Graduate Students' Communication Practices and Perceived Sense of Community: An Examination of Information Sources

This study explores the relationship between graduate students' communication practices and their perceived sense of program-level community within a graduate program. The program under study here consists of both traditional on-campus students, as well as students who take classes via a distributedlearning model. This particular program has made a concerted effort to improve its distance education offerings, and to improve information provision on its website to improve the students' sense of community within the program. Self-reported data were collected and analyzed from students (both on-campus and off-campus) enrolled in the graduate program at a midsized Southeastern public university. This study contributes to the body of empirical literature on distance education by providing evidence regarding the relationship between student communication practices and students' perceived sense of community within a distance education program.

Given the graduate program's efforts to improve information provision on its programs' website, our interest in examining the data on this program is to analyze communication behaviors, specifically information seeking, and how these information seeking behaviors relate to the students' perceptions of sense of community within the distance education program (hereafter “perceived sense of program community”). We analyze information seeking behaviors because it has been established that, in uncertain environments such as a distance education environment, many individuals use information seeking as a useful strategy for negotiating such an environment (Miller & Jablin, 1991; Morrison & Vancouver, 2000). We then relate the information seeking practices of distance education students to the students' perceived sense of community within the distance education pro• Julie Reinhart, Saint Xavier University, Chicago, 3700 West 103rd Street, Chicago, IL 60655. Telephone: (773) 341-5030. E-mail: jreinhart@sxu.edu gram. We examine perceived sense of program community because of the well-documented empirical evidence that students who have a greater sense of community tend to have an improved educational experience within the distance education courses, and the improved educational experience has the potential to improve retention (Hara & Kling, 2000; McCarthy, Pretty, & Cantano, 1990; Rovai, 2002).

The contemporary distance education literature, discussed in what follows, suggests that carefully-designed courses offered via distance education can be as effective as, or in some cases more effective than, those offered in a more traditional education setting (Overbaaugh & Lin, 2006; Hill, Raven, & Han, 2002; Karatas & Simsek, 2009). Our study goes one step further and investigates an entire program that is offered via distance education. While the field of distance education has established a great deal with regard to what is necessary to create and implement effective distance education courses, there is little empirical evidence as to what is required to create and implement an effective distance education program. In other words, the distance education field has empirically established what is necessary to create an effective distance education course. However, in the empirical literature it is still unresolved as to exactly what is required to assure that students continue taking distance education courses and complete a program. We contribute to the distance education literature by empirically establishing the relationship between the student's communication behaviors and the student's sense of program community.

The literature review discusses, persistence/burnout in distance education courses, sense of community as it relates to distance education courses, and distance education programs. While the literature review places our study in context, it is important to emphasize our focus on the program level, this program-level focus is an important innovation relative to the empirical distance education literature. The literature review discusses communication practices, and presents important information seeking models, and sources of information.

We focus our attention here upon persistence in distance education programs. Persistence here refers to “a college student's academic continuation behavior that leads to successful program completion” (Rovai, Wighting, & Liu, 2005, p. 362). The literature suggests that “students' strong feelings of social community with other students, faculty, and staff, throughout the school setting can produce positive outcomes” (Rovai et al., 2005, p. 372). Retention in distance education is an important issue in the academy, especially given the large number of institutions offering distance education programs. One report stated that 32% of all 2- and 4-year institutions offer college-level degree, or certificate, programs entirely via distance education (Parsad & Lewis, 2008). Clearly, the more programs that are offered completely via distance education, the more important retention becomes across higher education. The sheer scale of distance education programs, and their integral role across the academy motivate, in part, our investigation of perceived sense of program community.

The perception that student retention in distance education classes has the potential to be improved if students within such courses experience a greater sense of community is well documented (Hara & Kling, 2000; McCarthy et al., 1990; Rovai, 2002). Rovai (2002) found a positive, significant relationship between the distance learners' sense of classroom community and students' perceived sense of cognitive learning. It is reasonable to assume that students are less likely to quit their program of study if they perceive that they are learning. While another study found that sense of community in an online class was significantly related to student satisfaction, they did not find that it was significantly related to retention (Drouin, 2008). Clearly, more research needs to be done to understand the relationship between sense of community and student retention.

While the field has furthered our understanding of what is required to help students complete individual distance education courses, our study contributes to the body of research in distance education that examines what programs can do to help students successfully complete all courses within the program. We investigate communication practices and perceived sense of program community, which is one piece of the program retention puzzle.

We draw upon a landmark study by McCarthy et al. (1990) that investigated burnout in university students. They found a positive relationship between the student's levels of perceived sense of community and their levels of being “burned out.” Additionally, they found that burnout was also related to academic performance. McCarthy et al.'s (1990) research suggests that universities need to consider more than just academics when it comes to the university students' experiences; physical, social and organizational structure that affect students' experiences also need to be considered. McCarthy et al.'s research is related to our study in that it looks at the students' sense of community within their entire collegiate program, not simply one class.

Definition. Sense of community is an amorphous concept, to be sure, yet a widely accepted definition of “sense of community” is “a feeling that members have of belonging, a feeling that members matter to one another, and to the group, and a shared faith that members' needs will be met through their commitment to be together” (McMillian & Chavis, 1986, p. 9). We use the McMillian and Chavis (1986) definition.

Community Within the Classroom. Many comparative studies have tested for differences in the perceived sense of community across two or three different types of courses. Rovai and Jordan (2004) compared the perceived sense of community in blended distance education courses, to that perceived in exclusively online courses, and to that perceived in traditional, exclusively on-campus, courses. Rovai and Jordan conclude that students in the blended courses have a stronger sense of community than do students in the other two types of courses.

When comparing entirely online courses to entirely on-campus courses, Rovai et al. (2005) report that:

Furthermore, the results of the Bernard et al. (2004), meta analysis corroborate the Rovai et al. (2005) findings that online students' experience a weaker sense of connectedness. Bernard et al. (2004) report a positive and significant retention effect in favor of classroom over online instruction. In contrast, Pigliapoco and Bogliolo (2008) found no difference in sense of community when comparing online students with their face-to-face counterparts.

Sense of community in distance education programs is clearly an important variable that some argue varies between exclusively on-campus, blended, and exclusively online courses. Whether or not it varies systematically is still an unresolved question in the literature.

The empirical regularities in differences between the online versus the classroom environment, established by Rovai et al. (2005), and Bernand et al. (2004), with respect to “connectedness” and “retention,” respectively, might otherwise be assumed to persist in general in the absence of a positive effort to manage “connectedness” and “retention” in the online environment. However, the extant empirical evidence suggests that sense of community is something that can be positively affected by careful design of online instruction, which is discussed in what follows.

In fact, several studies have found that students within online courses have a stronger sense of community than students in face-to-face learning situations when online teachers and instructional designers create online instructional environments that encourage community building within the course (Hill, Raven, & Han, 2002; Overbaugh & Lin, 2006). One study found no significant difference in course satisfaction between online learning and face-to-face learning (Karatas & Simsek, 2009). The fact that online instruction can be designed so that there is no significant difference in course satisfaction is a step in the right direction. Equal levels of course satisfaction, in the traditional classrooms and online, imply that online learning, in and of itself, does not negatively impact student satisfaction.

Based on the empirical research, it is clear that instructional design is important for creating online classes in which students succeed and have positive learning experiences (Morris & Finnegan, 2008; Shea, Sau, & Pickett, 2006; U.S. Department of Education, 2009). In fact, much of the contemporary research in distance education has focused upon online learning; this is evidenced by the meta-analysis of online learning studies that was done by the U.S. Department of Education (2009). This landmark meta-analysis found that, “on average, students in online learning conditions performed better than those receiving face-to-face instruction” (U.S. Department of Education, 2009). It has become clear that with carefully designed instruction, and skillful instructors, online learning is a viable instructional medium for learning, and it has been empirically established that course satisfaction can be equal to that of traditional courses (Karatas & Simsek, 2009).

We investigate key determinants of sense of program community within a distance education program that uses the distributed model. The extant empirical evidence largely focuses upon online classes, as discussed above, and suggests that sense of community can be proactively managed. In our study we examine the empirical evidence to further our understanding of what type of program-level initiatives can be designed to improve perceived sense of program community.

Community Within a Program. The extant research indicates that carefully designed distance education courses, with an emphasis upon an online class, can achieve outcomes equal to, or greater than, those achieved in a traditional class. There is, in fact, scant empirical research on entire programs in education that utilize distance technologies. The widely disseminated literature on program-level sense of community, essentially two studies, was published subsequent to the data collection for our current study. Both of these studies, which analyze program-level sense of community, examine programs that are of relatively small scale.

Of these two small-scale studies, which are focused upon the program-level, one is a 2-year longitudinal study of a small cohort in an online program. This study found that communities evolve over time and that the growth in community was, for the most part, based on the community members' efforts to sustain the community (Conrad, 2008). These communities evolved over time because the relationships extended beyond course boundaries, and existed outside learners' formal courses (Conrad, 2008). While instructors carefully design and facilitate the learning environments, the strength of the community, according to Conrad (2008), is due to efforts made by the community itself.

The second of these two studies examining program-level sense of community is another small-scale study. It used a mixed-method design to investigate differences in perceived sense of program-level community between exclusively online students and residential students (Exter, Korkmaz, Harlin, & Bichelmeyer, 2009). The 2009 study found no significant difference in perceived sense of community between distance and residential students. The study did find, through an openended survey question, that

This desire for additional opportunities to connect with other distance students suggests that students in distance education programs see a need for different, more robust, communication tools in an effort to connect with other students within their program.

The previously discussed empirical evidence which reported that masterfully designed online classes can promote a greater sense of community motivates our investigation into what types of planned initiatives can be employed to promote the student's sense of community, and by extension, improve student retention. Our study is specifically focused upon communication practices, more specifically information sharing/seeking sources at the graduate program level. Our review of the literature has found a complete dearth of research studies examining program-level communication practices as it relates to the perceived sense of program community.

A related study established some empirical regularizes with regard to sense of community and frequency of communication (Dawson, 2006), but this was only at the course level, not the program level. Dawson has found that the sense of community is significantly related to the frequency of communication between members of a classroom community. The study measured many sources of communication that the student used, but concluded that further research needs to be done (Dawson, 2006). Dawson's (2006) results linking communication practices to the sense of community, combined with Exter et al.'s (2009) analysis of differences in program-level sense of community between online and residential students, highlights an important gap in the empirical literature. This gap in the literature is the need to improve our understanding of how communication practices are related to the perceived sense of community in distance education programs. The results of our study fill this gap in the literature.

Research studies on program-level information sharing and seeking for distance education programs are not ubiquitous in the literature. Therefore, a variety of information-seeking models developed in related lines of research were reviewed to identify a model that would be appropriate for use in our study.

Important models of information seeking (Leckie, 1996; Ramirez, Walther, Burgoon, & Sunnafrank, 2002; Wilson, Schraefel, & White, 2009) have been vetted in the literature. These models range from models that focus on information seeking only in a computer-based environment, to general models of information seeking in professional organizations (Leckie, 1996; Wilson et al., 2009), to a model of information seeking by newcomers to an organization (Miller & Jablin, 1991). We use Miller and Jablin's (1991) “Model of Newcomers' Information Seeking Behaviors During Organization Encounter” as a framework for the communication portion of the study.

Miller and Jablin's model is utilized for this study because it closely parallels graduate student experiences. A master's student traditionally spends a short period of time completing his or her graduate studies, but typically experiences a high degree of pressure to do well, and succeed within the organization. Miller and Jablin's model (1991) is motivated by the need to help organizations understand the importance of organizational assimilation. Key aspects of Miller and Jablin's model are uncertainty and social costs of information seeking, which are particularly relevant to the students in a master's-level distance education program (Hara & Kling, 2000). Uncertainty here refers to a persons' uncertainty as to their role within the organization, and their need to learn the informal and formal roles (p. 95). Social costs reflect a person's awareness of the potential for negative relational consequences associated with information seeking requests (e.g., worried about “bugging” people or appearing to not be knowledgeable enough; p. 97). Both factors are issues that master's-level distance education students confront, and have the potential to interfere with their learning experiences (Hara & Kling, 2000). Generally speaking “the extent to which newcomers experience behavior-outcome uncertainty should directly affect their communication behaviors” (Miller & Jablin, 1991, p. 95). Therefore, depending upon the perceived social costs of seeking specific kinds of information from particular information sources, newcomers may “select information-seeking tactics that are less overt (detectable by the target), and more covert, in nature” (Miller & Jablin, 1991, p. 97).

Information seekers make choices as to information sources and information content based upon the level of uncertainty and perceived social cost associated with each choice. Miller and Jablin (1991) discuss other influences on information seeking behavior, such as cross-sectional individual differences, and contextual factors. For purposes of our study we focus upon information sources.

Common categories for information sources include (Miller & Jablin, 1991, p. 97):

1. Official, downward, media-related messages from management;

2. Members of the newcomer's role set (i.e., immediate supervisor, coworkers, and subordinates);

3. Other organization members (e.g. acquaintances in different departments);

4. Extraorganizational sources (e.g., clients); and

5. The task itself.

It is generally believed that information seekers tend to focus on members of the newcomer's set, such as supervisors and coworkers, due to the relative availability and access to other sources of information (Miller & Jablin, 1991; Ostroff & Kozlowski, 1993). Additionally, Miller and Jablin (1991) argue that supervisors may be a higher source of socialization because newcomers want to gain their approval. Coworkers are more available to newcomers so they become an important source of information. Louis, Posner, and Powell (1983) report that, in the workplace, coworkers' availability as an information source “was significantly correlated to job satisfaction, commitment, and tenure intention” (p. 864).

Students in distance education programs are typically separated from other students and instructors, both geographically, and in time. This study investigates communication practices, specifically sources of information that students within a distance education program utilize throughout their program. The study examines the relationship between the students' use of information sources with the students' perceived sense of program community. Our goal is to gain a greater insight into the relationship between the information sources and a student's sense of community within their graduate program. Greater insight into, and a refined understanding of, this relationship are essential to proactive design to improve student outcomes. Using our insights, programmatic communication initiatives can be developed to further promote a sense of program community.

Our study investigates the following questions:

1. Is there a difference between on-campus students' perceived sense of community (including sense of social and learning community) as compared to off-campus students' perceptions of sense of community?

2. Is there a relationship between student's perceptions of program sense of community (including sense of social and learning community) and their communication practices?

The graduate program under study is a library and information studies program that consists of eight full-time faculty members with offices on the main campus. The graduate student population in the program is comprised of three groups that are identified by the location where they attend classes in person when they are required to do so. Classes in this program are offered in a variety of delivery modes. These modes include face to face, blended, completely online, or class taught via two-way audio/video conferencing technology. The program consists of five mandatory core classes, in addition to elective classes. Students who are off-campus take the five core courses via two-way audio/video conferencing, which are broadcast from the main campus. Electives are offered in one of these formats: completely online, in a blended format, or face-to-face. There are no cohorts in this program, because the students are not required to move through the program at the same pace and course sequence. Two part-time coordinators are responsible for the logistical coordination at the off-campus locations.

Study participants consist of 85 students enrolled in the graduate program at a major public university in the Southeast. The volunteer rate for this study is approximately 30%. The sample consists of 63 female students, 21 male students, and one student not reporting his or her gender.

Forty-seven of the study participants attend classes on the main campus. Thirty-three of the participants attend classes at a location 90 miles from the main campus. Five of the participants attend classes at a location 200 miles away from the main campus. Seventy-seven percent of the participants are part-time students who take one or two classes a semester, while 63% of the respondents work full-time outside of school.

Given that all the participants in this study are enrolled in this one graduate program at a major university in the Southeast, the findings may not apply directly to all other programs. While every effort has been made to avoid bias, the respondents in this survey study are, of course, volunteers and so the respondents may be predisposed to reflect the best perspectives on communication practices within the program.

Sense of Community. To measure the student's perception of sense of program community (SPC), the study used a modified version of the Sense of School Community Index (SSCI) by Rovai, Wighting, and Lucking (2004), which has been shown to be valid and reliable. The SPC contains two subscales; these subscales are: the learning community subscale, and the social community subscale. These subscales are in the original SSCI as well. The SPC inventory includes 10 items from the SSCI that were modified to pertain to the “program” rather than to the “school.” For example, one item from the SSCI is “I feel that this school satisfies my educational goals.” The modification for the current study is “I feel that this program satisfies my educational goals.” For each of the 10 items that address the sense of program community, participants are asked to choose one of five Likert scale responses that include: strongly agree, agree, neutral, disagree, and strongly disagree. These responses are then coded as strongly agree = 1, agree = 2, and so on.

Rovai, Wighting, and Lucking (2004) report that the SSCI is both valid and reliable. We use Cronbach's coefficient alpha to evaluate reliability by measuring the internal consistency for the modified version of the SSCI, the SPC, Cronbach's alpha is .853, which indicates that the SPC is reliable. The Cronbach coefficient alpha for the sense of learning community is .783, and the coefficient alpha for the sense of social community is .868; therefore, we conclude that sense of learning community and sense of social community are reliable.

Information Sources. Preferred information sources were measured using self-report measures at the same time that the participants completed the SCP inventory. Examples of items regarding preferred communication sources include: How often do you check your university email account? How often do you refer to the program website for news and information? We examine the relationship between various information sources and the perceived sense of program community (as well as perceived sense of learning community, and perceived sense of social community).

Miller and Jablin's (1991) common categories for information sources were coupled with communication sources utilized by information providers in the program (i.e., faculty, administration, etc.) that are involved in this study. The top three levels from Miller and Jablin's communication categories were used, as these are the three levels that directly relate to the community that is examined in our study. Miller and Jablin's top three levels communication categories are: first, official downward messages from administration, second, members of student's role set, and third, others within the organization. Each of the three categories is further disaggregated into two specific measures of communication. The first category is measured by frequency of checking website, and frequency of checking e-mail, respectively. The second category is measured by frequency of communication with faculty in current course, and frequency of communication with students in current course, respectively. The third category is measured by frequency of communication with faculty in the program but who are not currently teaching the course, and frequency of communicating with students in the program but not currently in course, respectively. Miller and Jablin's three levels, and the two specific measures used here for each level, as well as other examples of communication types frequently associated with each level, but not directly measured in the current study, are set forth in Table 1.

Our study uses a bivariate correlational design to investigate the relationship between the frequency with which a student utilizes each type of information source and the student's sense of program community. We also separately analyze the two subscales of sense of program community: social community, and learning community. Our bivariate correlational analysis explores the relationship between each information source and three dependent variables. These three dependent variables are: the student's sense of program community, the sense of social community, and the sense of learning community. For each of the three dependent variables ANOVA analysis is used to test for differences between the on-campus and the off-campus students. T tests are used to test for differences in student responses to sense of social community and sense of learning community.

All currently enrolled students in the program were asked to complete the online inventory. The inventory was located in a passwordprotected course management system to which all students in the program had access. The students' responses were anonymous, yet the course management system was able to track those who completed the inventory to assure that a student only completed the inventory once.

Descriptive statistics are provided in Table 2 for the overall sense of program community. The statistics for the full sample is shown in the second column. The sample is then partitioned into on-campus students and off-campus students. The results for the partitioned samples appear to the right of the statistics for the full sample. The maximum score is 40 for sense of program community. The maximum score is 20 for both sense of social community and sense of learning community. A higher score indicates a higher positive response for the items related to the construct of sense of program community. As with sense of program community, a higher score indicates a higher positive response to the items related to the constructs of sense of social community and sense of learning community, respectively.

ANOVAs indicate no significant differences between on-campus and off-campus groups in regards to each of the three dependent variables: sense of program community, social community, and learning community. T tests are used to test whether mean scores are different for the subscales: sense of social community and sense of learning community. For the full sample, T tests revel that the means for sense of social community and sense of learning community are significantly different at the .001 level of statistical significance (T = 11.545, df = 84). The mean is higher for sense of learning community.

T tests reveal that the means for the on-campus students were different for the sense of social community and sense of learning community subscales at the .001 level of statistical significance (T = 7.470, df = 46). The mean is higher for sense of learning community. For off-campus students the T tests reveal that the means for the subscales of social community and learning community are significantly different at the .001 level of statistical significance (T = 9.279, df = 37). The mean is higher for sense of learning community.

Correlational analysis showed positive, statistically significant correlations between sense of program community and both sense of social community, and sense of learning community, respectively. These correlations are .889 and .802, respectively. There is a positive, statistically significant correlation of .440 between the two subscales: sense of social community and sense of learning community.

Results of the correlational analysis indicate positive and statistically significant correlations between perceived sense of program community and each of the following: frequency of viewing the departmental website, frequency of communicating with instructors in current course, frequency of communicating with students in the current course(s), frequency of communicating with faculty who are not teaching the course, and frequency of communicating with students not in the course (the correlations are .354, .254, .388, .259, .290, respectively; see Table 3). Correlations are significant at the 5% level of statistical significance in a two-tailed test of the null hypothesis that the correlation is equal to zero.

Correlational analysis finds positive and statistically significant correlations between perceived sense of social community and each of the following: frequency of viewing the departmental website, frequency of communicating with instructors in current course, frequency of communicating with students in the current course(s), frequency of communicating with faculty who are not teaching the course, and frequency of communicating with students not in the course (the correlations are .336, .300, .467, .329, .420, respectively; see Table 3). Correlations are significant at the 5% level of statistical significance in a two-tailed test of the null hypothesis that the correlation is equal to zero.

Furthermore, correlational analysis finds only one positive and statistically significant correlation between sense of learning community and all the independent variables. The frequency of viewing the program's website is the only independent variable correlated with sense of learning community (correlation .257; see Table 3).

Is there a difference between on-campus students' perceived sense of program community (including sense of social and learning community) as compared to off-campus students' perceptions of sense of program community?

The descriptive results indicate no significant difference between on-campus and off-campus for sense of program community, social community, and learning community, respectively. Our findings indicate that the sense of community, whether program, social, or learning is not significantly related to the student being on-campus or off-campus. This finding supports the reports in the empirical literature that the student in a carefully designed online class perceives a sense of community that is no different than that perceived by the on-campus student (Karatas & Simsek, 2009; Overbaugh & Lin, 2006; Raven & Han, 2002). It is important to note that while the literature has largely focused more narrowly upon comparing the sense of community in the online versus on-campus class, we have, of course, taken a more broad approach, and examined the sense of program community. We establish empirically here that there is no important difference in the sense of program community regardless of whether the student is on-campus or off-campus.

Our results are therefore consistent with the assertion that the program's communication sources, in this study, have affected the same sense of program community for both the student that is on-campus, and for the student that is off-campus. This is important, because of what might appear to be an informational advantage that the on-campus students have over the off-campus students, given the physical proximity of on-campus students to departmental and faculty offices, and informal gathering areas on the main campus. However, we find that no important associated impact is in evidence with respect to on-campus versus off-campus.

Students, however, report a significantly higher sense of learning community than they do a sense of social community. These positive and significant differences were found in the entire sample of students in the program, and also in the partitioned samples of: on-campus, and off-campus students, respectively. Most likely this is due to the nature of the population for the study. Two thirds of the population of this graduate-level program attends part time. Given the nature of the population, it is likely that students are more focused on the learning community, as compared to the social community. However, further research should be done with other populations to examine any potential important systematic differences between learning and social communities.

Is there a relationship between students' perceptions of program sense of community (including social and learning community) and their communication practices?

We begin our discussion on this question starting first with looking at the results as they relate to communication sources outside of the students immediate role set, others in the organization, then as they relate to communication sources within their immediate role set, and then as they relate to official, downward messages.

The information sources from others within the organization include instructors and students outside of the student's current classes. The significant correlations between perceived sense of program community and both frequency of communication with instructors and frequency of communication with students outside the student's classroom make sense given Miller and Jablin's (1991) model. Generally speaking, individuals who seek information from outside their role set are typically looking for appraisal or relational information rather than information directly related to what they are currently working on (Miller & Jablin, 1991). The model provides support for our finding of a lack of a significant correlation between the students' sense of learning community and each of the two sources of information outside their role set. There is no reason to think that members outside their role set would be directly related to what the student is learning in class, and therefore our finding of no significant correlation between the sense of learning community and the two sources of information outside their role set is consistent with our intuition.

Our findings that communication with others within the organization is positively and significantly correlated with both sense of program community, and with the sense of social community, is consistent with Conrad's (2008) finding that community exists outside the boundaries of the classroom. Therefore, if community exists beyond the boundaries of the classroom, then it is not surprising that greater communication with members outside of the students' role set is significantly positively correlated with both the sense of program community, and the sense of social community.

It is not surprising that communication within the newcomers' role set, which, of course, includes teachers and students in courses in which they are currently enrolled, is significantly related to a student's perceived sense of program community and to the subscale of sense of learning community, respectively. This is likely because current teachers and students availability and their access to other sources of information (Miller & Jablin, 1991; Ostroff, & Kozlowski, 1993). This finding supports prior research findings on information-seeking behaviors and job satisfaction that job satisfaction and tenure intention are related to coworkers' availability as information resources (Louis et al., 1983). Furthermore, it seems reasonable that frequent communication with current teachers may be highly correlated with perceived sense of program community, as some students seek instructor approval within the community (Miller & Jablin, 1991).

However, it is interesting that the subscale of sense of learning community is not related to information sources within the members' current role set. Further research needs to be done to better understand why there is no correlation. A study that investigates the types of information students seek, as well as information sources, would help further our understanding of why sense of learning community is not correlated with the information sources that are in the students' current role set.

The major finding with the correlational analysis here relates to the official, downward messages from the university and program community. In this study the two sources are the university e-mail and the program's website. The frequency with which students view the program website is positively correlated with the perceived sense of social community, and also positively correlated with both subscales: sense of social community, and sense of learning community, respectively. In some sense the positive correlation between the frequency of website viewing and each of the community scales is consistent with our intuition, given the resources that this program, like many other programs across the country, has committed to making the website a current and viable source of information for many program aspects.

Our finding that the program's website is the only information source that is correlated with students' perceived sense of learning community is somewhat surprising. Another surprising result is that regularly reading university e-mail was not correlated with the student's sense of program community, nor with sense of social community. This result stands in contrast to the heavy emphasis most universities place on students checking the university e-mail. While it is clearly important that students check their university e-mail for logistical and official business, our study suggests that it is not e-mail communication that helps to promote a sense of community within the program (see Table 3).

Our surprising result with respect to checking university e-mail could be due to the fact that e-mail is traditionally considered a “push” form of communication, meaning that the information is pushed out to recipients. This push is in contrast to the information on a website being “pulled.” When individuals use websites they actively choose to go to the site and pull information, based on their specific informational needs. Depending on the perceived social costs of seeking information, it is possible that students prefer seeking specific types of information that are more covert in nature (Miller & Jablin, 1991). Using the program's website to find information is covert in nature, and can reduce social costs of information seeking, associated with “bothering” people with questions, or else the social costs of appearing to not be knowledgeable enough (Hara & Kling, 2000; Miller & Jablin, 1991). Therefore, the finding that the frequency of checking the university website is significantly related to sense of program community (including the two subscales) makes sense given that students are able to actively pull information from the website at the time of need, and accessing the website can be done at will, without incurring the associated social costs that might otherwise be associated with information seeking.

While most would agree that the program's website is an important communication tool, it is interesting that regularly reading the website is significantly correlated to students' perceived sense of community, while regularly checking the university e-mail is not. These findings imply a trend away from e-mail communication towards more web-based communication if the goal is to create a sense of program community. This finding is anecdotally supported by the exponential increase in popularity of social networking sites (Facebook, 2010a). Our results also imply that distance education students desire more social, Web-based forms of communication to promote a greater sense of community (Exter et al., 2009). It is important to note that data collection for our study was conducted prior to Facebook and other social sites reaching their current level of prominence as to communication media outlets for college students (Facebook, 2010b).

Further research needs to be done to refine our understanding of why the website is the only communication source significantly correlated with the sense of learning community. Future research should include a careful investigation into what types of information individuals seek when they use the website, as well as other sources. Our exploratory study does not focus upon examining the types of information.

The literature suggests that carefully designed instruction can create a stronger sense of community within a class that, in turn, has the potential to increase retention. This exploratory study goes one step further, investigating a distance education program, and considers information sources and how they are related to creating a sense of program community. Our findings suggest that program websites are important communication tools, the use of which is significantly correlated with the student's sense of program community. In this study, students who visited the program website more frequently reported a higher sense of program community. Our findings do suggest that programs should carefully utilize their website for program communication and take into account the influence that such communication has on student's sense of community within the program (including both sense of social community, and sense of program community).

Further research needs to be done to further refine our understanding of the relationship between information sources and sense of program community. Investigating the types of information that students seek from the various sources will be an important next step for future research. Additionally, expanding the study to different types of distance education programs would also enhance our understanding in this important area.

Given the current climate with web-based, social networking communication, the study supports current trends in web-based communication and suggests that programs that are interested in enhancing the sense of program community, with its attendant implications for student retention, focus their resources upon communicating via the program website.