Assigned Roles as a Structuring Tool in Online Discussion Groups: Comparison of Transactional Distance and Knowledge Sharing Behaviors

Abstract

The purpose of this study is to determine the effects of structured discussion by giving participants in online discussion groups different roles in transactional distance perception and knowledge sharing behaviors. In the study, a quasi-experimental design was utilized with pretest and posttest control groups. The study was conducted with 111 first-year undergraduate students at a university for 10 weeks. Quantitative and qualitative data collection techniques were used in the study, and the data were obtained from the Transactional Distance Scale, Knowledge Sharing Behavior Scale, and Student Opinion Determination Form. The findings indicate that online discussions carried out by assigning roles to discussion group members (starter, moderator, arguer, source searcher, and summarizer) decreased students’ transactional distance perceptions and improved knowledge sharing behaviors in online discussions. Different suggestions are offered about how to structure online discussions in accordance with the findings of the qualitative data analysis.

Keywords

online discussion assigned roles transactional distance knowledge sharing behaviors

Introduction

Social learning theories focus on learning that occurs in a social context, based on personal experiences, and observations of and interactions with other individuals in social communities. The social constructivist paradigm draws attention to the social dimensions of the learning process, stating that the environment and social interaction is important in learning. According to the social constructivist paradigm, learners construct knowledge through sharing and discussing information with their friends and teachers during the learning process and then assimilating it. The need to construct knowledge emerges when individuals try to make inferences from their experiences. In social constructivism theory, it is the social processes that lie on the basis of cognitive processes. According to this theory, it is not individuals but societies that construct knowledge. The meanings inferred from experiences are valid as long as they are accepted by the members of the society. To construct knowledge and create a consensus about the knowledge, it is necessary for the group members to interact (Vygotsky, 1988). Therefore, interaction and cooperation with others are important in the learning process. Today’s university students grew up familiar with the Internet and other forms of virtual technology (Prensky, 2001). Teachers currently try to benefit from their students’ digital literacy in order to generate engaging and innovative learning experiences, to increase interaction and cooperation between individuals in the learning process, and to enable knowledge to be constructed socially (Salter & Conneely, 2015). One of the tools that is frequently used for these purposes is asynchronous online discussions.

Asynchronous online discussions enable knowledge to be constructed socially and cooperatively, give students time to think before contributing, and allow them to progress at their own pace. In addition, asynchronous online discussions give generally introverted students chances to participate in discussions (Murphy & Coleman, 2004). Moreover, these online discussions enable classroom discussions to be carried on outside the class, and they provide opportunities for students and teachers to share knowledge and ideas without having to be physically in the same place. Despite the many benefits that asynchronous online discussions provide to the learning process, one of the problems faced in these online discussions is that students’ rate of participation in the discussions is low and that a strong sense of community cannot be created. In a study by Salmon (2004), it was found that only one third of the students in online discussions made a contribution, only one third read what was written, and the remainder never participated. According to researchers, the fact that the number of students participating in the online discussions is low and the sense of community lacking may be related to the fact that students’ transactional distance perceptions are high, and the quality of their knowledge sharing behaviors is low (Karaoglan Yilmaz, 2017a, 2017b; Remesal & Colomina, 2013; Rovai, 2007; Xie, Yu, & Bradshaw, 2014). Therefore, measures should be taken to decrease transactional distance perception of the students in online discussions and increase their knowledge sharing behaviors.

One of the strategies used in increasing the efficiency of the students in online discussions is role assignment. Accordingly, it is seen that such roles as starter, moderator, arguer, source searcher, and summarizer are assigned to students to improve the quality of discussion (De Wever, Van Keer, Schellens, & Valcke, 2010; Gu, Shao, Guo, & Lim, 2015; Salter & Conneely, 2015). On the other hand, how role assignment will effect students’ transactional distance perception and knowledge sharing behaviors in online discussions should be further examined. Therefore, the main objective of this study is to examine the impact of assigning such roles as starter, moderator, arguer, source searcher, and summarizer to the group members in online discussions on students’ transactional distance perceptions and knowledge sharing behaviors and thus contribute to the depth and flow of research on the topic.

Theoretical Background and Research Problems

Online Discussions and Transactional Distance

Transactional distance perception is defined as a psychological and communicational space that causes probable misunderstandings between students, teachers, and content elements in different places. In other words, transactional distance perception is not just the physical distance between student and teacher (Moore & Kearsley, 2011). Within the framework he developed, Zhang (2003) asserted that student–student, student–teacher, student–content, student–interface, and student–environment interactions have a determining role on the transactional distance perceptions that students have in online environments. According to Zhang, as the quality of these interactions increases, the transactional distance perceptions of individuals related to the environment decrease.

Studies have concluded that transactional distance perception was a significant predictor of positive attitudes toward student achievement, satisfaction, the permanence of what is learnt and content (Ekwunife-Orakwue & Teng, 2014; Goel, Zhang, & Templeton, 2012; Wengrowicz, 2014). Online discussions bring together students, teachers, and content in different places and enable the social construction of knowledge. Students’ transactional distance perception is an important factor effecting the efficiency of these. Transactional distance perception is one of the factors affecting online discussions designed to construct knowledge socially by bringing students, teachers, and content in different places together. Students’ transactional distance perception toward online discussion is an important factor regarding the efficiency of these environments. Studies reveal that transactional distance perception is influenced by the type of interaction and its structure (Karaoglan Yilmaz, 2017a, 2017b; Yilmaz & Keser, 2017). Ekwunife-Orakwue and Teng’s (2014) research also indicates that transactional distance perception may change according to the quality of student–student, student–teacher, and student–content interactions in online discussions and the interface and features used for the online discussions. Karaoglan Yilmaz (2017a) found a high correlation in a positive direction between community of inquiry among students in virtual learning community and its components such as teaching presence, social presence, and cognitive presence and transactional distance. In the light of these findings, it can be said that using strategies that will activate students’ teaching presence, social presence, and cognitive presence will decrease their transactional distance perceptions. In this context, assigning students such roles as starter, moderator, arguer, source searcher, and summarizer in online discussions will improve student–student and student–content interactions and reduce transactional distance perceptions. However, this hypothesis is a topic to be studied in the literature.

Online Discussions and Knowledge Sharing Behaviors

Another problem in online discussions involves knowledge sharing behaviors in these environments. An important dynamic of online discussion environments is knowledge sharing behaviors with the community. As for Alakurt (2013) knowledge sharing behaviors include such activities as replying to a general or private message, participating in a discussion, helping a member with a problem, and sharing document or file. In addition to being an important element for virtual communities of practice, knowledge sharing is considered as an important reason for motivation (Fang & Chiu, 2010). In the communities where knowledge sharing is the main activity, encouraging individuals to share knowledge and increasing their willingness and frequency of knowledge sharing are among the leading problems in online knowledge sharing process (M. J. J. Lin, Hung, & Chen, 2009). Therefore, it is indicated that studies toward identifying factors that encourage or discourage behaviors in knowledge sharing process (Alakurt, 2013).

Studies show that participants in online discussions taking part only superficially share knowledge and that the level of structured knowledge is low (Ng, Cheung, & Hew, 2009). According to Rovai (2007), various solutions to this have been proposed, such as having students spend more time online, fostering interaction, better meeting their expectations, and having more task-oriented and socially oriented discussions. One of the dimensions of knowledge sharing behaviors that should be studied is how online discussions can be designed to follow these recommendations (Alakurt, 2013).

Karaoglan Yilmaz (2017b) found that social presence perception and its subfactors, which are interaction, ownership, and affective statements, have an effect on students’ knowledge sharing behaviors. According to the study, knowledge sharing behaviors develop as social presence perception increases. Another finding in the study was that transactional distance perception was important factor on knowledge sharing behaviors. Accordingly, it is believed that assigning roles in online discussions will improve students’ interaction and ownership feelings in online discussions and contribute to realization of high-quality and comprehensive discussions. Besides, as the roles and responsibilities of everyone will be preset in discussions, student–student and student–content interactions will be more developed. This is believed to improve knowledge sharing behaviors in online discussions. However, this hypothesis should be further studied.

Assigned Roles in Online Discussions

To solve the problems related to transactional distance perceptions and knowledge sharing behaviors in asynchronous online discussions, the inclusion of different design elements has been discussed. According to Gu et al. (2015), one of these elements is how online discussions are structured.

Research suggests that appropriately structured online discussions can lead to meaningful interactions between students, teachers, and aid the delivery of content. The interest here is how to best structure these online discussions and processes. In structured online discussions, components such as the rules to be followed, tasks to be carried out, which roles will be assumed by whom, and so on, are determined before the discussion, and the participants have to comply with these rules and roles. These rules and roles are more flexible in unstructured online discussions (Moore & Kearsley, 2011; Yilmaz, 2014). The literature shows that the basic educational approaches, interaction environment and technology used, and the objectives and rules all have an effect on the structure of online discussions (Hollingsead et al., 2006; Strijbos, Martens, Jochems, & Broers, 2007; Moore & Kearsley, 2011).

Another method that can be used in structuring online discussions is assigning different roles to group members (De Wever, Schellens, Van Keer, & Valcke, 2008; Strijbos et al., 2007). Roles can be defined as tasks or responsibilities that organize group interaction and that guide students’ behaviors (Ho, Rappa, & Chee, 2009; Strijbos et al., 2007). Roles can include being the one who listens to or interprets a discussion; the one who starts a discussion, guides it, and provides a theoretical contribution; the one who looks for references; and the one who collects the data and summarizes the discussion (De Wever et al., 2010; Gu et al., 2015; Salter & Conneely, 2015).

Purpose of the Study

The literature shows that, in online discussions, assigning different roles to group members helps discussions get started, enables their sustainability, affects social construction of knowledge positively, and increases student participation and engagement (De Wever et al., 2010; Gu et al., 2015; Salter & Conneely, 2015). On the other hand, it is wondered how assigning roles would affect the main problems of online discussions, which are high transactional distance perception and lack of knowledge sharing behaviors in students. The purpose of this study conducted with aforementioned uncertainty and situations in mind is to investigate the effect of assigning roles to students as starter, moderator, arguer, source searcher, and summarizer on transactional distance perception and knowledge sharing behaviors. In line with this purpose, answers have been sought to the following research questions:

Is there a significant difference between Transactional Distance Scale pretest and posttest scores of students in structured and unstructured online discussions?

Is there a significant difference between knowledge sharing behaviors scale pretest and posttest scores of students in structured and unstructured online discussions?

What are the students’ opinions about structured and unstructured online discussions?

In this context, unstructured online discussions were those in which students were not assigned roles.

Method

Research Model

In this research, both the qualitative and quantitative research methods were used. Research was carried out according to a quasi-experimental design with a pretest or posttest control group. The data of this study were collected from the Transactional Distance Scale and Knowledge Sharing Behavior Scale, which were used as pretest and posttest. Moreover, a Student Opinion Determination Form was used in order to determine the students’ opinions at the end of the procedure.

Participants

The study was conducted with a group of political science and public administration undergraduate students taking the Computing II course at a state university. A total of 111 students (51 males and 60 females) took part in the study. In the first class for this course, students had already been divided into two different groups. Since students had already been split up in the term before the research began, it was planned according to the quasi-experimental design with regard to these groups. There were 53 students in the experimental group and 58 students in the control group. Determining which group would form the experimental group and which would form the control group was carried out at random. The experimental group used structured online discussions, in which roles were assigned to the students, and the control group used unstructured online discussions, in which students were not assigned roles.

The students were similar to each other in terms of age range (students in both groups were in the age range 17–20), their academic success in the Computing I class that they had taken the previous term (experimental group, $\bar{x} = 79.34$ ; control group, $\bar{x} = 79.62$ ), and the number of students in each group by gender. Those students in experimental and control groups were taught basic computer hardware, operating systems, spreadsheets, presentations, and Internet literacy within the course of Computing I in the previous semester. Moreover, digital citizenship issues which are among online discussion topics (see also online discussions and procedure) were in the curriculum.

The students in the experimental and control groups had engaged in discussions about the course individually in a forum integrated into the Moodle learning management system during the Computing I course in the fall term. Therefore, the students already had some transactional distance perceptions and knowledge sharing behaviors within online discussion environments.

Instruments

Transactional Distance Scale

In the study, the Transactional Distance Scale developed by Zhang (2003) and adapted into Turkish by Yilmaz and Keser (2015) was used to determine the transactional distance perceptions of the students in the experimental and control groups. The 5-point Likert-type scale consists of five subdimensions (student–interface interaction, 8 items; student–content interaction, 6 items; student–teacher interaction, 6 items; student–student interaction, 11 items; student–environment interaction, 7 items) and 38 items. The Cronbach α reliability coefficients calculated for this study were .80 for the student–interface interaction subdimension, .84 for the student–content interaction subdimension, .86 for the student–teacher interaction subdimension, .94 for the student–student interaction subdimension, and .87 for the student–environment interaction subdimension. The Cronbach α reliability coefficient calculated for the whole scale was .97. From these results, it may be said that the scale is reliable. The scale was used as a pretest and posttest in this study. High scores from the scale indicate a low transactional distance perception; in other words, there is a high degree of interaction and strong feeling of closeness.

Knowledge Sharing Behavior Scale

In the study, the Knowledge Sharing Behavior Scale developed by Alakurt (2013) was used to determine the knowledge sharing behaviors of the students in the experimental and control groups. The 5-point Likert-type scale consists of five subdimensions (usefulness or compliance, 4 items; sacrifice, 3 items; respect, 4 items; being happy to help, 7 items; trust, 3 items) and 21 items. The Cronbach α reliability coefficients calculated for this study were .92 for the usefulness or compliance subdimension, .87 for the sacrifice subdimension, .88 for the respect subdimension, .94 for the being happy to help subdimension, and .90 for the trust subdimension. The Cronbach α reliability coefficient calculated for the whole scale was .96. From these results, it may be said that the scale is reliable. The scale was used as a pretest and posttest in this study.

Student Opinion Determination Form

The Student Opinion Determination Form, which consists of semistructured questions, was developed by the researchers in order to determine the opinions of the students about the online discussions they were having. The aim of using the Student Opinion Determination Form was to identify students’ opinions on transactional distance perception and knowledge sharing behaviors in online discussions. In this regard, students in experimental and control groups were asked what positive and negative effects online discussions had on transactional distance perception (interactions of student–student, student–teacher, student–content, student–interface, student–environment) and knowledge sharing behaviors. In addition, students in experimental group were asked what the influences of assigning roles to group members were on transactional distance perception and knowledge sharing behaviors in online discussions. The questions were semistructured and used in the study after the necessary corrections had been made depending on expert review. The form was applied at the end of the experimental research process.

Online Discussions and Procedure

A forum which had been integrated into the Moodle learning management system was used for the online discussions. In the research, two different online discussions, structured and unstructured, were developed for the experimental and control groups, respectively. The main difference between the discussions was that students in the structured online discussions were given various roles and responsibilities. These roles were defined based on the literature. These roles were starter, moderator, arguer, source searcher, and summarizer (De Wever et al., 2010; Gu et al., 2015).

The starter starts the discussion by conducting preanalysis related to the subject and tries to make the discussion sustainable by starting discussions in different contexts and cooperating with group members whenever the discussion becomes mundane or reaches an impasse. The moderator follows the flow of the discussion, evaluates whether the discussion is relevant to the subject, budgets for time, checks whether group members are following the rules, and coordinates group members. The arguer formulates questions from the sources given by liaising with the source searcher about the topic of the discussion, asks questions by liaising with the starter in order to continue discussions started by starter, and helps the discussion remain focused by taking a critical approach to the answers that participants give. The source searcher searches for information related to the topic before and after the discussion. The summarizer notes the major points and issues brought up by summarizing the important information related to the topics at the end of the process. The students in the control group participated in an unstructured discussion. They were not assigned specific roles, and all of the students had a similar responsibility to participate in the discussion and carry out the work in a group. There was no difference between the structured and unstructured discussion groups other than the assignment of roles in the former. The online discussions were introduced to the students in both groups before the testing process and the discussion rules were described. In addition, the students in the structured discussion group were informed about their roles and responsibilities. The researcher did not interfere in either of the discussion groups directly. Unless there was a major problem, the process was simply allowed to unfold so that the experimental process was not affected. In the structured groups, the assignment of roles was determined according to the students’ wishes.

The students in the experimental and control groups were divided into subgroups of five people. Each subgroup was given topics relating to digital citizenship, and every week discussions were initiated through the presentation of materials and content that had been prepared by subgroups in the online discussions. Discussions relating to the topics in question were held during 1 week until the following week’s topic started to be covered, and they continued for 10 weeks, each week covering one topic. The study was carried out within the course of Computing II instructed through blended learning method. In face-to-face part of the lesson, students in subgroups came together in classroom environment and designed contents and materials they would use in online discussions. Within that period, the teacher informed students about instructional design principles, material design, and development processes. In online phase of the lesson, subgroups shared the contents and materials they had prepared in the online discussion platform and held asynchronous discussions regarding the topics. Within the scope of digital citizenship, subgroups developed contents and materials and discussed about digital access, digital trade, digital communication, digital literacy, digital ethics, digital law, digital health, digital security, digital rights, and responsibilities. At the beginning of the research, students in the experimental and control groups were informed about how online discussions would be carried out, and discussion setting was introduced. Students in the experimental group were also informed about how discussions would be held based on the roles assigned to group members in online discussions.

Data Analysis

Whether the scores of the students for the Transactional Distance Scale and the Knowledge Sharing Behavior Scale showed a normal distribution was tested with the Kolmogorov–Smirnov test. The results indicated that the data were normally distributed (p > .05). Therefore, parametric tests were used for data analysis. To determine whether the dependent variables were different between the experimental and control groups, posttest results of the groups were compared with the pretest scores of the groups. The analysis of covariance (ANCOVA) test was used for this comparison. Content analysis was used to analyze the qualitative data. The data obtained in the study were coded by a researcher, and then the data obtained were recoded by a second coder to test the reliability of the study. The reliability of the coding was obtained by dividing the number of common codes given by both encoders by the total number of codes. For the reliability of the study, the two researchers independently codified, and the percentage of compliance was calculated as 88% according to Miles and Huberman (1994). For the remaining 12% difference, the encoders came together and reached a consensus. After the written explanations of the students were examined, it was determined that this difference had arisen because some answers given by one student had been collected under several subthemes.

Findings

Descriptive Statistics of the Scales

Descriptive statistics calculated according to the answers the students gave in the Transactional Distance Scale and Knowledge Sharing Behavior Scale used as pretest and posttest in the study are presented in Table 1.

Table 1.

Descriptive Statistics on Pretest and Posttest Scores.

Group	Scales	Minimum score	Maximum score	Mean	Standard deviation
Experimental group	Transactional Distance Scale—pretest	81	190	139.04	24.87
	Transactional Distance Scale—posttest	107	190	154.91	19.68
	Knowledge Sharing Behavior Scale—pretest	31	105	76.42	16.16
	Knowledge Sharing Behavior Scale—posttest	53	105	85.70	12.36
Control group	Transactional Distance Scale—pretest	94	189	146.45	24.42
	Transactional Distance Scale—posttest	64	190	147.19	28.19
	Knowledge Sharing Behavior Scale—pretest	40	105	78.84	14.81
	Knowledge Sharing Behavior Scale—posttest	21	105	78.90	16.42

Examining the results in Table 1, it was seen that pretest average score for the Transactional Distance Scale of the students in the experimental group was $\bar{x} = 139.04$ , and the pretest average score for the Knowledge Sharing Behavior Scale was $\bar{x} = 76.42$ . The pretest average score for the Transactional Distance Scale of the students in the control group was $\bar{x} = 146.45$ , and the pretest average score for the Knowledge Sharing Behavior scale was $\bar{x} = 78.84$ . Although the pretest average scores of test and control groups were close to each other, it was decided to perform an ANCOVA comparing the posttest scores by assessing the pretest scores in order to monitor the effect of differences in the pretest scores on the research process. The significance of these differences was analyzed in accordance with the goals of the study.

Findings Related to Transactional Distance Perception

The first research question aimed to find whether there was a statistically significant difference between the transactional distance perception scores of the students in the experimental and control groups. The students’ pretest and posttest scores for the Transactional Distance Scale were analyzed using ANCOVA to assess whether there was a statistically significant difference between them. To determine whether the ANCOVA assumptions were met or not, the homogeneity of variances and the equality of intragroup regression coefficients (F = 2.294, p > .05) were tested and found to meet the assumptions. The adjusted average posttest scores of the groups were calculated as $\bar{x} = 156.94$ for the students in the experimental group and $\bar{x}$ = 144.56 for the students in the control group. The average pretest and posttest scores of the groups are shown in Figure 1.

Figure 1.

Average pretest and posttest results for each group’s Transactional Distance Scale scores.

The results of the ANCOVA used to test whether there was a significant difference between posttest scores are given in Table 2.

Table 2.

Results of Covariance Analysis Related to Students’ Transactional Distance Scores Based on Assigned Roles.

Source	Sum of squares	df	Mean square	F	p
Pretest	26,940.030	1	26,940.030	75.612	.000
Group	4,193.037	1	4,193.037	11.769	.001
Error	38,479.412	108	356.291
Total	2,593,753.000	111

When Table 2 was analyzed and the pretest scores of the experimental and control are taken into consideration, it was observed that the difference in the adjusted posttest scores of the groups was significant, F(1, 108) = 11.769, p < .01. In other words, it was observed that the adjusted posttest scores of the students in the experimental group ( $\bar{x}$ = 156.94) were significantly higher than the adjusted posttest scores of the students in the control group ( $\bar{x}$ = 144.56).

Findings Related to Knowledge Sharing Behaviors

The second research question aimed to find whether there was a statistically significant difference between the Knowledge Sharing Behavior Scale scores of the students in the experimental and control groups. The students’ pretest and posttext scores were analyzed using ANCOVA to determine whether there was a statistically significant difference between them. To determine whether the ANCOVA assumptions were met or not, the homogeneity of variances and the equality of intragroup regression coefficients (F = 0.033, p > .05) were tested and found to meet the assumptions. The adjusted posttest average scores of the groups were calculated as $\bar{x}$ = 86.49 for the students in the experimental group and $\bar{x}$ = 78.14 for the students in the control group. The average pretest and posttest scores of the groups are shown in Figure 2.

Figure 2.

Average pretest and posttest results for each group’s Knowledge Sharing Behavior Scale scores.

Results of the ANCOVA used to test whether there was a significant difference between the posttest scores of the groups are given in Table 3.

Table 3.

Results of Covariance Analysis Related to Students’ Knowledge Sharing Behavior Scores based on Assigned Roles.

Source	Sum of squares	df	Mean square	F	p
Pretest	10,647.834	1	10,647.834	90.715	.000
Group	1,920.738	1	1,920.738	16.364	.000
Error	12,676.715	108	117.377
Total	773,596.000	111

When Table 3 was analyzed and the pretest scores of both groups were considered, it was observed that the difference between the adjusted posttest average scores of the groups was significant, F(1, 108) = 16.364, p < .01. In other words, it was observed that the adjusted posttest scores of the students in the experimental group ( $\bar{x}$ = 86.49) were significantly higher than the adjusted posttest scores of the students in the control group ( $\bar{x}$ = 78.14).

Student Opinions Related to Structured and Unstructured Online Discussions

When all the students’ responses on the Student Opinion Determination Form were assessed, both negative and positive subthemes emerged as a result of the content analysis. These themes are evaluated later.

With regard to the online discussions, students in the experimental group stated that they provided some advantages. These included the opportunity to question badly expressed ideas, get feedback, learn new things, become aware of topics that had been wrongly or inadequately learned, participate when and where they wanted, and express themselves freely. Similar responses were given by students in the control group. Some student opinions are as follows:

S1 (Experimental group): It helped me participate in the lesson more actively and comprehend better by discussing the topics.

S2 (Experimental group): It increased my interest in the lesson and helped me understand it better.

S1 (Control group): We have seen what knowledge we lack and learned the details we did not know.

S2 (Control group): It encouraged us to discuss in online platforms and enabled active participation in the discussions. Here we could discuss the points we hadn’t been able to discuss in face-to-face classroom environment.

Unlike the responses from the control group, students in the experimental group stated that being given roles and responsibilities made the discussion period meaningful, effective, and fun. They also stated that having roles required them to be careful during the discussions, increased cooperation and communication between group members, made them more responsible, and enabled them to think in more versatile ways and come up with different solutions. Despite this, some students in experimental group stated that having roles and responsibilities could be an obstacle to learning about matters which were the responsibility of others, led to discussions becoming banal after a while and also decreased curiosity. Students in the control group also expressed the negative opinions that their interest and participation in the discussions decreased over time, that meaningful learning of new information did not occur, that there were no original ideas, and that discussions sometimes went off track. Students also stated that some dominant students tried to control the online discussions and that this was among the problems encountered. Some student opinions are as follows:

S1 (Experimental group): We had a particular order in the discussions depending on the roles we had. Everyone’s role and responsibility was determined, and it was an advantage for us to continue without deranging.

S2 (Experimental group): Everyone knew about how each other worked and how much they were interested.

S3 (Experimental group): All the powers and responsibilities were equally on each group member, and it facilitated our disciplined and organized work.

When the students’ views related to transactional distance perception were evaluated, students in both groups stated that the simple and useful interface, and advanced information and file-sharing features of the forum, increased student–interface and student–environment interactions. In addition, students in the experimental group stated that having specific roles and responsibilities increased student–student and student–content interactions and made them more meaningful. In this respect, it can be said that the structured discussions brought about a difference in the subdimensions of student–student and student–content interactions in terms of transactional distance perception. Some student opinions are as follows:

S1 (Experimental group): Questions asked and contents shared in discussions were more qualified because everyone had a responsibility. I can say that forum environment improved our communication among friends.

S2 (Experimental group): We could hold more comprehensive discussions due to the fact that everyone had a role and responsibility in the discussion process. Interaction between students was high.

S1 (Control group): Everyone could express themselves adequately thanks to the discussions and get help by asking when they couldn’t understand the topics.

S2 (Control group): The discussion platform was user-friendly and had a simple interface so I could use it comfortably.

When the views of the students about knowledge sharing behavior were compared, students in both groups stated that group discussions increased knowledge sharing. However, while the students in the experimental group stated that the process of sharing knowledge was meaningful and productive, students in the control group stated that sharing knowledge was not very meaningful and that knowledge was shared in a repetitive way. Furthermore, participation in discussions and student interest was greater in the students in the control group. In this respect, when the qualitative findings were evaluated, for the Knowledge Sharing Behavior Scale in particular, there was a difference in the subdimension of the usefulness or suitability of the discussions. Some student opinions are as follows:

S1 (Experimental group): As everyone had particular roles and responsibilities in online discussions, each question or request was responded in a fast and qualified way. In this way, a very good knowledge sharing setting was created.

S2 (Experimental group): As everyone’s role was determined, they exercised due care. Therefore, I was very pleased. I think there was mutual respect, and it increased our participation in the process.

S1 (Control group): I can say that discussions were a bit superficial, we could not go into details. There were mostly repetitions.

S2 (Control group): The discussion setting enabled a social environment where we could share everything comfortably. We could communicate with our friends easily and exchange information.

Discussion and Conclusion

This study was carried out to determine the effect of assigning roles (starter, moderator, arguer, source searcher, and summarizer) to the group members in online discussions on their transactional distance perception and knowledge sharing behavior. The findings indicate that assigning roles is an effective strategy on students’ transactional distance perception and knowledge sharing behavior.

The findings obtained from the first subproblem of the research show that structured discussion, in which different roles and responsibilities are given to the students, is a useful strategy in decreasing transactional distance perceptions. The qualitative findings indicate that giving various roles and responsibilities to the students leads to meaningful interactions and cooperation especially in student–student and student–content interactions. The transactional distance perception of the students subsequently decreases. Another strategy that Rovai (2007) suggests for increasing the quality of online discussions is to develop student–student interactions. It is, however, necessary to develop them using the proper strategy (Rovai, 2007). Otherwise, students begin to have meaningless discussions with each other, and this may cause an increase in transactional distance perception over time. The results of this research show that discussions structured by assigning various roles and responsibilities to the group members increases student–student and student–content interactions and therefore decreases the transactional distance perception. Xie et al.’s (2014) research confirms this result and shows that assigning roles is a strategy that can be used to increase interactions among participants. Similar research shows that assigning roles improves interaction and cooperation (S. J. Lin & Crawford, 2007; Marttunen & Laurinen, 2009; Schellens, Van Keer, De Wever, & Valcke, 2007; Strijbos, Martens, Jochems, & Broers, 2004; Weinberger, Ertl, Fischer, & Mandl, 2005).

When the results related to transactional distance are evaluated from in terms of roles, the starter, arguer, and summarizer roles are seen to act as catalysts for the development of student–student and student–student interaction because these roles are directly related to these kinds of interactions. As a matter of fact, the research shows that assigning roles increases participation in debates, which leads students to think more deeply about the content and to better analyze the issues (Salter & Conneely, 2015). This can reduce transactional distance perception by improving student–student and student–content interactions. The moderator, in coordinating the group, and the source searcher, in finding additional resources and sharing them in order to deepen the discussion, aid the development of student–student and student–content interactions. In this study, the role of the teacher as an experimental variable was limited to following the online discussions. In future applications, the teacher could also contribute to the development of student–teacher interaction by effectively participating, thereby achieving a further reduction in transactional distance perception.

The findings show that the knowledge sharing behaviors of the students in the structured online discussions were more developed. The results of qualitative data analysis show that the students in the structured group gave positive feedback for the effectiveness or convenience subdimension in particular. From these findings, it can be said that having discussions in which specific roles are given to the participants develops a community’s knowledge sharing behaviors. These findings correlate with the findings in the literature. For example, in the study by De Wever et al. (2010), assigning roles to the members in discussion groups was beneficial both in starting and in sustaining the discussion. Similarly, in the study of Salter and Conneely (2015), structured online discussions were effective in promoting student engagement. And in the research of Gu et al. (2015), structured discussions in which roles were assigned to the members increased the development of group cognition, collaborative problem-solving, and joint collaborative activities. In their study, De Wever, Van Keer, Schellens, and Valcke (2009) stated that giving roles to the students positively influenced the process of the social structuring of information, but that after a while this influence was no longer as effective. Conversely, according to the study of Fitzgerald, Mitchem, Hollingsead, Miller, and Koury (2005), the more structured discussions were, the more situation-oriented and shorter they would be. In this study, after the requirements for participation were identified, it was observed that some of the students only met those requirements and did not participate in all discussions. To solve this problem, Salter and Conneely (2015) stated that reassigning roles over time could make discussions more attractive and increase participation.

When the results for knowledge sharing behaviors are evaluated in terms of roles, it seems that the starter contributed in initiating knowledge sharing behaviors in the community, while the arguer contributed to knowledge sharing behaviors by providing detailed and sustainable knowledge related to the discussions. The role and responsibilities of each member in a structured discussion group require each member to “make a sacrifice” and undertake efforts to develop the performance of the group as a whole (Strijbos et al., 2007). In this sense, the responsibility that members assume for their roles may have contributed to the development of the sacrifice subdimension of the Knowledge Sharing Behaviors Scale. The presence of the moderator in the debate required the group members to be more respectful to each other, to use appropriate language, and to be polite. This may have contributed to the development of the respect subdimension. In her research, Gray (2004) concluded that the presence of a moderator in the community helped knowledge be shared. The role of source searcher in improving knowledge sharing is thought to work especially in the subdimension of trust because the content that the group members’ share is found by the source searcher and is structured according to the knowledge of the group members. It is believed that knowledge constructed through discussions contributes to the development of trust in the knowledge of the community and between community members. In their study, De Wever et al. (2010) concluded that the role of the summarizer made the biggest contribution to the process of structuring knowledge. The important points noted down by the summarizer at the end of the discussion about what has been learnt were seen as important by the members of the online discussion group. This may have contributed to the development of the usefulness or compliance subdimension of the Knowledge Sharing Behaviors Scale, leading to the online discussions becoming useful or relevant to group members.

The research shows that assigning roles is particularly valuable during at the beginning of a discussion. However, the effectiveness of the assigning roles can be reduced toward the end of the discussion (De Wever et al., 2010). Therefore, in time, transactional distance perception could increase and knowledge sharing behaviors could decrease in online discussions. In such cases, teachers should try to resolve these problems by, for example, providing metacognitive support to discussions and by applying methods such as reassigning group members’ roles on a weekly basis. This will contribute to keeping the group dynamics active and deepening the debate. Yilmaz and Keser (2017) have found that providing metacognitive support in an asynchronous discussion process helps to reduce transactional distance perception. In their study, Salter and Conneely (2015) concluded that teachers’ participation also increased students’ participation. Accordingly, it can be said that teachers’ participation in the discussion will also support the development of knowledge sharing behaviors.

Based on the findings from this study, the following recommendations can be made for the design of online discussions. Assigning different roles and responsibilities to the participants of online discussions will help solving problems related to transactional distance perception and knowledge sharing behaviors. Changing the roles and responsibilities of the participants may allow increased participation and make the discussions more attractive. Except for the roles and responsibilities, structuring the discussion by putting such criteria as participation rules, assessment criteria could diminish learners’ autonomy. This may cause a decrease in the quality of the discussions. Therefore, structuring the discussion should be made to a certain degree and should enable flexibility. The teacher should neither be the focus of much of the discussion nor be completely outside. The teacher should make his or her presence felt by following the flow of discussion, guiding the discussion at necessary points, and leading it where appropriate. To improve the quality of knowledge sharing and also reduce transactional distance perception, including reflective thinking activities and practices that will develop high-order thinking skills will be useful.

There are certain limitations of this research. As this is an experimental study, it was carried out on the limited number of students. Subsequent research could be performed with a larger sample and with groups of different educational experiences and age. This study was carried out during classes in which blended learning was used. This study can be repeated in classes which are completely carried out online, and the results from both studies can be compared. This research determined the effect of structuring on transactional distance perception and knowledge sharing behaviors through pretest and posttest questionnaires. In future studies, assessment formats such as peer assessment and self-assessment could be used to determine the effect of structured discussions. The research compared groups that were unstructured with groups that were structured. In the future, experimental studies could be conducted to determine the effectiveness of each role in the structured groups. By applying the research design to synchronous online discussions, results could be compared and studies could be conducted to determine the ideal research models. Also, wall analysis can be to used make the discussion comments. Today, such Social Network Sites as Whatsapp, Facebook, and Twitter are widely used to create virtual communities of practice and to make knowledge sharing (Çetinkaya & Sütçü, 2018; Yilmaz, 2016). In this context, studies toward determining the impact of different discussion environments on transactional distance and knowledge sharing behaviors can be carried out.

This study examined the impact of assigning roles (starter, moderator, arguer, source searcher, and summarizer) to the group members in online discussions on their transactional distance perceptions and knowledge sharing behaviors. The results indicated that assigning roles in online discussions reduced transactional distance perception and improved knowledge sharing behaviors of the students in the discussion. This study attempted to facilitate future research and practical applications involving online discussions. It is expected that the results from this study will contribute to increasing the efficiency of online discussions both within and outside the classroom and to guide the practitioners.

Footnotes

Declaration of Conflicting Interests

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

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Ramazan Yilmaz is working in the Department of Computer Education and Instructional Technology (CEIT) at Bartin University in Turkey. He received BS degree from the Faculty of Education, Department of CEIT, Hacettepe University, in 2007. In 2010, he received MA degree in CEIT from the Gazi University. In 2014, he received PhD degree in CEIT from the Ankara University. He is interested in instructional design, online learning, community of practice, multimedia learning, sociopsychological variables, eye-tracking, human–computer interaction, and pedagogical agents. He has published articles in various journals such as Computers & Education, Computers in Human Behavior, Interactive Learning Environments, Journal of Educational Computing Research, Internet Research, Journal of Computing in Higher Education.

Fatma Gizem Karaoglan Yilmaz is working in the Department of Computer Education and Instructional Technology (CEIT) at Bartin University in Turkey. She is the head of this department. She received BS degree from the Ankara University, Faculty of Educational Sciences, Department of CEIT in 2008. In 2010, she received MA degree in CEIT from the Cukurova University. In 2014, she received PhD degree in CEIT from the Ankara University. She is interested in distance education, online learning, eye-tracking, interactive learning environments, and human–computer interaction. She has published articles in various journals such as Computers & Education, Computers in Human Behavior, Interactive Learning Environments, Journal of Educational Computing Research, Internet Research.

References

Alakurt, T. (2013). An analysis of the factors influencing knowledge sharing behaviours between members of virtual communities of practice (Doctoral dissertation). Ankara University, Turkey.

Çetinkaya

Sütçü

S. S.

(2018) The effects of Facebook and WhatsApp on success in English vocabulary instruction. Journal of Computer Assisted Learning. doi:10.1111/jcal.12255.

De Wever

Schellens

Van Keer

Valcke

(2008) Structuring asynchronous discussion groups by introducing roles: Do students act in line with assigned roles? Small Group Research 39(6): 770–794.

De Wever

Van Keer

Schellens

Valcke

(2009) Structuring asynchronous discussion groups: The impact of role support and self-assessment on students’ levels of knowledge construction through social negotiation. Journal of Computer Assisted Learning 25: 177–188.

De Wever

Van Keer

Schellens

Valcke

(2010) Roles as a structuring tool in online discussion groups: The differential impact of different roles on social knowledge construction. Computers in Human Behavior 26(4): 516–523.

Ekwunife-Orakwue

K. C.

Teng

T. L.

(2014) The impact of transactional distance dialogic interactions on student learning outcomes in online and blended environments. Computers & Education 78: 414–427.

Fang

Y. H.

Chiu

C. M.

(2010) In justice we trust: Exploring knowledge sharing continuance intentions in virtual communities of practice. Computers in Human Behavior 26(2): 235–246.

Fitzgerald, G., Mitchem, K., Hollingsead, C., Miller, K., & Koury, K. (2005). Enhancing case-based learning using online discussion groups: Diversity of practice. Proceedings of International Special Education Conference, Glasgow, Scotland.

Goel

Zhang

Templeton

(2012) Transactional distance revisited: Bridging face and empirical validity. Computers in Human Behavior 28(4): 1122–1129.

10.

Gray

(2004) Informal learning in an online community of practice. Journal of Distance Education 19(1): 20–35.

11.

Shao

Guo

Lim

C. P.

(2015) Designing a role structure to engage students in computer-supported collaborative learning. The Internet and Higher Education 24: 13–20.

12.

M. L. C.

Rappa

N. A.

Chee

Y. S.

(2009) Designing and implementing virtual enactive role-play and structured argumentation: Promises and pitfalls. Computer Assisted Language Learning 22(5): 381–408.

13.

Hollingsead, C., Mitchem, K., Koury, K. A., Miller, K. J., Fitzgerald, G., & Tsai, H. (2006). Asynchronous online discussions within multimedia case based instruction. Proceedings of E-Learn 2006—World Conference on E-Learning in Corporate, Government, Healthcare, and Higher Education, Honolulu, HI.

14.

Karaoglan Yilmaz

F. G.

(2017a) Predictors of community of inquiry in a flipped classroom model. Journal of Educational Technology Systems 46(1): 87–102.

15.

Karaoglan Yilmaz

F. G.

(2017b) Social presence and transactional distance as an antecedent to knowledge sharing in virtual learning communities. Journal of Educational Computing Research 55(6): 844–864.

16.

Lin

M. J. J.

Hung

S. W.

Chen

C. J.

(2009) Fostering the determinants of knowledge sharing in professional virtual communities. Computers in Human Behavior 25(4): 929–939.

17.

Lin

S. J.

Crawford

(2007) Innovations in teaching: An online debate series for first-year pharmacy students. American Journal of Pharmaceutical Education 71(1): 1–8.

18.

Marttunen

Laurinen

(2009) Secondary school students’ collaboration during dyadic debates face-to-face and through computer chat. Computers in Human Behavior 25(4): 961–969.

19.

Miles

M. B.

Huberman

A. M.

(1994) Qualitative data analysis: An expanded sourcebook (2nd ed.), Thousand Oaks, CA: SAGE.

20.

Moore

M. G.

Kearsley

(2011) Distance education: A systems view of online learning, Boston, MA: Cengage Learning.

21.

Murphy, E., & Coleman, E. (2004). Graduate students’ experiences of challenges in online asynchronous discussions. Canadian Journal of Learning and Technology, 30(2). Retrieved from https://www.cjlt.ca/index.php/cjlt/article/view/26522/19704.

22.

C. S. L.

Cheung

W. S.

Hew

K. F.

(2009) Sustaining asynchronous online discussions: Contributing factors and peer facilitation techniques. Journal of Educational Computing Research 41(4): 477–511.

23.

Prensky

(2001) Digital natives, digital immigrants part 1. On the Horizon 9: 1–6.

24.

Remesal

Colomina

(2013) Social presence and online collaborative small group work: A socio constructivist account. Computers & Education 60(1): 357–367.

25.

Rovai

A. P.

(2007) Facilitating online discussions effectively. The Internet and Higher Education 10(1): 77–88.

26.

Salmon

(2004) E-moderating: The key to teaching and learning online, London, England: Psychology Press.

27.

Salter

N. P.

Conneely

M. R.

(2015) Structured and unstructured discussion forums as tools for student engagement. Computers in Human Behavior 46: 18–25.

28.

Schellens

Van Keer

De Wever

Valcke

(2007) Scripting by assigning roles: Does it improve knowledge construction in asynchronous discussion groups? International Journal of Computer-Supported Collaborative Learning 2(2–3): 225–246.

29.

Strijbos

J. W.

Martens

R. L.

Jochems

W. M. G.

Broers

N. J.

(2004) The effect of functional roles on group efficiency: Using multilevel modeling and content analysis to investigate computer-supported collaboration in small groups. Small Group Research 35(2): 195–229.

30.

Strijbos

J. W.

Martens

R. L.

Jochems

W. M. G.

Broers

N. J.

(2007) The effect of functional roles on perceived group efficiency during computer-supported collaborative learning: A matter of triangulation. Computers in Human Behavior 23(1): 353–380.

31.

Vygotsky

L. S.

(1988) Thought and language, Cambridge, MA: MIT Press.

32.

Weinberger

Ertl

Fischer

Mandl

(2005) Epistemic and social scripts in computer-supported collaborative learning. Instructional Science 33(1): 1–30.

33.

Wengrowicz

(2014) Teachers’ pedagogical change mechanism–pattern of structural relations between teachers’ pedagogical characteristics and teachers’ perceptions of transactional distance (TTD) in different teaching environments. Computers & Education 76: 190–198.

34.

Xie

Bradshaw

A. C.

(2014) Impacts of role assignment and participation in asynchronous discussions in college-level online classes. The Internet and Higher Education 20: 10–19.

35.

Yilmaz, R. (2014). The effect of interaction environment and metacognitive guidance in online learning on academic success, metacognitive awareness and transactional distance (Doctoral dissertation). Ankara University, Ankara, Turkey.

36.

Yilmaz

(2016) Knowledge sharing behaviors in e-learning community: Exploring the role of academic self-efficacy and sense of community. Computers in Human Behavior 63: 373–382.

37.

Yilmaz

Keser

(2015) The adaptation study of Transactional Distance Scale. Hacettepe University Journal of Education 30(4): 91–105.

38.

Yilmaz

Keser

(2017) The impact of interactive environment and metacognitive support on academic achievement and transactional distance in online learning. Journal of Educational Computing Research 55(1): 95–122.

39.

Zhang, A. (2003). Transactional distance in web-based college learning environments: Toward measurement and theory construction (Doctoral dissertation). Virginia Commonwealth University, Richmond, VA.