions of the students during the 13 activity weeks
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Figure 3. The social network with the high in-degree centrality at the center
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Figure 4. The social network with the high out-degree centrality at the center
ACCEPTED MANUSCRIPT Highlights Networked creation activities had a positive effect on students’ perceptual engagement. This study analyzed students’ engagement in social network to learn through creation. Dynamic pairing or grouping mechanism is helpful to sustain students’
AC C
EP
TE D
M AN U
SC
RI PT
engagement. Knowledge level had an impact on students’ position in the social network. In- and out-degree centrality showed significant influence on perceptual engagement.