Sense of Community in Web Environment Master’s thesis April 2001 Abstract The study of social phenomena in the World Wide Web has been rather fragmentary, and there is no coherent, reseach-based theory about sense of community in Web environment. Sense of community means part of one’s self-concept that has to do with perceiving oneself belonging to, and feeling affinity to a certain social grouping. The present study aimed to find evidence for sense of community in Web environment, and specifically find out what the most critical psychological factors of sense of community would be. Based on known characteristics of real life communities and sense of community, and few occasional studies of Web-communities, it was hypothesized that the following factors would be the most critical ones and that they could be grouped as prerequisites, facilitators and consequences of sense of community: awareness and social presence (prerequisites), criteria for membership and borders, common purpose, social interaction and reciprocity, norms and conformity, common history (facilitators), trust and accountability (consequences).
In addition to critical factors, the present study aimed to find out if this kind of grouping would be valid. Furthermore, the effect of Web-community members’ background variables to sense of community was of interest.
In order to answer the questions, an online-questionnaire was created and tested. It included propositions that reflect factors that precede, facilitate and follow the sense of community in Web environment. A factor analysis was calculated to find out the critical factors and analyses of variance were calculated to see if the grouping to prerequisites, facilitators and consequences was right and how the background variables would affect the sense of community in Web environment. The results indicated that the psychological structure of sense of community in Web environment could not be presented with critical variables grouped as prerequisites, facilitators and consequences. Most factors did facilitate the sense of community, but based on this data it could not be argued that some of the factors chronologically precede sense of community and some follow it.
... create conflict between even the closest individuals. However, sense of belonging in a community does not require the history and memories of ... whole life. Some people believe that technology is a leading factor in the destruction of communal values. Television and video games ... their families (Paige 13-14). Ease of mobility is another factor that leads to weakened communal values. According to Genaro C ...
Instead, the factor analysis revealed that the most critical factors in sense of community in Web environment are 1) reciprocal involvement, 2) basic trust for others, 3) similarity and common purpose of members, and 4) shared history of members. The most influencing background variables were the member’s own participation activity (indicated with reading and writing messages) and the phase in membership lifecycle (from visitor to leader).
The more the member participated and the further in membership lifecycle he was, the more he felt sense of community. There are many descriptions of sense of community, but the present study was one of the first to actually measure the phenomenon in Web environment, and that gained well documented, valid results based on large data, proving that sense of community in Web environment is possible, and clarifying its psychological structure, thus enhancing the understanding of sense of community in Web environment. Keywords: sense of community, Web-community, psychology of the Internet of contents Acknowledgements 1 Introduction 1. 1 Defining Community 1.
2 Defining Sense of Community 1. 3 Psychological Structure of Sense of Community 1. 3. 1 Prerequisites 1. 3. 1.
1 Awareness 1. 3. 1. 2 Sense of Social Presence 1. 3. 2 Facilitators 1.
3. 2. 1 Criteria for Membership and Borders 1. 3. 2. 2 Collective Purpose 1.
3. 2. 3 Social Interaction and Reciprocity 1. 3. 2. 4 Norms and Conformity 1.
3. 2. 5 Roles and Social Structure 1. 3.
... the school, family I am not saying that social environment is the definitive factor in someone's lifetime ... out and grab the recognition they deserve. Social environment as a determining agent is definitely a human ... testing. The student from the upper-class community will have the best chance of succeeding. ... one has a good mind, trying to study in a gang-ridden neighborhood with constant gunfire ...
2. 6 Common History 1. 3. 3 Consequences 1. 3. 3.
1 Trust 1. 3. 3. 2 Accountability 1.
4 The Aims of the Study 2 Methods 2. 1 Materials 2. 2 Participants 2. 2. 1 Studied Web-communities 2. 2.
2 Respondents 2. 3 Procedure 2. 3. 1 The Pilot 2.
3. 2 The Study 2. 4 Measures 3 Results 3. 1 The Critical Factors 3.
2 The Prerequisites, Facilitators and Consequences 3. 3 The Effects of the Background Variables 4 Discussion 4. 1 Reciprocal Involvement 4. 2 Basic Trust for Others 4. 3 Similarity and Common Purpose 4.
4 Shared History 4. 5 Improvements and Future Research Issues 4. 6 Conclusions 5 References 6 Appendices 6. 1 Appendix 1: Questionnaire 6. 2 Appendix 2: Screenshots of each site’s discussion pages 6. 3 Appendix 3: Graphical presentations od background variables’ effects 1.
Introduction For decades the fields of social psychology and anthropology have studied communities, and the characteristics of communities are rather well known (e. g. Napier & Gershenfeld, 1993; Carrol & Rosson, 1997).
Researchers interested in psychology of the Internet have recently begun to explore whether communities could appear in the Web environment as well (e.
g. Valtersson, 1996).
Most of the studies regarding online-communities are either case studies, concentrating on single online-communities like WELL (e. g. Smith, 1994), or on single online-community-related characteristic like awareness (e. g.
Dour ish & Bly, 1992).
Furthermore, these studies have focused on the community, not directly on sense of community of individual members. Sense of community means the part of one’s self-concept that has to do with perceiving oneself belonging and feeling affinity to a certain social grouping. The study of social phenomena in the Web environments is rather fragmentary and there is no coherent theory in this field.
The purpose of the present study is to give more understanding to the issue and look for evidence of individuals’s ense of community in the Web environment and find out what are the most critical psychological factors that affect the sense of community in the Web environment. It is not known how much the Internet really connects people, or if it just produces isolation and false sense of connectedness, thus decreasing psychological well being. There is evidence supporting both views. For example, greater use of the Internet is associated with a decline in social involvement and an increase in loneliness and depression (Kraut et al. , 1998) whereas a study by Hampton and Wellman (2000) suggests that the Internet strengthens social relations and community involvement. This study does not directly deal with the effect of the Internet on a person’s real life social involvement, but it looks for evidence that could confirm that individuals can feel sense of community in a Web environment.
... the most of volunteers: a study shows volunteers are giving their time in exchange for community and social benefits. Parks and Recreation. Trochim ... specific associations because they believe that such activities will render community and social benefits. The testable and more specific hypothesis, on ... actually stay for a long period of time, or leave groups which they became a part of is the theme of ...
If a person feels sense of community, he is socially (although virtually) involved, which is connected to psychological well being (e. g. Cohen & Wills, 1985).
1 Defining Community Most definitions of a community stress out the same aspects. For example, Schechter (1998) defines a community as a social grouping, which has the following properties: shared spatial relations, social conventions, a sense of membership and boundaries, and an ongoing rhythm of social interaction. According to Preece (2000), groups that share important resources, provide social support and show reciprocity can be considered communities. In terms of social dynamics, there are many similarities in traditional and Web-communities: both involve developing a web of relationships among people who have something meaningful in common e.
g. hobby, profession or political cause (Kim, 2000).
Although many characteristics of a traditional community also apply to Web-communities, they probably manifest themselves differently due to the computer-mediated nature of online interactions. In a Web environment there is a distinction between network communities and community networks. Network communities are born and exist only in the Web, whereas community networks are real life communities that have happened to make themselves Web sites (Carroll & Rosson, 1997).
In order to find out, whether individuals can feel a sense of community purely based on online interactions, this study concentrates on network communities.
... . Student Signature (electronic) Barry S Krembs II ABSTRACT While Social Media and Web Services are a part of life and an everyday ... are Strategic for Business (Imperva, 2012a). Project Scope While Social Media and Web Services are a part of life and an everyday ... to function as a business (Imperva, 2012a). Conclusion While Social Media and Web Services are a part of life and an everyday ...
1. 2 Defining Sense of Community Defining sense of community brings out the individual’s perspective: how do individual people feel about their social environment, and do they consider it as a community that they are a part of. Sarason (1976) defines the psychological sense of community as a sense of mutual responsibility and purpose – a feeling of being a part of a group one can depend on and contribute to. Sense of community has to do with the affinity of community members; they feel they belong together, are similar with each other and like one another. Tajfel (1981) defines social identity as that part of the individual’s self-concept, which derives from his knowledge of his membership in a social group, together with the value and the emotional significance of that membership. Therefore, the concept of social identity is closely related to sense of community.
In their social identity theory Tajfel & Turner (1986) suggest that in seeking self-enhancement or positive self-identity, individuals characteristically categorize people so that they favor members of the group to which they themselves feel they belong. Social identity is based on the human capability to perceive something that relates to us in symbols and language (Kaunismaa, 1997).
This capability is the same, whether the individual is in a real social environment or in a Web environment. In this study, the sense of community is understood in line with Sarason, Tajfel and Kaunismaa: sense of community is a part of one’s self-concept, that has to do with perceiving oneself as belonging and feeling affinity to a certain social grouping. It is known that in real life people who share a social identity perceive themselves to be more similar to each other. They are also more likely to co-operate, feel a stronger need to agree with group opinion, perceive in-group messages to be of higher quality and conform more in both behavior and attitude (Nass, Fogg & Moon, 1995).
Social identity can be manipulated with minimal cues (e. g. group members wear similar badges) – at least in real life and in small groups. It is also possible to induce a psychological group formation between a human and a computer (Nass et al. , 1995), a fact that emphasizes the ease and flexibility of group identity formation.
... and not engaging in bad habits. The second social group, which is made up of the whites who ... by the boy who makes it to a community college and has a chance of going further ... linguistic codes in comparison with various cultural and social classes; through this they will know that the ... see beyond their immediate despairs and conditions. Both groups experienced a persistent cycle of poverty with negative ...
Even though communities are larger formations than groups, it is still very likely that sense of social identity can be identified in a community of people who interact via the Web. 1. 3 Psychological Structure of Sense of Community Several factors can contribute to sense of community in the Web environments, and they can be extracted from the definitions of sense of community. Some of them can be extracted from the known characteristics of real life communities and the studies of Web-communities.
In the lack of any coherent framework of sense of community in the Web environment, it is suggested here that based on the known characteristics of these factors they could be grouped as prerequisites, facilitators and consequences of sense of community as shown in Figure 1. Justifications why each factor has been included to its group are presented in the following chapters. Figure 1. The possible contributing factors of sense of community grouped as prerequisites, facilitators and consequences.
1. 3. 1 Prerequisites Due to lack of usual social cues and the transitory nature of the online interactions, there are presumably certain prerequisites for the development of the sense of community in a Web environment. These prerequisites are awareness of others and sense of presence (both self and others) in a Web environment. 1. 3.
1. 1 Awareness There are many dimensions in awareness, but this study focuses on awareness as an ability to maintain and constantly update a sense of one’s social and physical context (Schlichter, 1998).
A simple example of awareness is that when on a certain Web site, a person knows that there are also others there at that moment. Greenberg, Gut win & Cockburn (1996) have specified the concept of group awareness that consists of four different kinds of knowledge of what is going on. Informal awareness is basic knowledge of who is around in general and where people are located relative to you.
Group-structural awareness involves knowledge about e. g. roles and responsibilities. Social awareness is the information that a person maintains about others in social or conversational context (e. g. are others paying attention).
... The chart below shows some of the social web sites available and the number of registered ... 1994) and Tripod.com (1995). These early communities focused on bringing people together to interact with ... quarter in 2008, Facebook reported 67 million members, Myspace occupied 100 million users, and YouTube ... Additionally, users may join common-interest user groups, organized by workplace, school or college, ...
Workplace awareness involves knowledge about how the others in the group interact. Awareness can be thought as a prerequisite for sense of community to develop for at least two reasons. Firstly, perceiving, recognizing, and understanding the activities of others is a basic requirement for adequate human interaction and communication in general (Sohlenkamp, 1998).
Secondly, when an individual knows others are there, it brings norms into play to guide behavior (Erickson & Kellogg, 2000).
1. 3. 1. 2 Sense of Social Presence Closely related to awareness is the sense of social presence. Short, Williams and Christie (1976, 65) have defined social presence as “a perceptual or attitudinal dimension of the user, a mental set towards the medium.” It is a feeling of being socially present with another person at a remote location.
The difference to awareness is that sense of presence also includes the notion of self being present at a given location. Media perceived as having a high degree of social presence are judged as being warm, personal, sensitive, sociable and active (Short et al. , 1976).
Even though sense of social presence is important, it has been noted that in order to develop social identity on the Web, actual spatial co-presence is not required (Lea & Spears, as cited in Preece 2000).
1. 3. 2 Facilitators Awareness of others and sense of social presence are not enough for sense of community to develop. There should also be certain identifiable characteristics of communities, which are likely to facilitate the sense of community. These facilitators include criteria for membership and borders, collective purpose, social interaction and reciprocity, norms and conformity, roles and social structure and common history. 1.
3. 2. 1 Criteria for Membership and Borders In most groups and communities there are criteria for membership (Carrol & Rosson, 1997).
When an individual has met the criteria and has been accepted as a member of the community, he will probably value the membership more and his perception of belonging can develop.
In consequence of the criteria, there are “borders” of community, determined by the knowledge of who belong to the in-group and who to the out-group (Napier & Gershenfeld, 1993).
This affects social behavior, emphasizing the affinity to in-group and possible prejudices against out-group (Helkama, Myllyniemi & Lieb kind, 1998).
From an individual member’s point of view, one indicator of the existence of borders is that he communicates differently with other members than non-members (Napier & Gershenfeld, 1993).
3. 2. 2 Collective Purpose An important facilitator to feelings of affinity is collective purpose (Carrol & Rosson, 1997).
The purpose can be for example a common interest, need, value, concern or activity that gives a reason to the community. When members can give common reasons for being in the group, there is a sense of purpose among the members (Napier & Gershenfeld, 1993).
The collective purpose also defines the goals of the community.
Recognizing the shared purpose helps the individual also feel a sense of identification and unity, and get a better idea of who he is and what he is a part of (Bressler & Grantham, 2000).
Naturally individuals have their personal goals as well. The more these goals are in line with those of the community, the stronger sense of collective purpose the individual can feel. 1.
3. 2. 3 Social Interaction and Reciprocity Social interaction is one of the obvious prerequisites for community to exist in any environment. If there is a collective purpose and shared goals, community members usually work together and interact to reach those goals. Social interaction is also likely to increase after there is sense of community among the members.
In the Web environment most interactions are text-mediated, and hence, face-to-face and online interaction are clearly different. In online interactions there is often a lack of co-presence (being in the same place at the same time), which reduces the non-verbal communication and makes interactions asynchronous. However, online interactions have the advantage of being reviewable and revisable, because the written messages persist (Preece, 2000).
From individual perspective reciprocity of information, support and services is one of the most meaningful characteristics of social interaction. A person who develops a strong social regard and identity with his social system is likely to make return offers for help, although anonymity may tempt to only take and not repay (Preece, 2000).
Relatedness, reciprocity and mutual understanding are among important facilitators on achieving sense of community (Bressler & Grantham, 2000).
It is not certain whether or not one’s own participation is obligatory for the sense of community in Web environments. According to a study by Valtersson (1996) the sense of community is difficult to achieve without one’s own participation, whereas a study by Nonnecke & Preece (as cited in Preece, 2000) concluded that it is, in fact, possible to feel a strong sense of community without ever participating. Nonnecke and Preece focused on lurkers in Web-communities (Lurker is a person who does no participate, only observes).
Their interviews of lurkers showed that lurkers can become so immersed in the community’s discussions, that they feel they know other participants and that they belong to the community.
In line with the view of one’s own participation as optional, Bressler & Grantham (2000) emphasize any kind of involvement over active participation. So presumably the more involved the individual feels with the community -whether he participates or not – the more likely he is to feel the sense of community. 1. 3. 2. 4 Norms and Conformity Communities have norms, implicit or explicit rules and standards, which guide the group and define the range of acceptable behaviors of individual members.
Norms regulate the performance of a group as an organized unit, keeping it on course towards its objectives. In real life norms are mostly invisible; they develop by subtle, beyond-awareness processes of inference (Napier & Gershenfeld, 1993).
In a Web environment they may have to be more explicitly stated. If the norms of the group are compatible with an individual’s values and goals, the person will conform to the norms of the group (Napier & Gershenfeld, 1993).
Conformity can appear in computer-mediated co-operation, although somewhat reduced (Smilowitz et al. , as cited in Wallace, 1999).
In general, the conforming influence of the group also reduces when the majority is not consistent and unanimous. If, however, group influences such as conformity, can be observed in the Web environments, it indicates that the individual strongly identifies himself as a member of the community (Helkama et al. , 1998).
1. 3. 2.
5 Roles and Social Structure Communities are held together by a network of social roles. A role defines behavior and responsibilities that others expect from the role-holder (Helkama et al. , 1998).
The social structure of communities becomes evident with the existence of core and periphery, subgroups and leadership (Carrol & Rosson, 1997).
On individual level, recognizing one’s own place in a community’s social structure may facilitate the forming of sense of community, assuming that the place is agreeable. Real life statuses can often be equalized in online environment, and communication skills and quality of ideas may determine one’s influence on others (Suler, 1998).
A participant in a Web community can credit and bolster his standing by contributing to the group. Closely related to roles is the concept of membership lifecycle, which consists of five roles. Before joining the community, a person is a visitor. He does not have a persistent identity in the community, and he probably seeks more information about the community in order to decide whether to join or not.
Once the person has joined, he becomes a novice, a new member who needs to be introduced to the community life. After a while the member becomes a regular member, an established member that is accustomed to participating in the community life. When the member becomes experienced, he may be promoted to a leader, who helps keeping the community running. Finally, the member becomes an elder who shares his knowledge, passes along the community traditions and values, and acts as an advisor (Kim, 2000).
It can be assumed that the further the individual is on the membership lifecycle, the stronger sense of community he feels. On the Web, the time it takes to advance in a membership lifecycle is condensed.
Although interaction is often asynchronous and stretches over time, the temporal life in the Web proceeds fast. Internet environments change rapidly, and so do memberships of online groups. A person’s subjective sense of time is intimately linked with the rate of change in the world one lives in. In the online environment, the experience of time seems to accelerate, and people can proceed from novice to elder in a relatively short time (Suler, 1998).
1. 3. 2. 6 Common History A community usually has a shared history, which gives the members a feeling of belonging together.
The historical ties that bind the members consist of defining episodes, which are occasions when something memorable or important has happened in the life of the community (Carrol & Rosson, 1997).
The more aware an individual member is of his community’s history, the more likely is a shared identity amongst the members, and the more likely it is for a member to feel that he belongs to the community. 1. 3.
3 Consequences Trust and accountability are among the possible consequences of feeling the sense of community in a Web environment. 1. 3. 3. 1 Trust In the community context trust is the expectation of the members that arises within a regular, honest, and cooperative behavior, based on commonly shared norms (Fukuyama, as cited in Preece, 2000).
In online environment trust indicates a positive belief about the perceived reliability, dependability, and confidence in a person, object or process (Fogg & Tseng, 1999).
Community structure and norms contribute to members’ feelings of security, stability, and trust. Thus, if an individual member expresses trust in other members, it may be a consequence of a sense of community in a Web environment. 1. 3. 3. 2 Accountability Accountability means a person’s willingness to accept responsibility for his actions, and as a consequence, being trustworthy.
In a Web environment where a person is aware of others being present and knows there are certain social rules to follow, he is likely to be more accountable than without these influences (Erickson & Kellogg, 2000).
Therefore, also accountability can be considered as a possible consequence of sense of community in a Web environment. 1. 4 The Aims of the Study The main purpose of the present study was to look for evidence of individuals’s ense of community in a Web environment. Previous studies of real-life communities and occasional studies of Web-communities have indicated that there are factors that are likely to be important in sense of community in Web environment (e. g.
Carrol & Rosson, 1997).
The specific questions that the present study aimed to answer were: 1. What are the most critical psychological factors that affect the sense of community in a Web environment? 2. Can the factors be grouped as prerequisites, facilitators and consequences of sense of community, as suggested in the introduction? 3.
Are there some background variables that affect the sense of community in a Web environment? 2. 1 Materials An online questionnaire consisting of three parts was used (Appendix 1).
First there were questions on respondents’ background information, second there were propositions regarding possible factors of sense of community, and third there was one open question. The questionnaire was in Finnish.
The eight questions about the respondents’ background information included gender and age, Internet usage experience and the duration of community membership. Also there were the member’s estimate of his own activity in the community indicated by the frequency of reading and writing messages, and the member’s estimate of his own place in the membership lifecycle: visitor, novice, regular, expert and leader. (These were otherwise the same as Kim’s (2000) roles, except the original leader and elder were replaced with expert and leader. ) There were 50 propositions, to which the user was asked to express his degree of agreement on a seven-point scale.
The propositions were composed so that they reflected different aspects that precede, facilitate or follow the sense of community. Translated examples of each kind of proposition can be seen below: . I usually know when there are others present in the site (prerequisite / awareness).
People here usually follow the rules of the site (facilitator / norms and conformity).
My words won’t be misused here (consequence / trust) Finally there was an open question: “Tell in your own words why you use the given site and its discussion groups, and what you feel you get from it.” The questionnaires were otherwise similar for each community, except that each community was referred to with its own name. For example the questionnaire for Njet’s members referred to that community by the name Njet.
In this way the propositions were clearer, and it was easy to classify the answers by community. The questionnaire was composed specifically for this study, based on possible factors of sense of community, which were presented in the introduction. This was because this kind of study has not been conducted before, so there were no ready-made questionnaires to use. 2.
2 Participants 2. 2. 1 Studied Web-communities Three Web-communities took part in the study. They all had characteristics that made them suitable for this study, and were selected on the following basis: 1. They all had to be network communities, i.
e. the community existed only in the Web, or at least was initially developed online. 2. The Web sites had to have similar tools for members to interact, preferentially the kind tightly located within the site, so that it would also offer a sense of place. All selected sites had asynchronous discussion forums.
(See Appendix 2 to see screenshots of each site’s discussion page) 3. They all had to have about the same amount of members and activity. 4. The purposes or at least the types of communities had to be similar. The selected communities were all communities of interest. The background information for each site is listed in Table 1.
Table 1. The background information for each site. Web-community K”arkiverkosto web Verkkoklinikka web Njet web Description Service that collects and mediates information about the Finnish information society and its development. Medical portal that provides information and guidance on physical and mental health. A portal dedicated to culture. Based on voluntary content creation.
Type Community of interest Community of interest Community of interest Members (approx. ) 4000 registered users, about 200 of them participate in discussions 11 000 registered users, exact number of active participants of discussions unknown. (Possibly less than 100) 600 registered users, exact number of active participants unknown (Possibly 100-200) Age 2. 5 years 5-6 years 2 years Critical tools Discussion groups Discussion groups, Who’s online-service Discussion groups, Who’s online -service Examples of discussion topics Modern times, locality, navigation, content creation, distant work. Adoption, dieting, childcare, depression, studying, relationships, exercising, smoking. Philosophy & science, graphical heaven, books and theatre, music, feelings, free time.
2. 2. 2 Respondents A total number of 162 persons filled the questionnaire. 49 of them were K”arkiverkosto members, 65 were Verkkoklinikka members and the rest 48 from Njet. 67.
3% of respondents were women. The majority of the respondents (64. 2%) were between ages 20 and 39. Most respondents (65. 4%) had over four years of experience in using the Internet, and almost everyone’s usage was versatile.
Most respondents (64. 6%) had been members of their Web-community from six months to two years. The respondents represented all the phases in membership lifecycle as seen in Figure 2. The largest group considered themselves experts. Figure 2. Distribution of respondents self-evaluated phase in membership lifecycle.
The respondents included both active participants of discussions that read and wrote messages daily and more passive participants who never wrote messages and read them seldom. The more active the participants were, the further they thought themselves as being in the membership lifecycle. Figure 3 shows the distribution of activity of reading and writing messages in different phases of the membership lifecycle. Figure 3. The distribution of activity of reading and writing messages in different phases of membership lifecycle. 2.
3 Procedure 2. 3. 1 The Pilot After the questionnaire was constructed, it was tested in order to validate it and identify the most suitable propositions to be included in the final questionnaire. The test respondents (n = 18) were instructed to think themselves as users of some Web site and its discussion forum. The test respondents were mainly psychology and cognition science students from University of Helsinki. The pilot responses were analyzed with a simple correlation.
Those items that did not correlate at all even though they were assumed to measure the same construct and therefore correlate, were taken into further evaluation. The answers to uncorrelated pairs were counted, and if one of the propositions had received significantly more “I don’t know” -answers, it was removed. Also, a one-way analysis of variance was calculated to see whether answers differed in different phases of the membership lifecycle. If one of the variables in an uncorrelated pair was further away from being significant, it was removed. As a result the original 60 propositions were reduced to 50. 2.
3. 2 The Study The gathering of data was conducted entirely online. The validated questionnaires were placed on the Web, and links to them were added to the communities’ home pages, together with a prompt text encouraging members to participate. This way all members from each community had access to the questionnaire from their own community homepage. Members had approximately three weeks to fill in the questionnaire, and they were able to fill in it when ever and where ever it was suitable for them. This procedure was chosen even though it does not necessarily result in a random or representative group of the general population, nor is the situation truly controllable.
However, the Web-questionnaire was the most suitable method for studying people in Web-communities, especially because the members of the communities had access to the questionnaire directly from their own Web page. Measuring via Internet also has the advantage of resulting in high response rate, and yielding quantitative data within a short period of time. 2. 4 Measures The data gained from the propositions were analyzed with factor analysis in order to find out how the variables would load on factors, and to see if there were factors that could be identified as characteristics of sense of community.
Factor analysis was chosen for the statistical method because it is suitable for analyzing the correlation of factors that cannot be identified directly and to detect the structure in the relationships between variables. The data were statistically analysed with SPSS 8. 0. A correlation matrix was composed out of the variables, the factor pattern matrix was calculated, the scales were formed and reliabilities were calculated for both of them.
In order to see how the open answers supported the data gained from propositions, the answers were counted and grouped in two ways. To see what kind of answers there were, similar answers were grouped together and the groups were titled. To see how the answers might support the factors found in the factor analysis, the answers were grouped according to which factor they supported, if any. To find out, whether some of the variables could be considered as prerequisites, some as facilitators and some as consequences of sense of community, the respondents’s cores in the different propositions were taken into further evaluation. The scores in awareness- and social presence -propositions were summed and then grouped to four groups according to the amount of scores so that group 1 had the lowest scores and group 4 the highest. The distribution of groups in different phases of the membership lifecycle was then examined with analyses of variance and error bars.
The same was done for scores in trust and accountability -proposals and for scores in facilitator-proposals. In order to find out how different background variables might be connected to respondents’s cores in the four factors, the scores in these factors were calculated for each respondent. The scores were then grouped to four groups according the amount of scores so that group 1 had the lowest scores and group 4 the highest. A one-way analysis of variance was then calculated to see whether the respondents who scored low would differ from those who scored high, and in which background variables. In order to see how the groups differed their mean scores and standard deviations were graphically examined. [ Continue to Results ] Results 3.
1 The Critical Factors The factor analysis used to explore the critical factors in sense of community was Maximum likelihood -analysis and the initial factor solution was rotated with Vari max-rotation. The rotated factor pattern matrix revealed four factors, which can be seen in Table 2. Table 2. Factor pattern matrix that shows the four critical factors indicating the sense of community in Web environment.
The P-numbers are the unique identifiers of the propositions. In parentheses are the assumed variables the propositions reflect. Characteristics F 1: Reciprocal involvement F 2: Basic trust for others F 3: Similarity, common purpose F 4: Shared history P 45: Others seeking guidance (Trust).
77 -. 03.
16 -. 10 P 55: People willing to participate (Social interaction, reciprocity).
73. 28. 07 -. 07 P 14: Others correcting inappropriateness (Norms, conformity).
65. 14. 18. 10 P 18: Self revealing personal issues (Trust).
62. 02. 15. 17 P 34: Sociable environment (Awareness, social presence).
28. 18 P 1: Others as reason for visiting the site (Social interaction, reciprocity).
00. 05 P 26: Active environment (Awareness, social presence).
30. 02 P 33: Self seeking guidance (Trust).
57. 09. 42 -. 03 P 9: Knowledge of others being present (Awareness, social presence).
52. 09 -. 07. 31 P 59: Existence of long-time participants (Social interaction, reciprocity).
27. 18 P 25: Existence of subgroups (Social structure, roles).
44 -. 08.
16. 17 P 46: Others lie (Trust).
43 -. 39. 02 -. 14 P 49: Self said something not say able in real life (Accountability).
41. 03. 22 -. 01 P 35: Being oneself (Trust).
40. 35. 15. 10 P 8: Experienced guide newcomers (Membership lifecycle).
40 -. 17.
20. 14 P 28: Existence of core and periphery (Social structure, roles).
33. 12. 19. 27 P 29: Expect others to be fair (Trust) -.
02. 76 -. 03 -. 02 P 51: People follow the rules (Norms, conformity).
63. 06. 01 P 42: Warm environment (Awareness, social presence).
45. 56. 31.
09 P 3: Existence of rules (Norms, conformity).
17. 55. 06. 04 P 2: Others wont misuse one’s own words (Trust) -. 06.
53. 01 -. 23 P 56: One’s own privacy is protected (Trust) -. 06.
53. 09 -. 01 P 53: Self not committed participant (Social interaction, reciprocity) -. 25 -.
47 -. 30 -. 30 P 23: Others not seem real (Awareness, social presence) -. 10 -.
42 -. 10 -. 19 P 21: Difficult to get accepted (Borders, criteria for membership) -. 10 -.
42. 00 -. 15 P 31: Respect for others (Social interaction, reciprocity).
42. 15 -. 10 P 10: Membership is important (Social interaction, reciprocity).
30. 41. 39.
37 P 6: Accountable for own words (Accountability).
04. 39 -. 01. 24 P 20: Self trying not to annoy others (Norms, conformity).
26. 24 -. 02 P 27: Not caring what others think (Norms, conformity) -. 01 -. 25 -. 13 -.
03 P 40: Not being able to get know anyone (Awareness, social presence) -. 16 -. 17 -. 10 -. 15 P 60: Shared goals (Common purpose).
01. 69 -. 03 P 41: Existence of the purpose (Common purpose).
13. 14. 62.
02 P 52: Knowledge of the purpose (Common purpose).
22. 07. 56. 05 P 30: Shared language (Norms, conformity).
09. 55. 13 P 38: Agree with others (Norms, conformity).
12. 44. 52.
13 P 13: Similar values (Common purpose).
14. 39. 51. 18 P 48: Membership as advantage (Social interaction, reciprocity).
36. 42. 48. 22 P 4: Similar interests (Common purpose).
47. 01 P 17: Not having much in common with others (Common purpose) -. 23 -. 29 -. 39 -. 13 P 32: Shared language (Borders, criteria for membership).
17 -. 14. 31. 03 P 24: Self committed to reciprocity (Social interaction, reciprocity).
27. 05 P 57: Knowledge of founders (Common history) -. 16 -. 01 -. 15.
83 P 19: Knowledge of founding (Common history) -. 12 -. 02. 04. 79 P 5: Knowledge of background (Common history).
15. 03. 09. 68 P 36: Impression of number of participants (Awareness, social presence).
03. 51 P 15: Getting to know others (Social interaction, reciprocity).
42. 10. 06. 47 P 47: Recognizing own role (Social structure, roles).
06 -. 11. 05. 39 P 58: Others affecting one’s opinion (Norms, conformity).
19. 19. 37 P 43: Existence of leaders (Social structure, roles).
07. 04. 27 The variables that loaded strongly on the first factor emphasized the existence and the respondent’s awareness of other members’ activity and willingness to participate specifically in reciprocal manner, but also the respondent’s own willingness to be involved. Hence, the first factor was named as reciprocal involvement. The variables that loaded strongly on the second factor accentuated the existence of norms and the impression that members follow them. They also emphasized the respondent’s trust for others and that his words and privacy would be protected.
Hence, the second factor was named as basic trust for others. The variables that loaded strongly on the third factor stressed out the existence and knowledge of shared purpose, and that the members had shared goals, language, values and interests. Also conformity was apparent in the variables. Based on these, the third factor was named as similarity and common purpose. The variables that loaded strongly on the fourth factor highlighted the knowledge of shared background, the founding and founders of the community. The factor was named as shared history.
The reliabilities of factor scales and the standard errors of measurement are presented in Table 3. The reliabilities that indicate the precision of the measurement were counted with the assumption that the errors do not correlate. The standard error of measurement indicates the average error of measurement made when using the scale. Reciprocal involvement -scale had the best reliability, but all reliabilities were noticeably high. Table 3. The reliabilities of factor scales and the standard errors of measurement (SEM).
F 1 Reciprocal involvement F 2 Trust for others F 3 Similarity, common purpose F 4 Shared history Reliability. 886. 873. 812. 819 SEM. 290.
263. 360. 383 The respondents’ most frequently occurring answers to the open question are summarized in Table 4. The question asked respondents to tell in their own words why they use the given site and its discussion groups, and what they feel they are getting from it. The results of grouping the same answers according to which factor they supported are shown in Table 5. Only part of the answers could be seen as reflecting the factors, so the sum of answers in the two tables is not the same.
The answers supported especially the importance of reciprocal involvement and similarity of members. Two respondents even mentioned directly sense of community as their reason for participating. Table 4. Respondents’ reasons for using the site.
N means the number of references. Reason for using / participating the Web-community N “Peer support”, reciprocity 51 Gaining information and tips 38 Having company, acquaintances and friends 29 Keeping up to date 18 Consuming time, just having fun 17 Testing own ideas, a way to affect others 16 Some subjects are easier to discuss via Web 11 Listening to others’ opinions and gaining new point of views 10 Good atmosphere 9 Curiosity towards others’ lives 8 Feeling of affinity 7 Expatriate Finns’ contact to Finland 3 Table 5. The number of answers that reflected different factors Factor N F 1: Reciprocal involvement 71 F 2: Basic trust for others 10 F 3: Similarity, common purpose 30 F 4: Shared history 6 3. 2 The Prerequisites, Facilitators and Consequences It can be assumed that if awareness and social presence are prerequisites of sense of community, then the Web-community members who are in the beginning of the membership lifecycle should stress out these variables as much as the members in the other end of the lifecycle. This, however, was not the case. The respondents’s cores to awareness- and social presence -propositions in different phases of membership life cycle differed (F (4, 156) = 16.