What Types of Learning may occur through Casual Use of a Social Network Site?The Case of Facebook

Over the past decade, World Wide Web has undergone a subtle but profound shift to Web 2.0 to become more of a social web. Social network sites (SNSs) form an integral part of everyday life for a plethora of people and represent a typical paradigm of this web evolution. Yet, little is known about how casual use of SNSs is related to learning, while most studies on SNSs revolve around adolescents and emerging adults. This study aims at casting light on the types of learning through casual use of SNSs by adults who seek for learning activities. For this purpose a quantitative survey has been administered to the most active members of a Greek Facebook group, whose main goal is to share announcements among members regarding learning events. The results showed that participants integrate Facebook in their daily practice, with women and younger adults spending slightly more time on site. Participants reported that several distinctive occurrences of learning emerge through casual Facebook use and it is concluded that everyday Facebook use may help in developing multiple types of learning.

online environments bear a more diverse set of information resources and provide increased opportunities for information sharing and idea genesis.

Conceptual Framework
Essential for the study was to employ a robust conceptual framework to compare and contrast users" perceptions of learning through casual use of Facebook. Foremost, it is necessary to define what learning is. Learning can be understood as a process and as the outcome of this process (Smith 1999a). When learning is conceived as an outcome, the focus is placed on the individual and the emphasis is on knowledge acquisition and skills development. Yet, learning does not occur in a vacuum; it is an inherently social and participatory process situated in the sociocultural milieu. While these conceptions are often treated independently as two typical levels of analysis, understanding their interrelationships yields a conceptually more satisfying picture of learning. Salomon and Perkins (1998) unfold the interplay between these two perspectives, providing an appropriate theoretical perspective from which to approach learning through SNS use: Active social mediation of individual learning. It is the socially facilitated acquisition of knowledge and skills and occurs when a person or a group of people help an individual to learn. It is one of the most fundamental social forms of learning and it can be considered the same as instruction: a teacher teaches its students; parents guide their child, and so on. Here, there is a rather clear distinction between the individual learner, the learning products carried away as transferable cognitive commodities and the social agents facilitating that learning.
Social mediation as participatory knowledge construction. While in the previous category the focus falls on the learner"s learning, here individual cognitive activity and the social mediation of learning are seen as an integrated and situated system in a particular context. Rather being transmitted from a knowledgeable authority to its apprentices, knowledge becomes jointly constructed through interaction. Accordingly, the learning products are distributed over the entire system, rather than being owned by the individual.
Social mediation by cultural scaffolding. When individuals" learning is supported by cultural artifacts, such as books or tools. Artifacts, being culturally and historically situated per se, convey the wisdom that directed their design. In that way, the learner and the artifacts form a learning system, which reorganizes action and determines  Salomon & Perkins, 1998).
Learning to be a social learner. A social dimension of meta-learning. Through social interaction, people develop knowledge, skills, and understanding not only in particular domains, but also about learning itself in social settings. Hence, for example, they learn where, when, and how to ask questions or how to participate in reciprocal learning relationships. Here, the individual extends its capacity to overcome learning obstacles by participating in and capitalizing on the social milieu.
Learning of social content. The last category includes matters such as how to get along with other people, how to collaborate harmoniously, and so on. Here, it is not introduced a new way of understanding learning systems rather the learning system would be one of the previously discussed, operating on social content.

Methodology
The study aims at casting light on the types of learning through casual use of SNSs by adults who pursue learning activities. As aforesaid, Facebook is the SNS prototype used in this study as it surged in popularity the recent years.
A quantitative survey method has been selected, as it is a simple and straightforward way to study attitudes, beliefs, and motives (Robson, 2002, p.233), while it is a time/cost saving method and affords access to unique populations (Wright, 2005).
The survey instrument administered to the most active members of a Greek Facebook group, whose main goal is to share announcements about learning events among its members. The study focused on this particular group for it seemed to constitute an appropriate study population as it will be explained below, and it is also a sheer Greek group, providing that way a good opportunity to study attitudes and perceptions of Greek Facebook users.

Instrument
Data was collected by means of an online survey developed on the Bristol Online Surveys service. The survey comprised three sections and is presented in the Appendix. In the first section, demographic characteristics of participants were collected through four questions. In the second section, data referring to participants" Facebook usage frequency and time length spent on Facebook on a daily basis were collected through two rating scale questions.
Additionally, a 4-point Likert scale with seventeen questions, largely derived from Subrahmanyam et al."s questionnaire (2008, p.427), inquired about participants" frequency of performing several Facebook activities.
The third section consisted of four 6-point Likert scales with 27 questions in total, aimed at gathering participants" perceptions about learning through casual Facebook use. The response scale for each item ranged from "strongly disagree" (1) to "strongly agree" (5), with middlepoint (3) of the scale anchored by "neither agree nor disagree", while, consistent with Ryan and Garland (1999) suggestion, an additional "not applicable" option allowed respondents to provide an answer that is true to their experience.
In the second and third section there was an optional open-ended question so participants could provide examples and other qualitative feedback. This self-reported questionnaire was grounded on the above presented conceptual framework. The formulation of the questionnaire was facilitated by several unstructured short "informal" interviews which were conducted through the Facebook chat tool. This approach was chosen for its relevance to the research context. As Robson proposes (2002, p.282), seizing opportunities for short "informal" chats in the "research setting" is likely to provide authentic perceptions about anything concerning the research. Along the same lines, synchronicity embedded in the medium accelerated interaction, stimulated memories, and helped participants comment more freely about their experiences.
The instrument"s validity was tested by three educators who are experienced Facebook users and who engage with learning in both their professional and personal life. By virtue of their profession they could reasonably be considered experts on what learning is possible in a given context, despite that the study is on a novel research field. Additionally, their experience on Facebook helped refine their understanding of the learning potentials afforded by the specific SNS. Therefore, their opinion was requested to check whether the questionnaire was conceptually on the right lines and whether the statements were understandable. Based on the received feedback, ambiguous items were revised and the instrument was modified prior to administration.

Study Group
The study focused on a Greek Facebook group, whose members post announcements on the group wall about impending learning events. When this group launched in January 2010 it attracted great numbers of users in a short span of time; the group listed more than 36,750 members nine months after launching. The group has no particular thematic focus, rather it embraces all disciplines and interests from astronomy to mathematics to history to poetry to dance. Interaction on the group wall among members is low as it is used mostly for posting and reading announcements about learning events. It is indicative that while 764 announcements were posted from group launch till September 2010, only 164 comments were made during the same period. All members are invited and welcome to post information on the wall about upcoming seminars, conferences, workshops, and so on, so as the rest of the group members can be informed about impending events.
However, the vast majority (more than 98%) of members do not contribute to the wall with any kind of information (i.e., post, comment, or even "Like" on others" posts). This is consistent with a study on Facebook which revealed that many users spend much of their time reading and viewing information without interacting in any way, while over half of users only rarely interact with groups (Pempek et al. 2009). This tendency of no contribution by the majority of group members can be explained by means of a study about reasons of not contributing in online communities, which revealed that the most popular reason among users for not posting is that they feel that their information needs can be satisfied without posting (Preece, Nonnecke, & Andrews, 2004). Hence, it can be safely hypothesized that most members of this group do not contribute to the group wall because they can satisfy their informational needs without explicitly posting.
This group was chosen as the study population mainly for two reasons. Firstly, the members of the group explicitly look for learning activities. So, they have a predisposition to attend to events which are directly related to their specific domains of interest, or to put it in other words, they hold an individual interest in relation to a particular topic or domain (Ainley, Hidi, & Berndorff, 2002). Individual interest is personal, robust, and often wide-ranging (Trend, 2005), therefore it is expected that members of the group will seek any kind of learning opportunity on Facebook which is related to their interests. It is also anticipated that they will engage in activities related to their interests while using Facebook, for example they Online Journal of Communication and Media Technologies Volume: 4 -Issue: 2 -April -2014 will follow relevant links, read discussions, chat with people who share the same interests, and so on. Additionally, it is hoped that they will be competent to identify any beneficial effect on their learning due to casual use of Facebook. Secondly, this group does not constitute an online community of practice neither purports to establish one. If this group had formed a community of practice then, inevitably, its members would have been involved in a learning process by engaging with the community"s ethos and by participating in day-to-day interactions. It is beyond the scope of the present study to explore what types of learning may occur in a community of practice on Facebook, or whether Facebook affords or empowers the development of a community of practice; rather this study aims to explore what types of learning may occur through casual use of Facebook. Surely, members of this group may participate in other Facebook groups whose members form a community of practice, still participation in communities of practice can be reasonably considered as part of their casual use of Facebook.

Participant Characteristics and Data Collection
Research was conducted with the most active members of the "announcements about learning events" Facebook group over an eight-week period in July-September 2010. In this paper, active members are defined as those who have posted information about a learning event, or have given feedback on posts from the group launch. Most active members were purposively selected as the focus group, based on the premise that they are pursuing learning activities more energetically than the rest of the group members.

Sampling Strategy.
The sampling strategy involved a twofold approach. A personal invitation message with a short description of the study, information about confidentiality, and a link to the survey was sent to the 554 most active members twice during the research period. Some people who were invited and took the survey forwarded the survey"s link to their friends voluntarily.
Additionally, the survey"s link with information about the research study was also publicised on the group wall twice and members were openly invited to participate. These mixed recruitment methods elicited a total of 250 responses, yielding a response rate of 45.1% of the total members invited. Yet, the number of respondents was very low (less than 2%) compared with the number of group members. Therefore it is impossible to know how well they represent the group as a whole (Wright, 2005). Of the 250 participants, forty-nine failed to Subsequently, any results must be seen in the light of these biases towards female participants.
In terms of age, respondents ranged from under 20 to over 50, with approximately 70% being older than 26 years old.
Respondents were highly educated; over 90% had a graduate degree or higher. This is consistent with Lenhart et al."s (2010) findings that adults with at least some college experience are more likely to use SNSs than other users and supports the initial hypothesis that respondents were interested in learning. Table 1 summarizes the demographic profile of the respondents in the survey.
methods were utilized, since only nominal (male/female) and ordinal (never/rarely/sometimes/often) data were used. However, in analysing participants" perceptions about learning on Facebook parametric methods were also employed, for they are more efficient and sophisticated than non-parametric methods (Carifio & Perla, 2008;Pell, 2005;Robson, 2002). Although data were once more ordinal in character, scholars have long argued for treating Likert scale data as interval in nature (Carifio & Perla, 2008;Pell, 2005); many studies use parametric methods to analyze attitude data (Göb, McCollin, & Ramalhoto, 2007), while a variety of studies have revealed that "the Likert response format produces empirically interval data" (Carifio & Perla 2008, p.1150).
The technique implemented in this section is carried out in two phases. In the first phase, a graph is usually presented in order to immediately compare data across independent variables (gender, educational level, age) and examine possible relations. Gradation of colour from the fair to the dark represents increase in frequency or duration. In order to facilitate chart interpretation, adjacent categories with few responses were grouped together, as sacrificing detailed information is often acceptable in return of complexity decrease of the results (Robson, 2002). Therefore, age, frequency and duration of Facebook usage categories were reduced to five, four, and four respectively. In the second phase, several tests statistically validate, or invalidate, visual observations.

Facebook Usage and Activities
This part summarizes the results on the effect of gender, age, and educational level on the frequency and duration of Facebook usage and on users" activities.

Differences in Facebook Usage Related to Gender.
According to Mann-Whitney U test there is no important difference in frequency of Facebook usage among genders, yet it was found that female users tend to spend more time on Facebook than males.

Differences in Facebook Usage Related to Age.
Sommers"d test marked that, compared to older, younger users check their Facebook accounts more frequently, and that users aged between 21 and 25 spent more time on Facebook everyday than users from other age group categories. Moreover, partial correlation

Differences in Facebook Activities Related to Gender.
Mann-Whitney and partial correlation test with the effect of age and time spent on Facebook in turn controlled indicated that women present a statistically significant higher tendency to listen to/find music than men (p=0.000; Correlation=-0.280), whereas men present a statistically significant higher tendency to create groups about specific topics (p=0.000; Correlation=0.230) and look for new friends (p=0.10; Correlation=0.169).

Differences in Facebook Activities Related to Age
Sommers"d test marked that older users create events more often than younger ones and that under 25 years old users tend to look at others" photos more often than the rest of the users.
Both findings were confirmed by partial correlation analysis with the effect of time spend on Facebook controlled (p=0.001, Correlation=0.207; p=0.028, Correlation=-0.143).

Learning through Facebook Use
This part is comprised of five subsections, each devoted to a distinctive occurrence of learning according to the conceptual framework and presents summarized results of the perceptions of adult users with respect to learning through Facebook. It is also explored whether gender, age, and educational level influence these perceptions. Participants had the chance to opt for "not applicable" regarding their pragmatic experiences to each questionnaire item of this section, yet less than 4% of the total responses, almost equally distributed over the twenty-seven questions, belonged to this category. Thus, since "not applicable" responses were not meaningful for the study, they were grouped with missing responses, and, as before, the actual number of responses in a particular analysis is indicated as "AN".

Differences in Perceptions Related to Demographic Characteristics.
As can be observed from the pie chart series below, users under 25 years old perceived, more than other age group categories, that they developed their writing skills and gained knowledge concerning foreign language use through Facebook.

Figure 1: Age group differences as regards perceptions about: (a) writing skills development, and (b) foreign language knowledge acquisition through Facebook use.
An analysis of variance test was conducted using age as the independent variable. For both findings it was confirmed that there is statistical significant effect of age (p=0.002 and p=0.008 respectively) on users" perception. As can be observed from the pie chart series below, lower educational level users tend to perceive more that they developed their writing skills and that they gained knowledge concerning foreign language use than higher educational level users.

Figure 2: Educational level differences as regards perceptions about: (a) foreign language knowledge acquisition and (b) writing skills development.
An analysis of variance test was conducted using educational level as the independent variable. Same as before, for both findings it was confirmed that there is statistical significant effect of educational level (p=0.002 and p=0.022 respectively) on users" perceptions.
Two multiple linear regression analyses were conducted using foreign language knowledge acquisition perception as the dependent variable in the first analysis, and writing skills development perception as the dependent variable in the second analysis, while age, educational level, and time spent on Facebook were independent variables. In both analyses the coefficient of determination was low (Ra2=0.120 and Rb2=0.095 respectively). Age was not statistically associated with their perceptions, while educational level and time spent on Facebook were jointly significant in predicting users" perceptions (Fa=10,643, pa=0.000, Fb=8.259, pb=0.000). Finally, standardized coefficients indicated that educational level was reversely related with users" perception while time spent on Facebook had a greater effect on the result.

Social Mediation as Participatory Knowledge Construction on Facebook.
Users" perceptions with regard to social mediation as participatory knowledge construction on Facebook were collected through three questions (see table 4 for descriptive statistics). More than half of the participants agreed that they chat on Facebook with people about issues related to their interests, whereas two-thirds of the participants concurred that they participate in wall discussions where people exchange information and opinions about issues related to their interests and that they actively participate in groups thematically related to their interests. It was explored whether gender, age, and educational level influence these perceptions, yet no statistically significant effect was found.

Social Mediation by Cultural Scaffolding on Facebook.
Users" perceptions with regard to social mediation by cultural scaffolding on Facebook were gathered through seven questions (see table 5 for descriptive statistics). Nearly one-third of the participants signified that Facebook helped them become more self-regulated in learning how to use ICT, gain experience in computer related tasks, and find new/better ways to accomplish computer related tasks. Over four out of ten participants credited Facebook for facilitating their understanding of several ICT issues. More than half of the participants asserted that posted videos/images on walls helped them reflect on or gain new knowledge.
Finally, two-thirds of the participants acknowledged that their personal wall facilitates their effort to directly present or express their ideas, while approximately eight out of ten recognised their personal wall as a way to present their ideas or express them through representative means. Same as before, gender, age, and educational level does not seem to influence these perceptions.

Learning to be a Social Learner on Facebook.
Users" perceptions with respect to learning to be a social learner on Facebook were gathered through five questions (see table 6 for descriptive statistics). Nearly two out of ten of the participants asserted that they learned how to seek help or when and how to pose questions due to Facebook use, three out of ten that they learned to accept different thoughts and perspectives, and four out of ten that they can more easily start a conversation with an unknown/little known person. Finally, approximately two-thirds of the participants affirmed that they observe other people"s behavior on Facebook and the consequences of those behaviors.

Differences in Perceptions Related to Demographic Characteristics.
As can be observed from the chart below, male perceive more that they can more easily start a conversation with an unknown/little known person on Facebook than female users.

Figure 3: Gender differences on perception regarding their easiness to initiate conversation with unknown/little known persons on Facebook.
A regression analysis was conducted with users" perception as dependent variable and gender, age, educational level, and time spent on Facebook as independent variables. The analysis revealed that all independent variables were jointly statistically significantly related to users" perception (R2=0.082).

Learning of Social Content on Facebook
Users" perceptions with reference to learning of social content on Facebook were gathered through three questions (see table 7 for descriptive statistics). Nearly one out of ten participants reported that they often present a different personality either in their profile or in chat discussions so as to promote themselves better and two out of ten that they learned how to get along with other people due to Facebook use. It was explored whether gender, age, and educational level influence these perceptions, yet no statistically significant effect was found. UK (Ofcom, 2008), participants in this study exhibit significantly increased SNS usage. In parallel with Lenhart et al."s findings (2010), younger adults use Facebook more frequently than older ones, while this study revealed that they spent more time on Facebook too. Women tend to stay on Facebook slightly more than men, whereas taking into account the striking gender differences around activity in the study group they are also more energetic users.
Consistent with Ofcom"s report concerning men"s attitudes on SNSs (2008), men reported looking for new friends more frequently than women, which is in harmony with their feeling that they can more easily start a conversation with an unknown/little known person on Facebook. Additionally, it was found that men are likely to create groups more frequently than women, whereas women tend to listen to music on Facebook more often. In agreement to some extent with both Subrahmanyam et al."s (2008) and Pempek et al."s (2009) findings regarding activities on Facebook, the most frequent activities among participants were found to be reading comments/posts, listening to/finding music, and posting video links, whereas creating groups and events were the most uncommon activities. Boyd argued that adolescents use SNSs for entertainment and "social voyeurism" (2008, p.127); it seems that this disposition only gradually fades away since watching video links and looking at others" photos were more common activities among participants under 25 years old.
Conversely, creating events was more frequently operated among older participants. This might be explained by qualitative feedback, as two participants over 41 years old noted that they use Facebook for profession-related matters or for advertising their activities, hence it can be hypothesized that creating events is an efficient way to achieve that.

Learning on Facebook
Current social learning theories posit that knowledge and learning are inherently social, while cultural activities and artefacts are regarded as integral to conceptual development (e.g., John-Steiner & Mahn, 1996;Palinscar, 1998;Prawat, 1996). Given the recent worldwide proliferation of SNSs, these sites presumably play a role in people"s learning.
This study brought to light that half of adult users perceive that socially mediated individual learning is facilitated on Facebook either by getting feedback on ideas expressed on their wall or by gaining knowledge, getting ideas, and seeking information by asking other users, while most of them perceive that through groups they keep updated on issues of interest.
Additionally, a significant number of participants acknowledged getting guidance by asking other people on Facebook, developing their writing skills and gaining knowledge concerning foreign language use. Moreover, it seems that young users and users with relatively low educational background are likely to acquire more writing skills and foreing language benefits due to Facebook use than their more educated or mature counterparts.Yet, the vast majority of users disputed that they gain knowledge concerning native language use.
Over half of users perceive that participatory knowledge construction takes place on Facebook by actively participating in groups, in wall discussions, and in chats. Social mediation by cultural scaffolding emerges on Facebook as well; most users recognise their personal wall as both a medium and a facilitator to express their ideas through representative means. In addition, a large part of users perceive that Facebook facilitates their efforts to become more self-regulated in learning how to use ICT, to understand several ICT issues, to gain experience in computer related tasks, and to find new/better ways to accomplish computer related tasks.
Users acknowledge that they are learning to be social learners on Facebook. The majority of users observe other people"s behavior and its consequences, whereas a significant part of users assert that they can more easily start a conversation with an unknown or little known person, and they have learned when and how to pose questions, how to seek for help, and to accept different thoughts and perspectives due to Facebook use.
Finally, a considerable number of participants identify learning of social content on Facebook as they often present a different personality so as to better promote themselves and assert that they have learned how to get along with other people due to Facebook use.
It appears, therefore, that everyday Facebook use can nurture multiple types of learning. This adds to the view that, by using digital technologies, online users learn "a whole range of skills", such as communication, general knowledge, multitasking, technical confidence, creativity, collaboration, and so forth (Green and Hannon 2007, p.35). However, this is not true for all users, in fact most users disputed that several distinctive occurrences of learning take place on Facebook and consequently it was not perceived as a holistically effective medium. Hence, what it works for some it might not be working for the rest. Even so, the results are promising as regards learning through casual use of a SNS and their potential for creating learning possibilities and opportunities needs to be further explored.

Limitations
This study has limitations in its methodology and results interpretation; these limitations indicate avenues for further research. As with any questionnaire-based survey, one concern is that participants" responses may have been subject to incorrect estimates, biases, faulty memories, and other similar problems (Robson, 2002). Another thorny issue is that several questions in the questionnaire were complex or double-barrelled, which is against good practices in question wording (Robson, 2002 11. Indicate your agreement (or otherwise) with the following statements related to developing your skills in several socialization aspects through Facebook use. 12. Indicate your agreement (or otherwise) with the following statements related to how Facebook as a software tool facilitates your learning.