Abstract
This experimental study within the scope of Anadolu University Open Education Faculty in Turkey aims to examine the effect of online group discussion techniques. In this context, the online group discussion technique, which is applied by dividing the learners into manageable groups in online learning environments, has been examined within the scope of intrinsic motivation, social presence, and perceived learning variables. Sequential explanatory mixed method design was used to conduct this study. The online group discussion process was applied in the study’s experimental group by dividing the learners into manageable groups. In this process, a field expert accompanied each group. In the control group, the discussion process was left to its natural state, and all learners who took the relevant course were included in a single group. Before and after the experimental process, data were collected from 250 learners through the study scales, and then five learners were interviewed. The findings obtained within the study’s scope show that the design to be applied in the online group discussion technique used in online learning environments is important in increasing the efficiency to be obtained from the process. In this context, it has been observed that to manage online learning processes effectively, attention should be paid to the fact that the groups must be less than 150 learners to not show mass features. Online group discussions in groups that do not show mass features positively affected the learners’ intrinsic motivation, social presence, and perceived learning levels. However, acting in line with the principles of social constructivism theory and the community of inquiry theory in learning processes has increased the discussions’ efficiency. Besides, it has been observed that the increase in the social presence and intrinsic motivation of the learners positively affects their perceived learning levels.
Keywords
Online learning environments provide active and attractive activities for learners by providing opportunities for knowledge generation rather than learners’ exposure to knowledge transfer (Horzum et al., 2015). Furthermore, these environments are used to ensure that learners can easily access learning environments, to allow introverted learners to participate in learning processes, to enable learners to express themselves, and to encourage dialogue and communication among learners, instructors, and groups (Keppell et al., 2011).
Studies on online learning environments, which is one of the important elements those educational institutions invest in, emphasize that learner communication and cooperation in these environments are important in meeting learners’ social needs (Chau and Xu, 2012). In online learning environments where collaboration is important, learners experience loneliness experienced in the learning process and the dropout problem that develops accordingly (Kotturi et al., 2015). Learners’ sense of loneliness also decreases their social presence and learning outcomes (Meyer, 2004). This situation is seen as a problem that will reduce the benefit of learners from learning environments. It is considered a crucial factor in eliminating this sense of loneliness felt in online learning environments by increasing solidarity among learners and increasing the level of social presence (Kotturi et al., 2015). When learners who participate in online learning environments collaborate by increasing the solidarity between them and other learners in the environment, their motivation levels increase. As a result, they become more actively involved in the learning process and have richer learning outcomes (Jensen, 2017). Therefore, the application of teaching techniques that will increase the level of social presence of learners who participate in online environments is a crucial factor in preventing learners from feeling lonely, increasing their motivation level, and consequently increasing their academic achievement. From this point of view, it is thought that the online group discussion technique, which allows learners to work collaboratively and gain richer learning outcomes in online environments, will play an effective role in remedying these problems.
Although it differs according to the course and the subject of the course, online group discussions generally enable learners to discuss their points of view with more knowledgeable people, regardless of time and place and create meaning in the light of criticism from their perspectives. According to the social constructivist learning theory, learners should create meaning from their new experiences with active participation, interaction, and dialogue and be involved in the knowledge structuring process (Jonassen et al., 1995). According to this theory, meaning develops when the individual tries to interpret the stimulus. Therefore, it is very important to design learning environments where learners can discuss their thoughts, experiences, and perceptions with other learners and teachers (Pena-Shaff and Nicholls, 2004). According to Vygotsky (1967), social and cognitive development first takes place on the social plane where the learner and teacher interact. In this process, dialogue functions as a thinking tool (Vygotsky, 1967). This situation is because cognitive processes such as those involving clarification, elaboration, self-defense, and the integration of opinions and thoughts become operational through dialogue (Jonassen et al., 1995). In other words, the creation of meaning happens as long as thoughts are expressed, reflected, and discussed.
The use of online group discussions based on written communication, which is a part of the transformation created by the constructivist approach in education, is gradually increasing (Akcaoglu and Lee, 2016). Thus, the learners are enabled to discuss their perspectives with other learners and trainers regardless of time and place, examine different perspectives, and create meaning in light of the criticisms coming from their point of view. In learning processes with online group discussions, which are an important part of online learning, the following are provided (1) facilitating access to information, (2) increasing learner success, (3) providing easy communication, (4) increasing interactions between online learners (Chau and Xu, 2012).
Related literature
One of the aims determined within the research scope is to examine the effect of online group discussion techniques on learners’ intrinsic motivation. The structure of intrinsic motivation that affects interaction and permanence in learning in online learning environments includes mastery, self-interest, and research-oriented elements that are very necessary for cognitive and social development and represents the main source of joy and vitality throughout life (Firat et al., 2018). Furthermore, feedback, communication, and fairness are factors that positively affect intrinsic motivation (Ryan and Deci, 2000). Therefore, employing communication elements, giving effective feedback, and fair evaluation facilitates the formation of intrinsic motivation (Ryan and Deci, 2000). In addition, the study conducted by Ryan and Deci (2000) shows that not only tangible gains but also threats, limited timeframes, obligatory orders, and strict evaluations reduce intrinsic motivation. Studies conducted in line with these results (Deci et al., 1981; Flink et al., 1990; Martens et al., 2004) have shown that giving importance to learners ‘autonomy in learning processes improves learners’ more intrinsic motivation, curiosity, and desire to challenge.
Another aim of the research was to examine how the online group discussion technique applied throughout the process affects social presence. Since the concept of social presence (Garrison and Akyol, 2013), which emphasizes the need to encourage social interaction, which is a tool to guide learners to critical thinking and high-level learning, is directly related to online learning processes, Researchers and practitioners are investigating not only what social presence is, but also the roles it plays in online learning processes (Annand, 2011; Oztok and Brett, 2011). With these studies, it was concluded that social presence affects various factors in learners’ learning experiences. According to these results, a social presence can positively affect the learners’ participation in learning processes and their motivation to participate (Jorge, 2010), satisfaction with the learning process (Strong et al., 2012), perceived learning (Joksimović et al., 2015), academic success (Joksimović et al., 2015) and their level of interaction (Reio and Crim, 2013). Like these results, in the study conducted by Richardson and Swan (2003), it was concluded that learners who have a high level of social presence perception in an online learning process are both more satisfied with the learning process, and their perceived learning levels are higher than those who have a low social presence. In online learning environments where social interactions are at the center of the learning process (Sung and Mayer, 2012), the groups’ size can affect learners’ participation and their communication and interaction with other learners (Aragon, 2003). In line with this situation, Akcaoglu and Lee (2016) concluded in their study that group size harms group cohesion and learner learning. Therefore, dialogues between learners in large groups are insufficient, and learners argue less with each other (Hamann et al., 2012). On the other hand, smaller groups provide more opportunities for learners to interact and interact and participate in learning processes more actively (Lowry et al., 2006). This situation is seen as a factor that will increase the social presence levels by encouraging a sense of community and harmony among learners (Rovai, 2002). In this context, the number 150 (Dunbar, 1992), which is determined as the maximum number of individuals who can interact in online learning environments composed of large masses, should be considered in determining the number of groups in these environments.
The last aim determined in the research process was to examine how the online group discussion technique affects perceived learning. Perceived learning, which is the basis of current learning beliefs and emotions and is based on learners’ self-assessment in learning processes where learners are at the center (Horzum et al., 2015), is independent of learners’ scores from exams. Perceived learning, which is independent of the test scores applied in learning processes and is most prominent in online learning environments (Glass and Sue, 2008), is associated with other variables in learning processes (Rockinson-Szapkiw et al., 2016). In one of the studies conducted on this subject, Baturay (2011) states that learners’ perceived learning levels are very strong in their satisfaction with the learning process and the learner-content interactions. A study by Trespalacios and Perkins (2016), shows that there is a positive correlation between perceived learning and awareness of being in a community. In addition, designs that allow interaction elements to occur in online learning environments also increase learners’ perceived learning levels (Garrison, 2007). Similarly, to this situation, designs that enable instructors to engage with learners in online environments have a high degree of interest in learning and positively affect perceived learning (Costley and Lange, 2016). In other studies, conducted on perceived learning, Richardson and Swan (2003) and Wighting et al. (2013) stated that as learners’ social presence levels increase, their perceived learning levels also increase. However, according to Ferreira et al. (2011), there is a positive and high relationship between learners’ motivation towards learning processes and perceived learning. In addition to these factors, the perceived learning levels of the learners are also related to variables such as the duration of the course (Ferguson and DeFelice, 2010), participation in discussions (Swan, 2001), and readiness for the tools used in learning processes (Haverila, 2010).
Importance and aim of the study
Although many studies are supporting the use of online group discussions (Hampton et al., 2017; Hudson, 2014; Murphy et al., 2013), which makes learning processes more efficient (Chau and Xu, 2012), the lack of sufficient studies on the situations in which this technique supports learners (Hudson, 2014; Smith, 2008) and which protocols and criteria should be used (Orton-Johnson, 2007) generate the research question of this study. In addition to this, the limited number of experimental studies on online group discussions in the relevant literature (Gokhale and Machina, 2018; Hudson, 2014) stands out as a shortcoming in the literature. Besides, the fact that most of the learners in online learning environments do not use discussion forums, and the learners who use them are not satisfied with the discussion forums, in general, generates another research question for this study. How the online group discussion technique should be used in online learning environments formed by large masses and how this technique can be effective stands out as a field of study that needs to be researched. Therefore, there is a need for experimental studies to make up for these deficiencies determined to contribute to filling the existing gap in the literature. In this context, it is thought that this study will make significant contributions to the relevant literature.
The aim of this study is to examine the effect of the online group discussion technique applied by dividing learners into manageable groups in online learning environments on learners in terms of various variables. At this point, it is considered that the manageable groups should consist of a maximum of 150 participants to not show massiveness in online learning environments (Dunbar, 1992). In other words, separating the online discussion process into manageable groups brings the necessity to form small groups. Considering all this information, answers to the question “Does the online group discussion technique applied by separating learners into manageable groups in online learning environments affect learners’ intrinsic motivation, social presence, and perceived learning levels?” is sought.
Method
In this study, mixed-method research was used to obtain more reliable results. The basic assumption of mixed-method research is that using quantitative and qualitative methods together instead of using a single method will better understand the research problem (Creswell, 2012).
Research design
In this study, the sequential explanatory mixed method design was used
Quantitative dimension
In this section, the quantitative dimension of the research is explained.
Population and sample
During this study, the experiment was conducted within the scope of the “Human and Society” course in the Sociology Undergraduate Program of Anadolu University Open Education Faculty and taught in the spring semester of the 2018–2019 academic year. The population of this research consists of 8.254 Open Education Faculty learners who take the related course. The study sample consists of learners who answered these measurement tools and volunteered to participate in the study. In this respect, it can be said that appropriate sampling is used among non-random sampling methods. In this sampling method, the researcher collaborates with individuals who are easy to reach and volunteers for the study. Three hundred and sixty (360) learners responded to the measurement tools presented as a pre-test to 8.254 learners within the study’s scope. Within the study’s scope, the number of learners who responded to the measurement tools applied as a post-test after the experiment was performed during a period was 286. As a result of clearing the extreme data, this number dropped to 260. Then, looking at learning analytics, ten more learners who responded to the final tests but never entered the discussion forums were removed from the list, and the final number of the sample was determined to be 250. Ninety-seven of the participants included in the human and society course in the second year of their education, which is in the compulsory course category, are women, and 153 are men. However, 46 of the participants are in the 15–24 age range, 175 of the participants are in the 25–54 age range, and 29 of the participants are in the 55–64 age range. The assignment process to the experimental and control groups was done randomly. Random assignment can be divided into two as restricted and unrestricted. The unrestricted method was used to ensure non-randomness. For this purpose, the assignment process was made with the help of the “Select Cases” command in the SPSS 23 package program, according to the procedure explained by Altman and Bland (1999). During the “Select Cases” process, the number of people in the experimental and control groups was entered into the SPSS software. Later, SPSS 23 software randomly assigned the numbers 0 or 1 to the participants. Within the scope of this study, those that were written 0 formed the experimental group, and those that were written 1 constituted the control group. Finally, the distribution of 250 learners was 129 experimental groups and 121 control groups.
Research model
In this research, experimental models were used in accordance with the aim of the research. The research’s independent variable is the teaching technique used, and the dependent variables are the learners’ intrinsic motivation, social presence, and perceived learning level. Perceived learning, the other dependent variable of the research, was measured at the end of the learning process due to its nature. For this reason, in the process of determining the perceived learning level, a post-test control group design, one of the real experimental models, was used. Data is collected only after the test from the individuals randomly assigned to the experimental and control groups in this design. The perceived learning scale, like other scales, was prepared as Google Form and presented to participants. For this purpose, the social presence and intrinsic motivation of the groups that were previously formed randomly were measured, as well as their perceived learning levels.
Experiment process
When the literature is examined, it is seen that 150 participants are a critical value in online environments, and groups above this value have the characteristic of massiveness (Downes, 2013). Marques (2013) stated that more than the number of people who can be contacted in a learning process is massiveness. In line with this view, Dunbar (1992) stated in his study that 150 is the maximum person an individual can communicate with. In this direction, the experimental group in which the experiment will be conducted during the study process was divided into forum groups consisting of a maximum of 150 learners. Learners were randomly assigned to these subgroups.
In this process, 186 learners in the control group consisted of a single group as currently applied within the Anadolu University Open Education Faculty, while the experimental group consisted of two groups divided into groups of at most 150. After the learning process started, the experimental group, which consisted of two groups of 87 learners each, was led by a guide by a field expert. In this process, by acting in line with the principles of social constructivism and community of inquiry theories, which constitute the theoretical dimensions of the study, the learners were not given the necessary information directly. However, instead, they were guided to access the correct information themselves. In addition to separating the learners into manageable groups and accompanying each group by an expert, it is important for the continuity of the discussions to act in a way that encourages the learners to interpret rather than give direct information. In this way, learners were allowed to express themselves in a better way. Considering that learner interactions have a key place in the social constructivism and community of inquiry theory adopted within the scope of the study, it can be stated that it is important to design a discussion environment where learners can express their own thoughts comfortably and thus interact with other learners. Another factor that was paid attention to during the experiment was that the instructors preferred the daily spoken language instead of the formal spoken language. In this way, it has been observed that it is easier for learners to communicate and interact. In addition, discussion topics open to interpretation were determined while creating discussion topics. Thus, an environment was created in which learners can express their thoughts more comfortably instead of short answers. Another factor that was taken into consideration during the discussion process was informing the learners before the discussion process. In this way, information such as which subject and scope of the discussion process should be, which points should be considered in this process, and how long the discussion process will be, were provided to the learners. This situation developed the sense of belonging of the learners towards the process.
In the control group, no change was made, and the ongoing method was applied. In this method, the course instructor completes the learning process with all the learners on the discussion forum.
Learners were not obliged to participate in the discussion forums implemented within the scope of this study. They did not take notes of their participation in the discussion forums. The topic of the discussion forum consisted entirely of the units of the relevant course. During the discussion process, the instructors also published a guide that set the boundaries of the discussions. In this guide, the learners are informed that they should not go beyond the relevant discussion topic, not use slang expressions, pay attention to the style of expressing themselves, use the standard font, do the necessary readings about the topic of discussion, and not avoid expressing their thoughts and that opposing ideas should be approached with respect.
Data collection tools
Within the scope of this study, Social Presence Scale (Çakmak et al., 2014), Perceived Learning Scale (Albayrak et al., 2014), and Intrinsic Motivation Scale (Firat et al., 2018) were used to determine learners’ (1) social presence levels, (2) perceived learning levels and (3) intrinsic motivations. The permissions for the use of these scales have been obtained from the relevant authors. In addition, all the scales used in the study are scales with proven validity and reliability.
Data analysis
As a result of the analysis conducted within the study’s scope, it was seen that multivariate normality was not achieved in the data set, so separate (2X2) ANOVA was applied for social presence and intrinsic motivation instead of MANOVA. Since the difference between the experimental and control group pre-tests of the affective dimensions of intrinsic motivation and social presence was significant in the study, the pre-tests were considered a control variable, and ANCOVA was used to measure these variables. At this point, if the analyses of the same independent variable and two different dependent variables are done separately with ANOVA, the first type of error will occur (Pallant, 2007). In such cases where multivariate normality assumption could not be achieved, Bonferroni correction was used, and first type of error was tried to be controlled. For normality, kurtosis and skewness must be between −2 and +2 (Tabachnick and Fidell, 2014). Data were collected from the experimental and control groups only after the experiment to determine the perceived learning level. In this case, independent samples t-test was applied to determine whether the perceived learning levels of the individuals in the experimental and control groups differed.
For the statistical analyses obtained because of the applied t-test and ANOVA, the effect size that enables us to interpret the practical equivalent of the significance level was calculated. In the analysis, the effect sizes were interpreted according to Cohen (1988). Confirmatory factor analyses within the scope of the research were conducted with the Mplus 7.0 package software. In addition, IBM SPSS Statistics v26 package software was used for independent groups t-test and ANOVA.
Qualitative dimension
In the qualitative dimension of this study, the phenomenology method was used to reveal the learners’ views about their experiences in the online group discussion process. Phenomenology design, which is a research design that focuses on phenomena that we are aware of but do not have an in-depth and detailed understanding of, is an inquiry strategy in which the researcher tries to define the essence of human experiences about a phenomenon explained by the participants (Creswell, 2003).
Working group
To be suitable for the research aim, the criterion sampling method, one of the purposeful sampling methods, was preferred. The basic understanding of the criterion sampling method is to study all situations that meet a predetermined set of criteria (Patton, 2002). In this study in which the criterion sampling method was used. At this stage, the participants were determined according to the average scores they obtained due on the scales they answered. At this point, five participants with the largest scale average and five participants with the lowest scale average contributed to the study. At the end of the study, the learners who had the highest and the lowest mean scale scores and wanted to participate in the study voluntarily were ranked according to their scores, and these learners were tried to be studied respectively. Until the lists of 5 people were completed, it was tried to reach the learners one by one. When the desired number of learners was reached, the part of the analysis of qualitative data started.
Data collection tool
To learn the learners’ views participating in the study about their experiences at the end of the learning process, constructed interview questions were asked to the learners. Before the interviews, the constructed interview form’s content validity was obtained by taking the opinions of three experts working on qualitative research methods, and the last version was given. Within this scope, attention was paid to the learners’ questions directed to the variables measured in the quantitative part of the study. The structured interview form, which was finalized in line with field experts’ opinions, consists of 3 open-ended questions. These are: (1) Could you please explain your thoughts on the reflections of the discussions you had in the discussion forum you attended as part of the human and society lesson on your learning? (2) What do you think about the reflections of your experiences in the discussion forum during the learning process on your communication with other learners? (3) What has changed in your desire to participate in open and distance learning environments due to your time in the discussion forum?
Data analysis
The content analysis method was used to analyze the data. The content analysis aims to reach the concepts and relationships that can explain the collected data. For this purpose, the data collected must first be conceptualized, then organized logically according to the emerging concepts, and the themes explaining the data must be determined accordingly (Patton, 2002). Content analysis to be conducted within the study’s scope was conducted using NVIVO 12 qualitative data analysis program. The codes were determined due to the raw data’s content analysis to be obtained from the interviews. The codes obtained and the findings obtained from the scales applied to learners in the study’s quantitative dimension were integrated into the study’s conclusion and discussion section.
Findings
As the first step in this section, the scales’ factor structure applied to the participants within the study’s scope was verified. In this context, confirmatory factor analysis (CFA) was conducted for scales determined for each sub-purpose.
Validity and reliability analysis of the scales used in the study
The scales in this study were developed previously, and their validity and reliability were reported. For this reason, the construct validity of the scales used for data collection within the scope of the research was revealed by CFA. In constructive equation modeling, firstly, the available data’s compatibility with the predicted model is examined. In this direction, a chi-square (χ2) value, one of the values obtained, is interpreted as an acceptable fit if it is below 5 and a perfect fit below 2.5 (Yuan, 2005). RMSEA and SRMR values obtained from 0.05 mean perfect, and lower than 0.08 means acceptable fit, while these values are greater than 0.1 means that model-data fit is poor (Curran et al., 2003). For CFI and TLI values, the perfect fit is for values above 0.95, and a good fit is for values above 0.90 (Curran et al., 2003). Since it may not always be possible for all fit values to indicate good-acceptable model data fit when examining fit statistics, the researcher should decide after examining all values. In addition to these values, within the scope of the reliability of the scales used, the Cronbach α coefficient, which shows the reliability mostly in terms of internal consistency, was obtained and reported.
Validity and reliability analyses were carried out before, and no problem was encountered regarding these three scales, and validity and reliability analyses were carried out within the scope of this study. Details of these analyses are presented in order.
Intrinsic motivation scale
Fit Indices of the intrinsic motivation scale.
When the values in Table 1 are examined, it is seen that the χ2/SD value shows a bad fit before the modification. Therefore, because of the first analysis, the relationship between the error variances of the 2nd and 3rd items, which is the modification suggested by the program, as defined since it could be explained theoretically; and the values in the second row of Table 1 were obtained. As a result of these values, it is seen that χ2/SD value indicates good model-data fit after modification (2.75). While the SRMR value indicates an excellent fit both before modification (0.031.95) and after modification (0.99 > 0.95), a TLI value of 0.93 before modification (0.93 > 0.90) shows a good fit, and a TLI value of 0.97 after modification shows a perfect fit (0.97 > 0.95). The RMSEA value obtained as 0.141 indicated poor model-data fit before modification, while it was close to good fit after modification, but it showed a moderate acceptable model-data fit (0.083).
Along with the statistics obtained because of CFA applied to the intrinsic motivation scale, reliability, mean score, and standard deviation of the scale items were also reported. It is seen that the factor loads of the intrinsic motivation scale vary between 0.53 and 0.89, and these values are statistically significant at the p<.001 level. Within this scope, it can be interpreted that factor loadings are quite high.
The Cronbach α coefficient obtained to determine the intrinsic motivation scale’s reliability in terms of internal consistency is 0.87. Considering that the lower limit for the Cronbach α coefficient is 0.70 (Takavol and Dennick, 2011), it was observed that the reliability of the intrinsic motivation scale in terms of internal consistency was ensured. Considering all these data and comments, it can be said that the fit indices between the one-dimensional model and the data are at an acceptable level.
Social presence scale
Fit Indices of the social presence scale.
When the values obtained before the modification in Table 2 are examined, it is seen that the χ2/SD value indicates a good fit (3.03). When the other values are examined, it is seen that the RMSEA value indicates the average fit in the model data fit statistics. Therefore, because of the first analysis, the relationship between the error variances of items 14 and 17, 10 and 11, 1 and 4, which were decided to be theoretically explainable from the modifications suggested by the program, as defined, and the values in the second row of Table 2 was obtained. When the values obtained after the modification are examined, it is seen that the χ2/SD value decreases as expected, and this value approaches the perfect model-data fit (2.65). The RMSEA value, on the other hand, reached a good model-data fit value (0.080). The SRMR value was also obtained as 0.056, indicating a good model-data fit (0.056 < 0.080). In CFI and TLI values, the CFI value shows a good fit (0.91 > 0.90), while the TLI value is remarkably close to a good fit (0.89). In the light of all these values, it was concluded that the model, which CFA was applied according to the three-dimensional related factors model, has acceptable model-data fit values.
The reliability, mean score, and standard deviation of each sub-dimension of the scale were calculated together with the CFA statistics applied to the social presence scale. In addition, the mean and standard deviation values of each item were also examined. In CFA, standardized path coefficients can be interpreted as factor loadings. According to Tabachnick and Fidell (2000), factor loadings should be 0.32 and above. Therefore, it was concluded that all the factor loads obtained were significant at the 0.001 level, and these values varied between 34 and 0.83.
To examine the reliability of the social presence scale in terms of internal consistency, Cronbach α coefficient was obtained for each dimension. As a result, the said coefficients obtained were obtained as 0.75, 0.80, and 0.88 for each sub-dimension, respectively. Accordingly, it was observed that the reliability of all sub-dimensions of the social presence scale in terms of internal consistency was ensured. Considering all these results, it was concluded that the construct validity and reliability of the scale were at an acceptable level. In other words, the scale was verified.
Perceived learning scale
Fit Indices of the perceived learning scale.
No modification was needed because of the fit statistics obtained because of the CFA for the perceived learning scale. In this context, the modifications suggested by the Mplus 7 package program were not applied. When the values in Table 3 are examined, it is seen that χ2/SD value shows good model-data fit (3.04.90). Finally, in the fit statistics, the RMSEA value is moderately acceptable (0.089 < 0.1). In the light of all these values, it can be said that the model, which CFA was applied according to the three-dimensional related factors model has an acceptable model-data fit
It is seen that the factor loadings of the perceived learning scale, which do not need any modification, vary between 0.10 and 0.89. In this scale, it is seen that the factor load of item number 2 is very low and is not statistically significant. This situation shows that the relevant item does not work in the participants who participated in this study. This item with a low factor load was not included in other analyzes conducted within the study’s scope. It is seen that the factor loads of the other items of the scale have high values between 0.55 and 0.89 and are statistically significant (p < .001).
The Cronbach α coefficient obtained to determine the perceived learning scale’s reliability in terms of internal consistency was found to be 0.71, 0.81, and 0.70 in all sub-dimensions, respectively. Since these values are 0.70 and above, it was concluded that reliability in terms of internal consistency was achieved. According to the results obtained, it was concluded that the perceived learning scale was verified by removing one of the items in the original form, consisting of eight items and three dimensions.
Descriptive statistics on variables
As a result of the analysis, it was seen that multivariate normality was not provided in the data set, so separate (2X2) ANOVA was applied for social presence level and intrinsic motivation. In addition, independent groups t-test was used for perceived learning levels.
Before analyzing, it was examined whether there was a difference between the experimental and control groups’ social presence levels and the intrinsic motivation pre-tests. Accordingly, it was concluded that the experimental group’s intrinsic motivation scores were significantly higher than the control group (t (248) = 2.11, p = .036). Similarly, for the social presence scale’s emotional dimension, it was determined that the control group’s scores were significantly higher than the experimental group (t (248) = −2.16, p = .032). There is no significant difference in other tests. After this stage, it was decided to make the pre-tests control variable in the tests be made for the intrinsic motivation and social presence scale’s emotional dimension. Since parametric tests were used in the study, normality conditions were examined. In this context, all study variables show the normal distribution in both the experimental and control groups. The skewness values vary between −0.01 and −0.90, while kurtosis values vary between 0.01 and 0.88. Besides, histograms and Q-Q plots related to the variables were examined, and it was determined that the normality condition was met.
Findings regarding the social presence levels and intrinsic motivations of the experimental and control group learners after the intervention
An answer was sought to the question of whether the intrinsic motivation levels and social presence levels of the learners differed in the experimental and control groups after the intervention. Since the intrinsic motivation pre-test results were found to be significant, the pre-test was used as a control variable, and one-way ANCOVA was performed. It can be said that factors such as the loss of the subject experienced during the study process and the cleaning of the extreme samples may play a role in the intrinsic motivation pre-test results being significant. As a result of the Levene test performed in the analyses, it was seen that the variances were equal (p = 0.19). However, according to the analysis results, it was observed that there was a differentiation between the groups (F (1,248) = 35.54, p< .001, η2 = 0.12). It is seen that the effect size is medium. Bonferroni protocol’s significance value was figured out as 0.0125 to prevent first type of error in the research.
Interaction dimension posthoc test results.
**p <.01; ***p <.001.
Since the pre-test results of the affective dimension of Social Presence were significant, the pre-test was used as a control variable, and one-way ANCOVA was performed. Here, similar to the situation in the intrinsic motivation scale, it can be said that factors such as the loss of the subject experienced during the study process and the cleaning of extreme samples may play a role in the affective dimension of social presence being significant in the pre-test results. As a result of the Levene test performed in the analyses, it was seen that the variances were equal (p = 0.43). According to the analysis results, it was observed that there was a differentiation between the groups (F (1,248) =64.09, p < .001, η2 = 0.19). Thus, it is seen that the effect size is large. In the research, the significance value of Bonferroni protocol was determined as 0.0125 to prevent first type of error.
Belonging dimension posthoc test results.
**p <.01; ***p <.001.
2X2 ANOVA was performed for the interaction dimension of Social Presence, experimental control, and pre-test post-test comparisons. As a result of the Levene test performed in the analyses, it was seen that the variances were equal (pre-test, p = 0.64, post-test p = .09). According to the results of the analysis, it was seen that there is a difference in both measurement time (F (1,248) =7.02, p = .008, η2 = 0.01) and measurement * group context (F (1,248) =58.59, p < .001, η2 = 0.08). Furthermore, it is seen that the effect size is small for the measurement time and medium for the measurement * group. In the research, the significance value of Bonferoni protocol was determined as 0.0125 to prevent first type of error.
Findings regarding the posthoc test results made to show the difference are given in Table 4. According to the Tukey test results, the experimental group post-test scores are significantly higher than the experimental group pre-test scores (p = .001), and the effect size is small. Experimental group post-test scores are significantly higher than the control group post-test scores (p = .001), the effect size is small. On the other hand, experimental group post-test scores are higher than the control group pre-test scores (p = .001).
2X2 ANOVA was conducted for the belonging dimension of Social Presence, Experimental-control groups, and pre-test-post-test comparisons. As a result of the Levene test performed in the analyzes, it was seen that the variances were equal (pre-test, p = 0.15, post-test p = .37). According to the analysis results, it was seen that there is a difference in both measurement time (F (1,248) =11.06, p = .001, η2 = 0.013) and measurement * group context (F (1,248)=68.23, p = .001, η2 = 0.08). It is seen that the effect size is small for the measurement time and medium for the measurement * group. At this point, the Bonferroni protocol was used to prevent the first type of error in the study, and the significance value was determined as 0.0125.
Findings regarding the posthoc test results made to show the difference are given in Table 5. According to the Tukey test results, the experimental group post-test scores are significantly higher than the experimental group pre-test scores (p = .001), and the effect size is medium. Experimental group post-test scores are significantly higher than the control group post-test scores (p = .001), the effect size is medium. On the other hand, experimental group post-test scores are higher than the control group pre-test scores (p = .001).
Findings related to the perceived learning levels of the experiment and control groups after the intervention
Perceived learning levels t-test results.
***p <.001.
As seen in Table 6, within the scope of the cognitive dimension, which is one of the perceived learning levels, the scores of the experimental group are significantly higher than the control group (Welch (229.30) = 9.24, p = .001) with a large effect level (Cohen d = 1.18). Similarly, within the scope of the affective dimension, which is one of the perceived learning levels, the scores of the experimental group are significantly higher than the control group (Welch (217.47) = 12.47, p = .001) at a high level of effect (Cohen d = 1.59). Finally, within the scope of the psychomotor dimension, which is another of the perceived learning levels, the scores of the experimental group are significantly higher than the control group (Welch (237.44) = 12.33, p = .001) and have a large effect (Cohen d = 1.56).
Findings obtained as a result of qualitative interviews
In this section, the themes obtained in line with the opinions of the participants are included. Participant views were conveyed in the discussion and conclusion part to support the findings obtained from quantitative analysis. Thus, it was provided to convey the findings part in a more concise way.
Examining the experimental group learners’ opinions
When looking at the mean scores obtained from the scales, it was observed that the learners with the highest mean consisted of the learners in the experimental group. Therefore, all the highest five learner scopes in the mean score ranking obtained from the scales consisted of experimental group learners. The experimental group learners were first asked the question, “What do you think about the reflections of your experiences on the discussion forum during the learning process on your communication with other learners?” to evaluate their social presence levels. As a result of the content analysis performed considering the answers received, three different sub-themes were reached. These sub-themes are shown in Figure 1. Experimental group learners’ views on their social presence levels.
The opinions obtained on the theme of “Communications in the discussion forum increasing cohesion and belonging”, which is the first sub-theme reached within the scope of this main theme, are as follows:
Esra: “This situation positively affects our sense of belonging to the process, with other students in the learning process, and with our teacher…”
Gülay: “…. As a result, discussion forums useful for communicating and interacting with our friends…”
Okan: “…The communication and interactions in this process make the process more beneficial…”
Esma: “…I think that reflecting the experiences in the discussion forum are valuable sharing and contribute to the communication with other learners…”
The learners’ statements about the theme “Discussion forums reducing the sense of loneliness”, which appeared as the second sub-theme, were as follows.
Esra: “…Discussion forms allow us to socialize with other students and our teacher who are in the learning process, and to develop sincere relations with them.…”
Gülay: “…With the discussion forums, it is a pleasure to get to know many invisible, intangible people and spread throughout Anatolia like you. This situation also prevents isolation…”
Esma: “…In this way, you realize that you are not alone in distance learning environments…”
The third sub-theme obtained because of examining the opinions of the experimental group learners on the levels of social presence was “Learners being able to produce their own ideas and benefiting from other generated ideas.” The expressions collected under this sub-theme are as follows:
Ali: “…Discussion forums are beneficial in terms of reading the comments and generating ideas…”
Okan: “…In this process, thanks to mutual communication and interactions, they made me think with points of view and facts that I would never have thought of while putting forward ideas…”
Esma: “…I think that reflecting on the experiences in the discussion forum is valuable sharing and contributes to the communication with other learners. …”
The second open-ended question directed to the experimental group learners was “How did you change your desire to participate in open and distance learning environments due to your process in the discussion forum?” to evaluate their intrinsic motivation. As a result of the content analysis performed considering the answers received, three different sub-themes were reached. These sub-themes are shown in Figure 2. The opinions of the experimental group learners on their intrinsic motivation.
Opinions about the first sub-theme reached within the scope of this main theme, “Satisfaction with the learning process,” are as follows:
Gülay: “…All the teachers were professors in the formal education, but you could not understand and ask questions at the same time while listening to the lecture…”
Okan: “…Whether it is remotely or formally, these discussion forums allow people to learn and express themselves. They are activities that develop the possibilities of being able to do so, it is always difficult to find this opportunity…”
Esra: “…Moreover, the opportunity to think about the subject and participate in that way is something that cannot be found in a face-to-face forum…”
The views obtained for the second sub-theme, “Sharing of thoughts increases self-confidence and willingness to participate in the learning process”, were as follows:
Esra: “…Discussing the issues and sharing information increases one’s self-confidence, it makes one more confident, and I really feel like I am a university student even from afar, and I want to contribute to the process…”
Gülay: “...After finishing this program, I’ve already started dreaming of another program…”
Okan: “…I was looking forward to the start of the discussions on the discussion form…”
Esma: “…I would like to participate more in open and distance learning environments by expressing my ideas as much as I can spare time for it…”
The third sub-theme obtained as a result of examining the opinions of the experimental group learners on their intrinsic motivation was “Increased desire to participate in the learning process with the sense of belonging”. Opinions on this sub-theme are as follows:
Ali: “…The discussions made increased my willingness to participate. Feeling the presence of other friends in the learning environment in distance education positively affected our sense of belonging…”
Okan: “… In fact, I can even say that it increased my desire to reach scientific knowledge even more because it aroused curiosity…”
The third open-ended question directed to the experimental group learners was the “Could you please explain your thoughts on the reflections of the discussions you had in the discussion forum you attended within the scope of the human and society lesson on your learning?” question to evaluate their perceived learning levels. As a result of the content analysis performed considering the answers received, six different sub-themes were reached. These sub-themes are shown in Figure 3. Experimental group learners’ views on their perceived learning levels.
The view obtained for the first sub-theme reached under this main theme, “Learning facilitated due to rapid response to comments” is as follows:
Ali: “… It is good to respond quickly to the comments in terms of learning…”
The second sub-theme obtained was the theme of “Satisfaction with the learning process” as under the theme of intrinsic motivation. The views expressed under this theme are as follows:
Esra: “…Thank you for your contributions in the process…”
Gülay: “…I have been looking forward to the opening of discussion forums this term. I hope they will open in the next term too…”
Ali: “…The discussion form is an innovative idea. It is a nice application to share our own thoughts…”
Okan: “…I had the opportunity to express my conclusions by combining the knowledge that I learned in class and that I experienced in my life…
Esma: “…I think that the discussion form has positive reflections on my learning…”
The third sub-theme obtained because of examining the opinions of the experimental group learners on the perceived learning levels was the theme of “Learning environments to be more efficient”. The views expressed under this theme are as follows:
Gülay: “…Discussion forums provide a great excitement and enthusiasm to the learning environment…”
Esma: “…Discussion forums make us think. In this case, it is easier to understand the contents…”
The fourth sub-theme obtained under the main theme of perceived learning was “Increased learning experience with different thoughts”. The opinions gathered under this theme are as follows:
Ali: “… Since it is a more suitable environment for expressing opinions, you also master different thoughts from your own. In this case, it opens new horizons in the learning process…”
Okan: “…Discussion forums are the activities that improve people’s learning and expression opportunities.
I had the opportunity to express my conclusion by combining the information I learned in the course and the information I experienced in my life. I have presented my own idea to others and have also experienced declaring them. I have also experienced being in a discussion forum with thinking and answering other people’s thoughts…”
Esma: “…Discussions in the discussion forum brought a distinct perspective to the subject and significantly impacted the learning process. In this context, the discussion form has positive reflections on my learning…”
The fifth sub-theme obtained under the main theme of perceived learning was “Making positive use of more knowledgeable individuals.” The views expressed under this theme are as follows:
Gülay: “…The fact that there are teachers who are more knowledgeable, well-equipped, and well-versed than you make you prepare your questions, or rather think…”
Okan: “…I have experienced being in a discussion forum by examining and answering other people’s thoughts…”
Esma: “…Discussions in the discussion forum brought a unique perspective to the subject, but had a significant impact on the learning process…”
The sixth and last sub-theme, which is related to the main theme of perceived learning, was the theme of “Solving incomprehensible issues”. The views expressed under this theme are as follows:
Esra: “…It was very productive for me. I found the topics that I was stuck on and could not understand more understandable thanks to the attention of my teacher…”
Esma: “… As in most courses, I have difficulty in expressing the contributions of the discussions within the scope of the human and social course to my life. In short, I learned a lot about the subjects I did not know…”
As a result of these interviews, it is seen that both the social presence levels, intrinsic motivation levels, and perceived learning levels of the experimental group learners, who have higher averages than the scales presented to them within the scope of the study, are positive as in the quantitative findings. In other words, it is seen that the experimental group learners, who have higher scores than the control group in the quantitative findings section, have a positive view of the variables examined within the scope of the study.
Examining the control group learners’ opinions
When looking at the mean scores obtained from the scales presented to them within the study’s scope, it was observed that the learners with the lowest mean consisted of the learners in the control group. Therefore, all the lowest five learner scopes in the mean score ranking obtained from the scales consisted of control group learners. Therefore, control group learners and experimental group learners were first asked the question, “What do you think about the reflections of your experiences on the discussion forum during the learning process on your communication with other learners?” to evaluate their social presence levels. As a result of the content analysis performed considering the answers received, four different sub-themes were reached. These sub-themes are shown in Figure 4. Control group learners’ views on their social presence levels.
The opinions obtained regarding the first sub-theme reached within the scope of this main theme, “Inability to benefit due to not participating sufficiently in discussions” are as follows:
Mustafa: “…I didn’t get a chance to communicate much because I didn’t spend enough time in the discussion forum…”
Resul: “…I did not have the opportunity to communicate because I did not use it much…”
The second sub-theme obtained under the main theme of social presence was “Discussion environment not suitable for exchange of ideas.” The views of the learners regarding this sub-theme are as follows:
Emine: "… I do not think it is suitable for me to exchange ideas…"
Tuna: “…Undoubtedly, the ideas of learners influence each other in discussion forums. But since I was lost among comments, I could not see much of its effect…”
The third sub-theme obtained was “The need for instructor to take a more active role.” The learner’s view on this sub-theme was as follows:
Burak: “…The instructor should be more active in the discussion forum. Otherwise, there is hardly to make any communication …”
The fourth sub-theme obtained under the main theme of social presence was “Inability to get appropriate responses to the thoughts expressed in order to communicate.” The opinions obtained in this context are as follows:
Burak: “…There is an opportunity to communicate with other participants in the discussion forum. However, you usually do not get the answer you expect to the statement you write…”
Tuna: “…I can’t express myself and I can’t get satisfactory answers because I’m lost in comments…”
The second open-ended question directed to the control group learners was question “How did you change your desire to participate in open and distance learning environments as a result of your process in the discussion forum?” to evaluate their intrinsic motivation. As a result of the content analysis performed in the light of the answers received, three different sub-themes were reached. These sub-themes are shown in Figure 5. The opinions of the control group learners on their intrinsic motivation.
The first sub-theme reached under this main theme, as in the main theme of social presence, was the theme “Inability to benefit due to not participating sufficiently in discussions.” Opinions about this theme are as follows:
Mustafa: “…I only use the e-campus to study the course subjects. Apart from that, due to my busy work life, I have difficulties allocating time or having problems focusing on the subjects at that moment. So, I don’t spend a lot of time on discussion forums either…”
Burak: “…It didn’t affect my participation in the learning process much because I didn’t spend enough time in the discussion forum…”
The second sub-theme obtained was “Inability to benefit due to the elaborate discussion structure.” The views expressed within the scope of this sub-theme are as follows:
Resul: “…The discussion forums did not have a positive effect on my participation in the process, as it was a very complex environment…”
Tuna: “…Truth emerges from the collision of ideas.” But the discussion forums are not effectively managed. As such, we could not have a healthy discussion environment with our fellow students…”
The third and last sub-theme obtained under the main theme of intrinsic motivation was “Having a negative attitude towards the process”. The views expressed under this theme are as follows:
Resul: “…The discussion form has a very complex structure…”
Emine: “…Honestly, I benefited more from other materials. The discussion forum did not make any significant contribution…”
Tuna: “…I think the discussion forums are not properly managed…”
The third open-ended question directed to the control group learners was the “Could you please explain your thoughts on the reflections of the discussions you had in the discussion forum you attended within the scope of the human and society lesson on your learning?” question to evaluate their perceived learning levels. As a result of the content analysis performed considering the answers received, four different sub-themes were reached. These sub-themes are shown in Figure 6. Control group learners’ views on their perceived learning levels.
The opinions obtained for the first sub-theme “Inability to benefit due to not participating sufficiently in discussions”, obtained under this main theme, are as follows:
Mustafa: “…Since I haven’t been in the discussion forum much, these environments have not had much effect on me…”
Resul: “…I can’t evaluate because I did not use it much…”
Tuna: “…I did not participate in the discussion forums much. Because I do not use the e-campus program often. Also, during the periods when I used the e-campus, it did not attract my attention and interest…”
The second sub-theme reached under the main theme of perceived learning was “ Too crowded discussion environment”. The opinion expressed under this sub-theme is as follows:
Burak: “…Discussion forums seem too crowded for me…”
The third sub-theme obtained was the theme of “Having a negative attitude towards the process”, as in intrinsic motivation. The views expressed under this theme are as follows:
Burak: “…I cannot find the answers I am looking for…”
Emine: “…I do not warmly welcome participation in forums. I think it is verbiage…”
Tuna: “…When I used the e-campus, discussion forums did not attract my attention and interest…”
The fourth and last sub-theme reached under the main theme of perceived learning was “Failure to get an appropriate response to the stated opinion.” Opinions on this sub-theme are as follows:
Burak: “…You can’t get enough answers for a written comment…”
Emine: “…There is a lot of talk in the forums. No one is giving a clear answer…”
As a result of these interviews, it is seen that the social presence levels, intrinsic motivation levels, and perceived learning levels of the control group learners, who have lower averages than the scales presented to them within the scope of the study, are negative as in the quantitative findings. In other words, it is seen that the control group learners, who had lower scores than the experimental group in the quantitative findings section, had a negative view of the variables examined within the scope of the study.
Discussion and conclusion
As a result of the analysis conducted to determine the social presence levels of the learners, it is seen that the results obtained by the experimental group, which was divided into manageable groups and each group was led by a field expert, were significantly higher than the control group. Considering the five elements of social presence (knowledge and experience, interaction density, instructor’s interest, community harmony, and emotional unity) defined by Whiteside (2015), the reason for the higher social presence levels of the experimental groups, which consist of manageable groups and each group was assigned as a guide of a field expert, can be explained. It has been difficult for a single field expert to intervene and guide a group of learners with more than 150 learners effectively. This situation brought about the inability to transfer knowledge and experiences, weakening the interaction, the instructor’s indifference, and the absence of community harmony, and emotional unity.
In online learning environments where social interactions are at the center of the learning process (Sung and Mayer, 2012), the groups’ size can affect learners’ participation and their communication and interaction with other learners (Aragon, 2003). At this point, subgrouping the control group in a manageable manner did not show the expected effect since only one field expert looked at each subgroup. As a result, a field specialist was held responsible for 186 learners in the control group. This situation caused the control group to show massiveness and difficulty managing. Like this result, Akcaoglu and Lee (2016) concluded in their studies that group size harms group adjustment. In line with these results, Hamann et al. (2012) stated that dialogues between learners in large groups are insufficient, and learners have less discussion with each other. On the other hand, as can be seen in this study, it can be expressed that smaller groups offer more opportunities for learners to communicate and interact and participate in learning processes more actively (Lowry et al., 2006), which positively affects social presence (Kotturi et al., 2015).
The findings obtained from the structured interviews conducted with them on the learners’ social presence also supported the findings obtained from quantitative analysis. Within this scope, the control group learners’ opinions support that the learners in this group do not have enough communication opportunities in the learning process, and accordingly, their social presence levels decrease. In line with this result, Horzum et al. (2015) stated that “When learners ignore the discussions during the online group discussion process and do not participate in the discussions on time, the quality of communication and interaction will decrease and as a result, social presence levels will be negatively affected.” In addition, the learners’ opinions have been an indication that the learners do not have any difficulties in communication and interaction with each other. This situation contributes to the increase of the social presence level of the learners. This result obtained overlaps with the intensity of interaction and community cohesion, which are among social presence elements, as stated by Whiteside (2015). Considering that the level of interaction of learners with increased social presence also increased (Reio and Crim, 2013), it can be said that experimental group learners did not have any difficulties in interaction and communication. This situation coincides with the statement expressed by Lowry et al. (2006) that “smaller groups provide more opportunities for learners to interact and participate in learning processes more actively.”
As a result of the analysis conducted to determine the intrinsic motivations of the learners, it is seen that the results of the experimental group, which were divided into manageable groups and each group was led by a field expert, were significantly higher than the control group. Considering that factors such as feedback, communication, and fairness are factors that positively affect intrinsic motivation (Ryan and Deci, 2000), it is understandable that the intrinsic motivation was higher in the experimental group, where the teachers spent more time for the learners and communicated more by giving more feedback.
Kim et al. (2020) stated that the reasons that hinder or reduce learners’ motivation are content, lack of communication and interaction, and the instructor’s indifference. Within this scope, it can be stated that in the control group, the lecturers’ inability to deal with the ideas shared by the learners sufficiently brought along the lack of communication and interaction. In this case, it is considered as a reason that decreases the intrinsic motivation of learners. In this study, the control group learners stated that communication and interaction were incomplete under the sub-theme of “inability to benefit due to the complex discussion structure”. Considering these opinions obtained from learners, it is seen that learners experience a lack of communication and interaction that will cause a decrease in their intrinsic motivation. However, based on the expressions stated under the sub-theme “failure to benefit due to the complex discussion structure” and “negative attitude towards the process” obtained as a result of the opinions from the control group learners, it can be stated that there are problems in the management of the learning process in which learners are involved and that due to this situation, learners are insufficient in communication and interaction. In this context, it was concluded that groups with massiveness characteristics are perceived as a complex structure by learners, and this situation negatively affects the motivation in the learning process. This finding coincides with the finding that “large groups were seen as complex and problematic by learners” obtained in a study conducted by Lowry et al. (2006).
Statements under the sub-themes of “satisfaction with the process” increased self-confidence and willingness to participate in the process “and” increased desire to participate in the learning process with a sense of belonging obtained as a result of the opinions from the experimental group learners, show that learners do not have any difficulties in communication and interaction during the learning process. In such a case, learners’ willingness to take part in the process increased; thus, their motivation towards the process increased. In this context, it can be said that the qualitative findings obtained coincide with the findings obtained because of the quantitative analysis conducted within the scope of the research.
As a result of the analyses conducted to determine the perceived learning levels of the learners, it is seen that the results obtained by the experimental group, which was divided into manageable groups and each group was led by a field expert, were significantly higher than the control group. This situation can be explained by the fact that the experimental group learners were more satisfied with the process and were more likely to interact. This opinion also coincides with Baturay’s (2011) statement that “the perceived learning levels of the learners are in a very strong relationship with their satisfaction with the learning process and the learner-content interactions”. Therefore, the satisfaction of the process and the learners’ interaction were among the factors that increased the perceived learning level in the experiment group. The experimental group learners’ satisfaction with the learning process was expressed under the sub-theme of “satisfaction with the process” obtained because of their opinions. As a result of these opinions obtained from the experimental group learners, it is possible to say that the online group discussion technique, which is conducted by dividing into manageable groups, has positive results on the learners’ perceived learning levels. In addition to this situation, a study conducted in the relevant literature (Costley and Lange, 2016) indicates that the designs that make instructors engage with learners have a high degree of influence, and positively affect perceived learning.
In another study conducted on perceived learning, Richardson and Swan (2003) stated that as learners’ social presence levels increase, their perceived learning levels also increase. In addition to this situation, according to Ferreira et al. (2011), there is a positive and high relationship between learners’ motivation towards learning processes and perceived learning. These findings obtained in the relevant literature align with the findings obtained within this study’s scope. Within this study’s scope, the fact that the experimental group, which had significantly higher scores in terms of intrinsic motivation and social presence, had significantly higher scores in the context of perceived learning level is a situation that coincides with the findings in the literature.
The findings obtained from the structured interviews with learners within the study’s scope supported the findings obtained because of quantitative analysis. Accordingly, the opinions of the experimental group learners, who reached higher scores due to quantitative analysis in the context of perceived learning, were that the experimental process made a positive contribution to their perceived learning. Based on the themes obtained because of the experimental group learners’ opinions, it can be stated that the experimental group learners experienced factors such as instructor interest (Costley and Lange, 2016), satisfaction (Baturay, 2011), the realization of interactions (Trespalacios and Perkins, 2016), which would increase their perceived learning level during the experiment period.
As a result, it can be stated that the design to be applied in the online group discussions process, which is used as an effective teaching technique in online learning environments, is important in increasing the efficiency of the process. In this context, it has been observed that to manage online learning processes effectively, attention should be paid to the fact that the groups do not show mass features. However, developing dialogues by interacting with each other and thinking critically for those who learn in online discussion forums is closely related to the teaching techniques that the instructors will apply. At this point, it can be said that instructors should act in line with the principles of social constructivist theory and the theory of communities of inquiry, which coincide with the nature of online learning environments. In this context, instructors should prefer the daily spoken language instead of using formal language. In this way, learners will be supported to exhibit a more participatory behavior. Social interaction has a particularly key place in the theories adopted within the scope of the study (Jaramillo, 1996). Therefore, it can be said that in the context of the language and teaching techniques used in the online discussion process, attention should be paid to each element that enables learners to engage in social interaction. Another factor that should be emphasized at the point of enabling learners to engage in social interactions is that instructors should function as guides rather than providing direct information in the online discussion process. In other words, it is important for the learners to be able to express themselves and interact with a question based on interpretation to create an atmosphere for discussion. According to the theories that form the basis of the research, the learning of the individual requires a special social environment and a social process that includes the upbringing of individuals in this environment (Vygotsky, 1967). Therefore, it is an essential element that learners can exhibit their own interpretations and support this process with their comments. In this context, it can be said that the questions directed to the learners during the discussion process should be open to interpretation. Another point that the instructors should pay attention to in the online discussion process is that the learners should be informed before starting the discussion process. Conveying the points to be followed in the learning process and giving information about what is the scope of the subject to be discussed, are evaluated as points that increase the sense of belonging of the learners in the process. The efficient use of online discussion forums in learning processes will increase communication and interaction between learners and enrich learning outcomes. The fact that learners fulfil the interaction elements and participate more in learning processes creates a positive effect on social presence, intrinsic motivation, and perceived learning situations.
Recommendations
Considering that the experimental studies conducted for group discussion in online learning environments are insufficient (Gokhale and Machina, 2018) within the scope of future studies, It is thought that the technique of group discussion, which will enable learners to realize the learning process, especially in the principle of community of inquiry and social constructivism theories, should be used more and subject to more scientific research. Furthermore, many variables such as sense of community, learner satisfaction, transactional distance, learner attitude, and self-directed learning skills, which should be examined in online learning environments, need to be employed in the online group discussion process. In this way, it will be determined which aspects of the online group discussion process benefit learners in online learning environments and which aspects do not.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Author’s note
The current study utilized the dataset of the first author’s PhD dissertation.
Intrinsic motivation:
1- I enjoy working in e-learning environments. 2- I prefer to work in e-learning environments even if I have printed materials. 3- I look forward to working in e-learning environments. 4- I am satisfied with my work in e-learning environments. 5- I determine my own learning needs.
Social presence:
1- I use emoticons such as smiley face [:) , ;), :P ] to express my feelings while chatting. 2- I use capital letters to express my feelings [angry, surprise…etc.] when texting in a chat environment. 3- I express my feelings by using different colors in the chat environment. 4- I use humorous elements in my communication with other learners in the virtual environment. 5- I talk to my friends online about my personal issues. 6- I give my opinion on whether I agree with the ideas expressed in the virtual environment. 7- I do not react to correspondence in the virtual environment. 8- I express my ideas clearly to other learners. 9- I hesitate to ask other people questions online. 10- When I enter the virtual environment, I greet other people in the environment. 11- When leaving the virtual environment, I use expressions stating that I will be separated from other people [good evening, see you tomorrow, I’m leaving… etc.] 12- I use their names when addressing other people in the virtual environment. 13- I share information with other people in the environment. 14- I am happy to be remembered with my group mates. 15- I feel close to other learners. 16- I feel like a part of the team. 17- When I talk about my friends in the virtual environment, I use the expressions we, ours, our group… etc.
Perceived learning
1- I can organize educational materials in a logical structure. 2- I cannot produce a study guide for future learners for this training. 3- I can use the physical skills I learned in this training in my daily life. 4- As a result of this training, my attitude about the subjects changed. 5- I can logically criticize the texts used in this training. 6- I feel more confident thanks to the subjects I learned. 7- As a result of this training, I could not improve my physical skills. 8- I can show others the physical skills I learned in this training. 9- As a result of this training, I feel that I am a more intellectual thinking person.