Analysis of peer interaction in learning activities with personal handhelds and shared displays.
The ratio of computers to students in classrooms is rising steeply as the price of handheld devices such as PDAs, laptop computers, and Tablet PCs decreases. The ratio of one computer per student is increasingly becoming a reality in some schools, creating the one-to-one (1:1) computing environment (http://www.g1to1.org/). In classrooms equipped with wireless networks, technology is beginning to enrich conventional classrooms and enhance the learning and teaching practices within them. Handheld devices have been utilized to support lively learning, create rich learning scenarios in such technology-enriched classrooms, and encourage socially promotive interaction (Roschelle, 2003; Zurita & Nussbaum, 2004; Liu et al., 2006). However, handheld devices are designed for personal applications, rather than collaborative scenarios. Whether handheld devices facilitate or impede face-to-face social interaction is an important research issue (Hwang, Tsai & Yang, 2008).
Creating a situation and space that enables students to interact and learn is the main concern of design-based research (Collins, 1992). Design-based research is a series of approaches with the intent to produce new theories, artifacts, and practices that account for, and potentially impact, learning and teaching in naturalistic settings (Barab & Squire, 2004). Even though there is always a need to construct an environment of interaction, design-based research investigates how students learn in their learning environment and revises and optimizes the design of the environment accordingly (Barab et al., 2001). Recently, most studies focus on learning scenarios in which students learn and interact with peers solely through handheld devices. Although the resultant information is necessary to the construction of a space for students to learn in a 1:1 environment, few studies have as yet investigated how students interact with each other when provided with a combination of handheld and peripheral devices. There remains, therefore, a need to understand the situation in which collaborative learning can be effectively implemented in a 1:1 computational environment.
Peer discussion and interaction have been found to facilitate collaborative learning. Collaborative problem solving has gained significant attention among educators for improving student learning (Dillenbourg & Traum, 2006; Liu & Kao, 2007). Instead of passively receiving knowledge from teachers alone, students can engage in problem solving and knowledge construction activities during peer discussion and interaction (Blumenfeld, Marx, Soloway, & Krajcik, 1996). However, 1:1 learning scenarios sometimes overemphasize the affordance of handheld devices to facilitate face-to-face collaboration, particularly if the scenarios disregard the fact that cognition may be distributed among various artifacts including large displays, whiteboards, printers, and learners in the classroom (Hutchins, 1995). Therefore, the analysis of peer interaction plays a significant role in the understanding of the critical learning process and the effectiveness of peer collaboration in a 1:1 computing environment.
Collaborative learning activities encourage members to share the learning experience, gather learning resources to help others, and share learning achievements. Extensive collaborative learning systems have been developed on computers and the Internet to enhance learning (Suthers, 2005, Liu & Tsai, 2006). Despite the significant advancement of online collaboration, educators still emphasize the importance of face-to-face collaboration, because promotive interaction is a critical factor in the success of cooperative learning (Johnson & Johnson, 1994). Additionally, face-to-face interaction is the most common interaction style in classrooms. The interaction of students with their peers through different technological devices, and the influence of these devices on interaction among students, have become important research issues as technology gradually comes into classrooms. Researchers have analyzed conversations between students in online collaborative learning scenarios to investigate how students interact in online discussion forums (Scardamalia, 2004; Liu & Tsai, 2006). Nevertheless, face-to-face collaboration has many interaction cues, such as visual focus and hand-pointing behaviors, which are not visible in online peer interactions. Therefore, the development of an appropriate methodology is necessary to reflect and explore correctly the effects and atmosphere of face-to-face peer interaction. Consequently, this study attempts to investigate the peer interactions involved in face-to-face collaborative learning with handheld devices.
The effects of human activities may depend on the designs of the technological devices used and on the settings in which these devices are applied. For example, when students are performing collaborative activities with individual tabletop computers, the lack of shared displays may lead to a loss of eye-contact and an unawareness of visual focus (Scott et al., 2003). The screens of handheld devices may be counterproductive to the promotion of interaction among groups of learners, thereby causing difficulty in establishing effective peer interaction during collaborative activities. Meanwhile, display technology has significantly changed human application of information technologies in recent years. Liquid Crystal Displays (LCD) are becoming increasingly cheap and popular peripheral devices. Institutes and organizations increasingly set up LCD devices in public areas. They are increasingly important peripheral devices, working with personal devices to support group work and learning. The LCD becomes a shared display to augment group learning in the 1:1 computing environment. Therefore, this study examines how students interact with handhelds and how shared displays in classrooms affect peer interaction in face-to-face collaborative learning.
One salient feature of social collaboration and presence is intimacy, which depends significantly upon non-verbal cues such as eye contact and smiling (Short et al., 1976). A mutual gaze moderates interpersonal distance and the sense of intimacy (Argyle & Dean, 1965). In addition, gesticulated interaction frequently takes place along with verbal utterances in meaningful processes, resulting in meaning creation (Klerfelt, 2007). Scott et al. (2003) analyzed the non-verbal interpersonal interactions of group members during discussion, in order to assess whether the current collaborative learning environment setting is beneficial to discussion. This study proposes an analytical scheme to investigate how environments influence interpersonal collaborative interaction in terms of participation, non-verbal social cues, visual focus, and a conversation log. The following research issues, based on interaction analysis, are considered:
* Research question 1: How the process of participation and negotiation differs when the collaborative activity is supported by a 1:1 computing environment and shared displays?
* Research question 2: How social interaction differs when collaborative activity is supported by a 1:1 computing environment and shared displays?
* Research question 3: How the argument process differs when collaborative activity is supported by a 1:1 computing environment and shared displays?
One-to-one and shared display groupware
The portability and communication capabilities of handheld devices enhance classroom dynamics and promote face-to-face interactions (Zurita & Nussbaum, 2004; Liu & Kao, 2007). Students engaged in collaborative problem solving activities need both personal workspaces for doing their own work and a public workspace, where they can share personal work and discuss group work together. Many collaborative learning models, such as Think-Pair-Share (Lyman, 1981) and Jigsaw (Aronson, 1975), can adopt personal and public workspaces to enforce personal accountability and shared group goals respectively (Lai & Wu, 2006). However, handheld devices are designed for individual-user mobile applications, rather than for collaborative applications, and may therefore restrict face-to-face collaborative learning interaction. Due to the small screen size of handheld devices, students intuitively interact most frequently with the peers located adjacent to them, while neglecting peers sitting further away. The lack of public workspace may impede communication among members and lead to tete-a-tete or fragmented communication patterns (Milson, 1973).
To investigate how shared displays affect collaboration socially, this study provided a technology-enriched collaborative classroom (Figure 1) with shared display groupware, including six LCD displays, as a public workspace, while students performed their personal learning tasks using handheld devices. All displays in a classroom were connected to a shared display groupware system with which students could edit documents, search the Internet for learning materials, and send their ideas, as well as upload documents onto the shared display groupware. The groupware also allowed students to project their handheld screens onto the shared display via a wireless network (Figure 2).
The shared display groupware for each of the LCD displays in the classroom was based on client-server architecture and was designed in three layers: the network layer, the individual workspace layer, and the coordination and presentation layer (Figure 2). The network layer enabled students to connect the shared display groupware and individual handheld devices through the TCP/IP and wireless network. The individual workspace layer provided an interface on the student handheld devices (client) for editing their documents and searching for learning materials from the Internet and also allowed students to send their ideas and upload documents onto the shared display groupware (server). The coordination and presentation layer (Figure 2) provided those functions necessary for students to interact with the shared display server. With the functionality provided by the coordination and presentation layer, students could project their handheld screens onto the shared display via a wireless network, enabling them to share ideas and documents on the shared display instead of the small screen of the handheld devices.
Buxton's "less is more" design principle (Buxton, 2001) asserts that the key issue of design is to focus less on technology and engineering and far more on the human element. Adding technological functionality to the tools or channels between humans does not always add value to those tools for the mediated people. For example, in Buxton's article, the Super Appliances 5-in-1 power tool and Swiss Army Knife that integrated many functions in a single tool did not always guarantee a value-adding effect on these tools (Buxton, 2001). Thus, there is a need to examine the human reactions when shared displays are added to 1:1 computational environments.
Most studies have confirmed that the affordance of individual devices in learning scenarios, clickers and PDAs for example, do improve interaction in the classroom (Roschelle, 2003; Zurita & Nussbaum, 2004). Adding shared displays to collaborative learning scenarios, however, could possibly cause problems with distraction, because students would then have to switch between different workspaces and screens. The shared displays could also possibly result in problems with competition for control among students, because there is only one input device (i.e. one mouse and keyboard) for each shared display. Therefore, shared displays do not inevitably have a beneficial effect on collaboration. There is clearly a need to examine students' reactions to the integration of handheld and peripheral devices.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
The environment necessary for mobile learning research generally requires a significant equipment investment, such as Tablet-PCs, PDAs, and peripheral devices. It is difficult to conduct studies on large numbers of students. Many mobile learning studies (Zurita & Nussbaum, 2004, Liu & Chou et al., 2006) follow the design-based research approach (Collins, 1992). Because this approach involves rich, collaborative social interaction in a real world, not all variables of interest are well known in advance. For this reason, design-based research emphasizes finding and identifying specific phenomena that improve the learning environment and student practice, rather than just reporting the learning outcomes of controlled experiments. This type of study investigates learning practices deeply, revealing an advisable direction of further research. It focuses on the explanations for certain phenomena and closely examines communication and collaboration patterns. The intent is to determine why and how certain behaviors happen under specific conditions and to generalize these findings to apply to more loosely related cases.
Participants and the course
The participants were fifteen graduate students enrolled in the course "Statistics and Data Mining Techniques," at National Central University (Taiwan). The class was composed of weekly 3-hour classes and was focused on statistical analysis and data mining. During each class, the teacher first outlined each week's learning content. A student then presented the learning content related to the topics assigned. Students then began to solve the problems assigned by the teacher collaboratively after the student presentations. This study assessed peer interaction as influenced by the use of different equipment in the classroom. Thus, the experiments were carried out in two different environmental settings, namely 1:1 and Shared-Display. In the 1:1 setting, students used only the Tablet PC for both individual learning tasks and collaborative learning activities in the classroom. In the Shared-Display setting, students could utilize shared display groupware as well as handheld devices. The entire experiment lasted eight weeks.
Collaborative problem solving activity
The students were divided into 3 groups of five. In practice, the number of students participating in the collaborative learning activities varied between three and five, because some students did not attend some of the classes during the experiment. The teacher presented problems, which the students had to collaborate to solve. To enforce personal accountability, students had to solve the given problems by themselves before discussing them with their peers. Students had appropriate resources at their disposal in the classroom for solving the problems, such as statistical tables, calculators, and Microsoft Excel. Along with the resources available in the classroom, students also had to search Internet resources, such as statistical tools, to solve complicated problems. Group members then conferred with each other to organize a group solution, each had submitted a personal solution to the given problem. The interaction between group members and the process of discussion was observed in order to gain an understanding of how they interacted with the aid of handheld devices.
Argument process analysis
During collaborative learning activities, students have to express and discuss divergent ideas in order to construct and share knowledge collaboratively. In a face-to-face collaborative learning scenario, all these communications occur in the conversation among group members, because each member is required to externalize his/her ideas. Conversation is an important tool for knowledge mediation within a culture (Saljo, 2000). Conversation, negotiation, and the sharing of perspectives are carried out to build knowledge (Suthers, 2006; Stahl, 2005). A graphical representation, based on chat log analysis framework (Stahl, 2005), was adopted for this study to identify the role of uptake within online discussion forums. However, when students are solving complex problems collaboratively, they rely on a domain-specific argumentative model to solve problems within a domain. For example, design-based problems involve issue generation and position expression to solve complex problems. "Uptake" is therefore, is too general a term to describe the structure of discussion for the specific domains.
This study applied IBIS (Kunz & Rittel, 1970) as a model to analyze collaborative problem solving activity, rather than as a structure for discussion. In face-to-face interaction, there were rich physical gestures and facial cues during group discussion, which cannot be expressed in online communication. IBIS involves only task-oriented information (i.e. issue, argument, and position). It does not involve the social aspect of activities, such as the group development process and social cues. Therefore, this study proposed a methodology that combines conversation and social cues to explore face-to-face group discussion. Utterances within student conversations were sorted into the following categories based on their purpose in the problem solving process (Liu & Tsai, 2006):
* Issues: What needs to be done and what questions need to be answered. Issues relate to the concepts and skills being learned by students. For instance, students may seek peer support in solving a learning problem by asking questions such as "How about your decision on fe (formula)?", or "How did you solve Question 1? "
* Positions: Methodologies for resolving an issue or a question. Positions constitute answers from peers in response to issues that have been raised. A student may help others by responding to issues with comments such as "The answer is normally written like this", or "Your answer to 5.56 is unlikely to be correct."
* Arguments: Opinions that support or oppose a position. For instance, students may comment on the positions of others with statements such as, "So, let me explain it by giving an example: would the average scores of third-grade students in one junior high school be lower than the national scores?", or "You can compare it with the second question, which is solved by the second method."
* Group development: Suggestions for the progress of collaborative activities. For example, "I've found an example for solving the two-sample question", or "We should apply the example in Question 1."
Social interaction analysis
Non-verbal interpersonal behaviors facilitate interaction among group members in face-to-face collaborative learning (Short et al., 1976; Scott et al, 2003). Group members in this study did not restrict themselves to interaction through handheld devices. Instead, they could point at the shared display, make eye contact with each other to improve discussion, and watch each other's responses. This study analyzed hand-pointing and visual focus behaviors to confirm how environments influence interpersonal interaction during collaboration. The video of each group was recorded individually and analyzed by two independent observers.
Analysis of participation and negotiation
To integrate non-verbal interaction with conversational analysis, this study codified the events that occurred during the collaborative problem solving activities in a group dynamic chart (Figure 3 and 4). The heading denotes the group members A, B, C, D, E, respectively. There are two columns representing conversational and non-verbal interactions. The left column represents the conversational interactions, the right column the non-verbal interactions. The Y axis represents the chronological order of the conversational utterances. Conversational interactions are depicted in the conversational interaction chart (the left sequence of each chart in Figure 3 and Figure 4). Each number denotes an utterance generated by the group members. Each dashed line in the chart denotes a response from a certain student (or the student's) to another student's conversational utterance. For example, student D provided position utterance 2 in response to student A's group development utterance 1 in Figure 3.
In addition, the videos recorded of collaborative problem solving activity were analyzed to identify precisely how students interacted with the handhelds and with other students. Non-verbal cues demonstrating student visual focuses were identified using the videos. Non-verbal interactions are depicted in the non-verbal interaction chart (the right sequence of each chart in Figure 3 and 4). These non-verbal cues included (1) watching personal handheld devices, (2) pointing at personal handhelds, (3) watching the shared display, and (4) pointing by hand at the shared display. The visual focus cues are depicted on the right of the conversational interaction chart. For example, Figure 3 illustrates student B watching student A's handheld device while student D was responding to student A's group development statement with utterance 5. Students responded to other students' issues, positions of discussion, arguments, and group development, and even to their responses, thus forming discussion threads. A new discussion thread was started when a student presented a new group development statement, position, or argument that was not related to previous discussion threads. The following interaction items were measured to investigate student interactions in discussion threads.
* Discussion threads: A discussion thread is a set of connected utterances made by students. For instance, utterances 1 and 2 in Figure 3 constitute a short thread, while utterances 3-21 form a long thread.
* Thread depth: The thread depth is the number of utterances students gave in a thread. The depth of the thread constituted by utterances 1 and 2 in Figure 3 is 2, and utterances 3-21 form a thread with a depth of 19.
* Shared visual focus: Students participating in a thread established shared visual focus by watching/sharing the screens of personal handhelds and by watching the shared display together. For example, students A and B established shared visual focus on student A's handheld screen, because student B was watching student A's screen while student D was responding to student A with group development statement 5 in Figure 3. Another example is the establishment of shared visual focus on the shared display by all students while student A argued his opinion in argument 23 in Figure 4 with his hand pointing to the shared display.
* Participants in threads: This study examines student participation in discussion threads. A student was considered to be actively participating in a discussion thread when he presented opinions/responses or paid attention to other students proposing opinions or responses within the thread by having a shared visual focus with them. For instance, all students participated in the first discussion thread in Figure 4 (utterances 1-22) because all of them proposed opinions/responses. These students also actively participated in the second thread (utterance 23-25) even though only student A articulated his opinion, because all students had the same visual focus as student A.
* Informed agreement: Informed agreement indicates the number of students who actively participated in an activity and reached an agreement in a discussion thread. The participants of a thread were considered to have reached an agreement when some of them demonstrated their agreement by expressing opinions of agreement, supporting others' opinions, or by the non-verbal cue of nodding. For instance, four students demonstrated informed agreement in the thread starting with argument 61 in Figure 4, because student D nodded to support student A's argument when students A, C, D and E had established shared visual focus.
Criteria elicitation analysis
In order to understand user requirements and concerns regarding a system design, engineers gather direct user feedback through requirement elicitation (Browne, 2001). The elicitation process is integral to understanding individuals' conceptions of a design (Ford et al., 1993; Liu & Tsai, 2005). Therefore, the elicitation process could help in developing understanding of user perception and requirements regarding a building component. In order to investigate the student conception of the learning environment for the purpose of improving the design and, as a result, the learning, this study adopted a student-derived questionnaire as a means of elicitation. The information thusly gathered helps reveal students' concerns and requirements concerning 1:1 learning environments. The purpose of the questionnaire was not to perform a formal comparison between the Shared-Display system and 1:1 learning environment, but to understand, based on students' self-report, the important features that a 1:1 computational environment must have.
Students were required to fill out a questionnaire with questions in the form of the 5-point Likert scale, concerning their perception of the 1:1 and Shared-Display environments for collaborative learning activities. Working in their groups, the students developed questionnaires and then ask their peers to answer the questions. The Wilcoxon signed-rank test was used to test for differences. These questionnaires accurately reflect the students' concerns about effectiveness and design issues of the two environments, because they were produced by the students.
Data collection and data analysis
The design issues and evaluation methods regarding the research questions of this study are shown in Table 1. Data were collected by recording video and audio while students performed collaborative problem-solving activities in the classroom. This study analyzed a total of 102 minutes of discussion activity videos. All analyses of conversational and non-verbal interactions were performed by two independent researchers (coders). The analysis tasks included the classification of student conversation utterances, the identification of non-verbal cues, and the segmentation of conversations. The inter-coder reliability (agreement) for each analysis was at least 77%, indicating that the analysis was sufficiently reliable. Researchers resolved disagreements on analysis or categorization by discussion.
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
Results and discussion
Results of analysis of participation and negotiation
Table 2 and Table 3 give an analysis of thread interaction, based on the group dynamics exhibited by the three collaborative problem solving groups in the two different settings. The number of discussion threads in which students engaged did not vary significantly between the two different settings. Students generated a total of 159 discussion threads, of which 81 occurred in the 1:1 setting and 78 in the Shared-Display setting. Moreover, the depth (i.e. length) of the discussion threads occurring in the 1:1 and Shared-Display settings was also similar, the average depth of a discussion thread being 4.21 in the 1:1 setting and 4.32 in the Shared-Display setting (t=-.308, p=0.758), as shown in Table 3. Students did, however, exhibit different degrees of participation in the two settings. In the 1:1 setting, an average of 2.20 students joined in each discussion. In other words, most interactions occurred between only two students. In the Shared-Display setting on the other hand, each discussion thread attracted the participation of an average of 2.97 students, significantly exceeding the thread participation rate of the 1:1 setting (t=-5.777, p<.001). The participative difference demonstrated by students in the two different settings reveals that the shared display encouraged students to interact with one another and engage in group problem solving activity.
Students demonstrated different degrees of informed agreement in discussion threads depending on their setting. A comparison of the discussion threads in the 1:1 and Shared-Display settings indicates that students exhibited informed agreement behaviors in nearly equal numbers of discussion threads in the two settings (i.e., 39 and 38 respectively). However, more students were involved in the informed agreement occurring in the discussion threads in the Shared-Display setting than in the 1:1 setting. On average, 3.11 students were involved in the interaction threads when informed agreement was reached in the Shared-Display setting. Conversely, students in the 1:1 setting generally only reached informed agreement with the peers sitting immediately adjacent to them. Consequently, an average of only 2.41 students reached informed agreement in discussion threads in the 1:1 setting (t=-3.627, p=.001). This finding indicates that, in order to facilitate collaboration, a group workspace is necessary in addition to personal learning devices.
Results of social interaction analysis
The group dynamic chart indicates that the shared visual focus of students differed between the 1:1 and Shared-Display environments. Students displayed a shared visual focus in more discussion threads in the Shared-Display setting than in the 1:1 setting, 60 in the former as opposed to only 25 in the latter. In addition, more students were included in the shared visual focus occurring in the Shared-Display setting than in the 1:1 setting. On average, each shared visual focus involved 2.46 students in the Shared-Display setting, compared with the significantly smaller average of 0.80 students in the 1:1 setting (t=-7.323, p<.001). This finding indicates that, in the Shared-Display setting, students were more likely to be engaged in common discussion topics than distracted from the discussion by individual work.
Students displayed more hand-pointing behaviors in the Shared-Display environment than in the 1:1 setting. Students exhibited a total of 54 hand-pointing behaviors in the Shared-Display setting, compared with just 8 in the 1:1 setting. The hand-pointing comparison indicates that students interacted with each other in the Shared-Display setting in a livelier manner than when using the 1:1 learning style. Analysis of the group dynamic chart also reveals that informed agreement was sometimes indicated by hand-pointing. Of the 77 occurrences of informed agreement, 18 instances were accomplished by means of hand-pointing behaviors. Hand-pointing behaviors and shared visual focus were co-present in a coherent manner. The establishment of shared visual focus was accompanied by the occurrence of hand-pointing behaviors. Figure 4 shows one example of the co-presence of non-verbal interaction and informed agreement. Five group members reached agreement (utterance 59-64) when they pointed at and watched the shared display during discussion. In contrast, students who watched the screens of handheld devices incited informed agreement in few other participants. Therefore, the Shared-Display setting facilitated the collaboration and discussion process by encouraging students to participate in socially promotive activities, rather than restricting their focus to their own computers.
Results of argument process analysis
An analysis of the conversations that occurred in the 1:1 and Shared-Display settings (Table 4) demonstrates that group members were involved in more question interactions in the 1:1 setting (44%) than in the Shared-Display setting (32%). By contrast, group members presented more positions and arguments (11% and 17% respectively) when utilizing a shared display than when limited to Tablet PCs (6% and 12% respectively). An examination of conversational records reveals that, when they encountered problems, students presented them as issues and called for all group members to contribute their opinions. Since the shared displays provided a shared information space, students' personal work could be rendered on them. Hence, students did not frequently ask for input from peers and proposed fewer issues and questions in the Shared-Display setting than in the 1:1 setting. Nevertheless, students proposed more arguments and positions in the Shared-Display setting than in the 1:1 environment. Analysis of the discussion threads in table 2 also demonstrates that the average depth of threads was 4.32 in the Shared-Display setting, exceeding that of the 1:1 setting (4.21). It may be concluded, therefore, that the shared display enabled group members to articulate their ideas. Furthermore, in conversation, students used indexical words (Table 5) more frequently in the Shared-Display setting (31 times) than in the 1:1 setting (7 times). Brown et al. (1989) defined indexical words as those that index or point clearly to a part of the situation in which communication is being conducted. This phenomenon, as it exists in this study, demonstrates that students can easily point out information by using indexical words.
Results of criteria elicitation analysis
Students produced a total of 25 question items. These items concern six different dimensions of the collaborative learning environment supported by 1:1 technologies. Table 6 summarizes the six dimensions and the criteria included in each, as adapted from the student-derived questionnaires. Of the 25 items on the students' questionnaire, five items relate to the four criteria in the dimension of discussion, seven items to the three criteria in the dimension of facilitation of group work, one item to the one criterion in the dimension of supporting activity awareness, two items to the two criteria in the dimension of individual work, two items to the two criteria in the dimension of general learning interface design, and seven items to the five criteria in the dimension of general HCI (Human Computer Interaction) design. From these criteria, one can infer that students feel that effective 1:1 learning environments must have both an adequate interface design to support individual work on individual devices and adequate functions to support efficient and effective collaborative work (i.e. discussion, group work, and activity awareness).
Judging by their proposed criteria, students perceive the Shared-Display setting as providing more substantial support than the 1:1 setting in the criteria of ease of discussion, increase in discussion frequency, learning through discussion, and shortening of discussion time. Their feedback was consistent with the findings of interaction thread analysis in that the shared display encouraged more members to contribute to the discussion threads. Furthermore, students considered the shared display beneficial to group work in the criteria of aiding group work efficiency and generation of group answers, facilitating group shared understanding, and facilitating collaboration. Student feedback was consistent with the findings of the social interaction analysis, in that students were more likely, in the Shared-Display setting, to be engaged in common discussion topics than distracted from the discussion by individual work. Students perceived that the shared-display environment significantly promoted group awareness for the collaborative work. This might explain the significant increase in shared visual focus in the Shared-Display setting.
On the subject of general learning interface design, students regarded facilitating learning and increasing the motivation to learn as two important features of a learning environment. Students considered the Shared-Display setting to be satisfactory in relation to both criteria and regarded it as a favorable environment for learning. In the dimension of general HCI design, according to interview results, students regarded the facilitation of group discussion, rather than merely the presence of an easy-to-use interface, as an important factor in easy manipulation. They also thought that shared displays did not add extra usage load, but served to facilitate group discussion. Therefore, students preferred the Shared-Display setting with respect to the general HCI design dimension. Consistent with their opinions, students showed higher average scores in the Shared-Display setting.
Conclusion and implications
Capturing how students learn with peers through different technological devices and how these devices influence the interaction among students is a difficult task, because the representation of information and knowledge is distributed among different ubiquitous devices. This study captures both the conversation logs of groups and their social cues, which are often not examined by traditional online interaction analysis methodologies, as mediated by different technological devices (Liu & Tsai, 2006; Stahl, 2005). This integrated analytical methodology was found to be helpful for both quantitatively and qualitatively analyzing students' face-to-face discussions and collaborative problem solving activities.
Effectiveness of shared-display environment on participation and negotiation
The results of integrated analysis revealed that students in the 1:1 environment frequently exhibited fragmented communication patterns, i.e. most interactions occurred between two students. The shared displays attracted more students to participate in discussion threads than 1:1 environment. Furthermore, students showed informed agreements with more peers in shared-display environment than they did in 1:1 environment. Therefore, the shared display could not only facilitate students' participation but also negotiation process in the collaborative problem solving activity. These findings support that shared workspaces attracted students' participation and facilitated negotiation process in the scenarios of using handhelds to support collaborative problem solving activities.
Student questionnaire feedback reveals that shared displays could provide substantial support for improvement of awareness of group answers, ease of discussion, generation of group answers, facilitation of collaboration, efficiency of discussion, and willingness to use the system. The shared display helps students to integrate different ideas and to construct shared visual focus (attention). Therefore, learners in 1:1 classrooms need not only an individual learning space for gathering their learning resources and organizing their thoughts, but also a shared learning space to construct shared attention.
Effectiveness of shared-display environments on argument process
The integrated analysis reveals that students proposed more arguments and positions when collaborating in the Shared-Display environment than they did in the 1:1 environment. This finding indicates that the 1:1 learning environment needs not only individual workspaces, but also a shared workspace for students to contribute and explain their ideas. These findings are consistent with Greenberg's argument that a physical shared workspace is required to improve social interaction and collaboration (Greenberg, 1996). Due to the lack of shared workspace, students provided only with personal handheld devices are restricted to interacting with peers on their individual devices and thus cannot take part in lively promotive discussion. This study indicates the necessity for environmental transformation in classroom layout and equipment in order to establish effective and socially promotive collaboration with personal learning devices.
Effectiveness of shared-display environments on social interaction
The results of integrated analysis demonstrate that students in the Shared-Display setting often explained their thoughts by hand-pointing. Moreover, students frequently established shared visual focus and informed agreement in the Shared-Display setting. However, students in the 1:1 setting rarely demonstrated hand-pointing behaviors or shared focus. Because hand gesture and eye contact behaviors are two indicators of social presence and mutual awareness, it is apparent that the shared display facilitated mutual awareness to augment social presence and socially promotive collaboration (Argyle & Dean, 1965; IJsselsteijn et al., 2000). Handheld devices may impede socially promotive interaction if the design of the devices and software is not conducive to the communication these same devices are used to mediate (Gunawardena, 1995; Richardson & Swan, 2003). Previous studies have confirmed that social presence is a key factor in effective and promotive computer-mediated learning activities (Garrison et al. 2000; Kim et al., 2006). Roth & Welzel (2001) also confirmed that gestures help students to construct complex learning activities. Students in the 1:1 setting tended to concentrate on using their own devices, disregarding the activities of the other participants, while students with access to a shared display were able to link complicated information and to exchange ideas on a shared workspace with the use of lively non-verbal cues. Thus, the shared display can encourage students to participate in socially promotive activities. It is our recommendation that 1:1 environments and activities should be designed to encourage all participants to participate jointly in shared socially promotive activity.
Student responses state that looking at the shared display increased their awareness of collaborative work. Activity awareness was achieved through the shared display rather than the handheld devices. This finding confirms the importance of activity awareness (Greenberg et al., 1996; Hill & Gutwin 2003; Liu et al., 2007) and of social clarity in face-to-face situations (Erickson & Kellogg, 2000). Students rarely established shared focus in the 1:1 setting, where information and activities were often concealed in individual devices. The lack of activity awareness impeded collaboration between group members. Students demonstrated more effective interactive behaviors and active engagement in the Shared-Display setting than in the 1:1 setting, because the shared display facilitated interaction by raising activity awareness. This finding confirms the requirement for visibility and awareness of activity processing in 1:1 learning environments.
The authors would like to thank the National Science Council of the Republic of China, Taiwan for financially supporting this research under Contract No. NSC 96-2524-S-008-001 and NSC96-2520-S-008-006-MY2, NSC 962524-S-008-002 and NSC 97-2631-S-008-003.
Aronson, E., Blaney, N., Sikes, J., Stephan, C., & Snapp, M. (1975) Busing and racial tension: The jigsaw route to learning and liking. Psychology Today, 8, 43-59.
Argyle M., & Dean J. (1965). Eye-contact, distance and affiliation. Sociometry, 28, 289-304.
Barab, S., MaKinster, J. G., Moore, J., Cunningham, D., & the ILF Design Team. (2001). Designing and building an online community: The struggle to support sociability in the inquiry learning forum. Educational Technology Research and Development, 49 (4), 71-96.
Barab, S., & Squire, K. (2004). Design-based research: Putting a stake in the ground. The Journal of the Learning Sciences, 13 (1), 1-14.
Browne, G. J. (2001). An empirical investigation of user requirements elicitation: Comparing the effectiveness of prompting techniques. Journal of Management Information Systems, 17 (4), 223-249.
Buxton, W. (2001). Less is more (more or less): Some thoughts on the design of computers and the future. The Invisible Future: The Seamless Integration of Technology in Everyday Life, New York: McGraw Hill.
Blumenfeld, P. C., Marx, R. W., Soloway, E., & Krajcik, J. (1996). Learning with peers: From small group cooperation to collaborative communities. Educational Researcher, 25 (8), 37-42.
Collins, A. (1992). Towards a Design Science in Education. In Scanlon, E. & O'Shea, T. (Eds.), New Directions in Educational Technology (pp. 15-22), New York: Springer.
Dillenbourg, P., & Traum, D. (2006). Sharing solutions: Persistence and grounding in multimodal collaborative problem solving. Journal of the Learning Sciences, 15 (1), 121-151.
Erickson, T., & Kellogg, W. A. (2000). Social translucence: An approach to designing systems that support social processes. ACM Transactions on Computer-Human Interaction, 7 (1), 59-83.
Ford, K., Bradshaw, J., Adams-Webber, J. & Agnew, N. (1993) Knowledge acquisition as a constructive modeling activity, International Journal of Intelligent Systems, 8, 9-32.
Greenberg S., Gutwin C. and Cockburn, A. (1996). Using Distortion-Oriented Displays to Support Workspace Awareness. In Sasse, A., Cunningham, R. & Winder, R. (Eds.), People and Computers XI (pp. 299-314), New York: Springer.
Hutchins, E. (1995). Cognition in the Wild, Cambridge, MA: MIT Press.
Hill, J., & Gutwin, C. (2003). Awareness support in a groupware widget toolkit. Proceedings of the 2003 international ACM SIGGROUP conference on supporting group work, New York: ACM Press, 258-267.
Hwang, G. J., Tsai, C. C., & Yang, S. J. H. (2008). Criteria, Strategies and Research Issues of Context-Aware Ubiquitous Learning. Educational Technology & Society, 11 (2), 81-91.
IJsselsteijn, W.A., de Ridder, H., Freeman, J., & Avons, S. E. (2000). Presence: Concept, determinants and measurement, retrieved March 20, 2009 from http://www.ijsselsteijn.nl/papers/SPIE_HVEI_2000.pdf.
Johnson, D. W., Johnson, R. T., & Holubec, E. J. (1994). The New Circles of Learning: Cooperation in the Classroom and School, Virginia: ASCD.
Klerfelt, A. (2007). Gestures in conversation-the significance of gestures and utterances when children and preschool teachers create stories using the computer. Computers & Education, 48 (3), 335-361.
Kim, K., Tatar, D., & Harrison, S. (2006). Handheld-mediated communication to support the effective sharing of meaning in joint activity. Proceedings of the 4th International Workshop on Wireless, Mobile and Ubiquitous Technologies in Education, Los Alamitos, CA: IEEE Computer Society, 171-173.
Kunz, W., & Rittel, H. (1970). Issue as Elements ofInformation Systems, Berkeley, CA: University of California.
Lyman, F. (1981). The Responsive Class Discussion. In Anderson, A. S. (Ed.), Mainstreaming Digest, College Park: University of Maryland.
Lai, C. Y., & Wu, C. C. (2006). Using handhelds in a Jigsaw cooperative learning environment. Journal of Computer Assisted Learning, 22, 284-297.
Liu, C. C., Tao, S. Y., & Nee, J. N. (2007). Bridging the gap between students and computers: supporting activity awareness for network collaborative learning with GSM network. Behaviour and Information Technology, 27 (2), 127-137.
Liu, C. C., & Kao L. C. (2007). "Do Handheld Devices Facilitate Face-to-Face Collaboration?": Handheld Devices with Large Shared Display Groupware to Facilitate Group Interactions. Journal of Computer Assisted Learning, 23 (4),285-299.
Liu, C. C., & Tsai, C. C. (2006). 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., Chou, C. C., Liu, B. J., & Yang, J. W. (2006). Improving mathematics teaching and learning experiences for hard of hearing students with wireless technology-enhanced classrooms. American Annals of the Deaf, 151 (3), 345-355.
Milson, F. (1973). An Introduction to Group Work Skill, London: Routledge and Degan Paul.
Richardson, J. C., & Swan, K. (2003). Examining social presence in online courses in relation to students' perceived learning and satisfaction. Journal ofAsynchronous Learning Networks, 7 (1), 68-88.
Roth W. M., & Welzel M. (2001). From activity to gestures and scientific language. Journal of Research in Science Teaching, 38, 103-136.
Roschelle, J. (2003). Unlocking the learning value of wireless mobile devices. Journal of Computer Assisted Learning, 19 (3), 260-272.
Saljo, R. (2000). Larande i praktiken. Ett soczokulturellt perspektiv (Learning in practices. A sociocultural perspective) (in Swedish), Stockholm: Prisma.
Scardamalia, M. (2004). CSILE/Knowledge Forum. Education and technology: An encyclopedia (pp. 183-193), Santa Barbara: ABC-CLIO.
Suthers, D. D. (2006). Technology affordances for intersubjective meaning making: A research agenda for CSCL. International Journal of Computer-Supported Collaborative Learning, 1 (3), 315-337.
Suthers, D. (2005). Collaborative knowledge construction through shared representations. Paper presented at the 38th Hawaii International Conference on the System Sciences (HICSS-38), January 3-6, 2005, Waikoloa, Hawaii.
Scott, S. D., Mandryk, R. L., & Inkpen, K. M. (2003). Understanding Children's Interactions in Synchronous Shared Environments. Journal of Computer Assisted Learning, 19 (2), 220-228.
Stahl, G. (2005). Group Cognition: Computer Supportfor Collaborative Knowledge Building, Cambridge, MA: MIT Press.
Short J., Williams E., & Christie, B. (1976). The social psychology of telecommunications, London: Wiley.
Zurita, G., & Nussbaum, M. (2004). Computer supported collaborative learning using wirelessly interconnected handheld computers. Computers and Education, 42 (3), 289-314.
Chen-Chung Liu (1), Chen-Wei Chung (2), Nian-Shing Chen (3) and Baw-Jhiune Liu (4)
(1) Graduate Institute of Network Learning Technology, National Central University Taiwan //Tel: +886-4-227151-35412 //firstname.lastname@example.org
(2) Department of Computer Science and Engineering, National Central University Taiwan //Tel: +886-4-227151-57880// jerryjongggmail.com
(3) Department of Information Management, National Sun Yat-sen University, Taiwan //Tel: +886-7-5252510 // email@example.com
(4) Department of Computer Science and Engineering, Yuan-Ze University, Taiwan // Tel: +886-4-638800-2366 // bjliu@firstname.lastname@example.org
Table 1. The research questions, issues, and evaluation methods Research Questions Issues Evaluation methods Research Question I Process of participation Group dynamic chart and negotiation analysis Research Question 2 Social interaction Non-verbal analysis Research Question 3 Argument process Conversation analysis Table 2. Statistics of artici ation, negotiation and social interaction Group 1 Group 2 1:1 * Shared 1:1 Shared Number of students 5 5 4 5 Number of threads 27 21 23 14 Average depth of threads 2.85 3.38 6.04 6.43 Average number of students 1.96 2.19 2.17 3.78 participating in threads Number of threads demonstrating 4 12 8 11 shared visual focus Average number of students 0.37 1.41 0.70 3.07 showing shared visual focus in threads Frequency of hand-pointing 5 13 1 24 Number of threads demonstrating 8 8 13 10 informed agreement Average number of students 2 2.25 2.31 3.90 showing informed agreement in threads Group 3 Total 1:1 Shared 1:1 Shared Number of students 5 4 14 14 Number of threads 31 43 81 78 Average depth of threads 4.03 4.08 4.21 4.32 Average number of students 2.41 3.09 2.20 2.97 participating in threads Number of threads demonstrating 13 37 25 60 shared visual focus Average number of students 1.26 2.74 0.80 2.46 showing shared visual focus in threads Frequency of hand-pointing 2 17 8 54 Number of threads demonstrating 18 20 39 38 informed agreement Average number of students 2.67 3.05 2.41 3.11 showing informed agreement in threads * 1:1 represents one-to-one setting Table 3. Difference test between 1:1 and Shared-Display environments Difference test of depth of threads N Mean SD t-value p-value 1:1 81 4.21 3.48 -0.308 0.758 Shared 78 4.32 2.82 Difference test of number of participants in threads 1:1 81 2.20 0.86 -5.777 .000 *** Shared 78 2.97 0.84 Difference test of number of students showing shared visual focus in threads 1:1 81 0.80 1.34 -7.323 .000 *** Shared 78 2.46 1.52 Difference test of number of students showin, informed agreement in threads 1:1 39 2.41 0.82 -3.627 .001 ** Shared 38 3.11 0.86 ** p<0.01, *** p<0.001 Table 4. Statistics of conversational utterances Group 1 Group 2 1:01 Shared 1:01 Shared Issue/ questions 21 (58% 16 (48%) 21 (42%) 7 (22%) Position 1 (3%) 4 (12%) 4 (8%) 6 (18%) Artlument 4 (11%) 2 (6%) 6 (12%) 10 (31%) Group 10 (28%) 11 (33%) 19 (38%) 9 (28%) development Group 3 Total 1:01 Shared 1:01 Shared Issue/ questions 21 (38%) 20 (29%) 63 (44%) 43 (32%) Position 4 (7%) 5 (7%) 9 (6%) 15 (11%) Artlument 7 (13%) ll (16%) 17 (12%) 23 (17%) Group 24 (43%) 33 (48%) 53 (37%) 53 (39%) development Table 5. Occurrence of indexical words One-to-one (1:1) Shared-Display Issue/questions 2 14 Position 0 8 Argument 0 6 Group development 5 3 Table 6. Students' perception of the 1:1 and Shared-Display environments Dimension Criteria # of 1:1 Shared item (+) Discussion Easy for discussion, 5 2.64 4.50 ** shortens discussion time, increases discussion frequency, aids learning through discussion Group work Aids group work efficiency 7 2.65 4.29* and generation of group answers, facilitates collaboration, facilitates group shared understanding Activity Aids awareness of others' 1 3.15 4.30 ** awareness answers Individual work Aids editing of individual 3 3.61 4.00 ** answers, aids learning through generation of individual answers General learning Facilitates learning, 2 3.12 4.30 * interface design increases motivation to learn General HCI Easy to manipulate, user 7 3.03 4.03 ** design friendly, eases off usage load, robust, evokes willingness to use the system again, shortens learning time of svstem functions * p<0.05 **p<0.01 (+) represents the average score of all the items in a dimension.
|Printer friendly Cite/link Email Feedback|
|Author:||Liu, Chen-Chung; Chen-Wei, Chung; Nian-Shing, Chen; Liu, Baw-Jhiune|
|Publication:||Educational Technology & Society|
|Date:||Jul 1, 2009|
|Previous Article:||Effects of the ISIS recommender system for navigation support in self-organised learning networks.|
|Next Article:||Reflections of students in their use of asynchronous online seminars.|