Introducing a twitter discussion board to support learning in online and blended learning environments

  • Published on

  • View

  • Download


  • Introducing a twitter discussion board to supportlearning in online and blended learning environments

    Brian Thoms & Evren Eryilmaz

    # Springer Science+Business Media New York 2013

    Abstract In this research we present a new design component for online learningcommunities (OLC); one that integrates Twitter with an online discussion board(ODB). We introduce our design across two sections of upper-division informationsystems courses at a university located within the U.S. The first section consisted offull-time online learners, while the second section met face-to-face twice a week.Guided by a working theoretical model for how individuals learn and interact withinOLCs, we measure student perceptions of learning, social interaction and coursecommunity before and after our intervention. Initial findings were largely positiveand students across both sections experienced high levels of learning, interaction andcommunity. Our results pave the way for more integrated learning environments thatincorporate online social networking (OSN) technologies.

    Keywords Social learning . Online social networking . Twitter . Online discussionboard . Constructivism

    1 Introduction

    For most individuals, imagining a world without the Internet seems impossible. Fortodays traditional college students, aged 1824 and who comprise 60 % of collegeenrollment, this notion is impossible (U.S. Census Bureau 2009). Referred to asmillennial students, these students have grown up with computers and the Internetwith over 86 % participating in some form of online social networking (OSN) (Jenks

    Educ Inf TechnolDOI 10.1007/s10639-013-9279-3

    B. ThomsSUNY Farmingdale, Farmingdale, NY, USA

    E. EryilmazBloomsburg University, Bloomsburg, PA, USA

    B. Thoms (*)Computer Systems Department, School of Business, Farmingdale State College,2350 Broadhollow Road, Farmingdale, NY 11735-1021, USAe-mail:

  • 2011; Williamson 2007; Oblinger and Oblinger 2005). And todays institutes ofhigher education are learning to adapt for these new dynamic learners. In recent years,colleges and universities around the country are relying on a wide variety of Internettechnologies to facilitate everything from course registration to course delivery.

    In this research we focus on new pedagogical software; specifically asynchronoussocial technologies that provide students and instructors with opportunities to extendlearning outside of the classroom. More specifically, we explore those technologiesthat are widely used by traditional college students on the most popular OSNs such asFacebook, Google and Twitter. Commonly referred to as Web 2.0 software, thesequick and simple-to-use technologies offer students and instructors opportunities tocreate and share knowledge outside the classroom.

    As design science researchers, we are continually exploring IT artifacts acrosshigher education with the greater goal of enhancing learning through interaction andcourse building. In previous design iterations, we integrated Twitter, a microbloggingengine, with a traditional blog within an online learning community (OLC) used atour university. While we found Twitter was successful in providing students with ameans to explore and share new information, there was a lower level of agreementthat our design enhanced levels of interaction, learning and community.

    During this design cycle, we have redesigned our software and integrated Twitterwith our OLCs online discussion board (ODB) and measured its usage acrossmultiple sections of upper-division information systems courses. The first coursewas a fully online course, while the other course met face-to-face two times eachweek. Initial results were largely positive across both sections and we have discoveredthat this new design provides students with a powerful mechanism for building coursecommunity, increasing course interaction and aiding in learning.

    2 Background

    Within higher education, online technologies play increasingly important roles inlearning. While some institutions offer complete online degree programs, more tradi-tional institutions utilize the web to supplement in-class learning, with some offeringcomplete courses online. Market research estimates that 81 % of all institutions will beproviding some form of online learning by 2014. Across each of these institutions,specialized software, known as course management software (CMS), is in a constantstate of flux to meet the demands of academic institutions, classes and, of course,students. With an adoption rate above 96 %, this software plays a critical role in howcourse content is delivered and managed (Educational Marketer 2003). CMS software isdesigned specifically for the facilitation and management of academic courseworkproviding instructors and students with critical resources for learning.

    Our universitys preferred CMS platform is Angel, which provides participantswith a number of interactive features, including email, blogging and an ODB.However, even with its vast range of features, Angel still lacks even basic socialcomponents that traditional college students utilize on a day-to-day basis. And thosesocial components Angel does offer lack even basic features such as avatars and userprofiles. Yet, as the numbers of social software users rises, so, too, will expectationsthat it play a more central role in educational software. And while CMS platforms

    Educ Inf Technol

  • look to find the right blend of social media tools, practitioners and researchers arecurrently developing and integrating existing social technologies into education.

    2.1 Social software across education

    Studies in online collaboration have shown that virtual communication patternscorrespond in similar fashion to real-life communication (Rhode et al. 2004; Redfernand Naughton 2002). Research by Stacey (2002) found that a higher quality ofelectronic communication helps to engage students and aids in their learning of thecourse material. As in face-to-face communication, members of online social learningenvironments are able to state what they think, comment on what others have said,collaborate on common statements, and share information in many forms. Addition-ally, as members of a learning community, students have the right to comment onwhat others have said, collaborate on common interests, and share information inmany forms. Accordingly, online social learning environments offer a valid form oflearning and offer many different methods for students to interact with instructors andtheir peers (Quan-Haase 2005).

    Web 2.0 technologies, such as blogs, microblogs and wikis along with peer-to-peernetworking, discussion and file sharing, empower individuals to take ownership ofthe content they create while also making it easier to pursue social or scholastic tieswith their peers. And increasingly, more individuals are gaining access and familiarizingthemselves with these technologies making their introduction into the classroom more-or-less seamless (Oblinger and Oblinger 2005). In one study, Brescia and Miller (2006)found that the benefits to using blogging in the classroom included enhanced studentreflection, increased student engagement, portfolio building, and better synthesis acrossmultiple activities. In another, Kirkup (2010) argues that academic blogs can lead to theconstruction of an intellectual identity.

    As researchers in social learning software, we look to enhance our existing OLC,one that has already shown success blended learning environments, to discover hownew technologies can further engage students. In prior research we have shown that acorrect formula of software, along with proper alignment with course learningobjects, social software offers students the ability to reflect on course material andexpand in-class discussions to virtual spaces, in addition to improving learning, enhanc-ing social interaction and helping to build course community (Thoms 2012, 2011). Inthis research we explore how the integration of an asynchronous ODB and microblogcan support these same constructs.

    2.2 Online discussion boards

    Online forums have existed since the early days of the Internet and their usagecontinues to grow. From YouTube comments to the Facebook Wall, online forumshelp facilitate discourse across web content. These same tools are represented in theODBs of the most widely used CMS platforms, including Blackboard and Angel.And it is no wonder since ODBs have become the pivotal social component thatprovides numerous advantages.

    For starters, the discourse that takes place across an ODB is not performed in realtime, which offers students more opportunities to prepare, reflect, think, and search

    Educ Inf Technol

  • for additional information before contributing to a discussion (Chen and Chiu 2008;De Wever et al. 2006; Liaw and Huang 2000). Secondly, discussion content persistslong after a discussion concludes, which allows supports to analyze the contributionsof others while collaboratively expanding and deepening their understanding of aparticular subject matter (Hull and Saxon 2009; Solimeno et al. 2008). Lastly, inmany types of online forums, users are presented with certain levels of control, whichprovides users with a sense of empowerment and autonomy.

    In the light of these advantages, the success of an asynchronous ODB acrosseducational environments can be conceptualized as the ability of a system to facilitatecognitive, on-topic, on-task, and sustained discussion among a community oflearners. And these interactions has a reciprocal effect, as identified in Balaji andChakrabarti (2010), where a strong sense of course community among students canincrease interactions across the discussion board. And, as identified in La Pointe andGunawardena (2004), these interactions have a higher impact on students learningoutcomes, than merely studentinstructor interaction.

    3 Microblogging

    Microblogging sites such as Twitter, Tumblr and Jaiku are largely popular and provideindividuals with innovative ways to share information. Microblogs, by their definition,limit the amount of information that can be shared for any given post. This, in turn,requires that the content generator be more selective with what he or she shares.

    We chose Twitter as our microblogging engine for many reasons. A primary reasonwas because it is the most popular and popularity has its advantages, such as a greaterchance students would already have accounts or at least be familiar with the Twitterinterface. Additionally, Twitter has an advanced application developer interface (API),which allows third-party developers to easily integrate Twitter into their own applica-tions. Finally, Twitter still largely focuses on text based posts (140 characters or less),while other microblogs such as Jaiku and Tumblr focus more on media transmissions,including images or videos.

    3.1 Twitter

    The microblogging engine, Twitter, was founded on March 21, 2006 and allows anyuser with a valid email address to create an account for free (Official Twitter Blog2011). Once a user opens an account they are able to post content, view or subscribeto other Twitter feeds and can do so through mobile means as well. Detailed in Fig. 1,Twitter has a very user-friendly web interface.

    Currently, the average number of Twitter posts, or tweets, per day is in theneighborhood of 140 million and, roughly, 1 billion tweets made each week. Addi-tionally, the site averages over 175 million accounts. While it is easy to argue againstthe quality or content of these tweets, for good or bad, Twitter has been successfullyutilized across many industries. Some industries utilize the microblogging site as aprimary mechanism for marketing and to alert customers to upcoming events orpromotions. Another popular area for the microblogging site has been real-time newscoverage. Twitter, along with other social media outlets like Facebook and YouTube,

    Educ Inf Technol

  • was at the forefront of the Occupy Wall Street movement, or the series of movementsacross the U.S. aimed at protesting the economic inequality within the U.S. show-cases how effective Twitter was for communicating information in real-time and whothe dominate social media activists were and how they used it (Chen and Pirolli2012). Within more cooperative settings, such as the workplace, Twitter helps createvirtual water coolers and helps colleagues to get to know one another better (Zhaoand Rosson 2009).

    3.2 Twitter used in education

    With the widespread adoption of Twitter, it was only a matter of time before it wasintroduced into education. Recent research suggests that within an academic environmentmicroblogging adds to community building by offering individuals the ability to contin-ually inform others to what you are doing, discovering or experiencing (Betta 2007).

    This notion is supported empirically, and numerous studies have included Twitterin academic environments. Grosseck and Holotesco (2008) argue that the incorporationof a microblog models good pedagogy and can be responsive to a students learningneeds. Specifically, they argue that Twitter can change the classroom dynamic and offera useful tool to share information. Ebner et al. (2010) concluded that microblogging cansupport learning beyond the traditional classroom through a constant flow of informa-tion between students and between students and teachers. Dunlap and Lowenthal (2009)found microblogging to be a powerful tool for enhancing social presence in addition toestablishing informal, free-flowing, just-in-time communication between and among

    Fig. 1 Twitter system

    Educ Inf Technol

  • students and faculty. Furthermore, Wakefield et al. (2011) found that Twitter can helpincrease understanding of course materials as supported by the interactive environmentand affiliated rapid feedback.

    4 Theoretical model

    Theory plays an integral role in how we create and manipulate our OLC design andhelps guide how new sub-components can facilitate learning, social interaction andcourse community within an academic OLC. Illustrated in Fig. 2, this model iscomprised of three distinct but integrated constructs: individuals, activities andcommunity. Our model incorporates theories of individual learning and classroominteraction (constructivism), peer-to-peer interaction and community (social presence)and the various technology-based activities students perform to accomplish courseobjectives (activity theory). Together, they help guide the design of information systemsartifacts and identifies how individuals will utilize a wide range of technologies to shareideas and information with the larger community.

    Academic communities are a subset of what Lave and Wenger (1991) have coinedcommunities of practice (CoP). In such communities, individuals work togethertowards common goals, collaborating on common problems, sharing best practices,supporting one another and sharing a common identity. At the heart of an academiccommunity is the individual, which we represent in our model by constructivism.Prior research has traced the roots of a CoP to constructivism (Johnson 2001; Palloffand Pratt 1999; Savery and Duffy 1996). Constructivism has proven to be a widelyeffective theory for understanding the multiple dimensions each learner brings into

    Fig. 2 Theoretical model for OLCs in education

    Educ Inf Technol

  • the classroom. Existing research on constructivism attributes learning to the experiencesand interactions a learner encounters within a specific context.When dealing with onlineversus face-to-face learning environments, establishing a dynamic context, one that theuser can manipulate as they see fit is essential. Since our OLC is based primarily on thesocial networking model, learners have a greater freedom to approach learning frommultiple angles, choosing when and where they want to participate.

    Our second layer focuses on the different tools learners will interact with within anOLC. Activity theory provides the appropriate lens for viewing how individualsutilize specific technologies to accomplish course goals. In activity theory activitiesare goal-directed, where multiple ways exist to achieve those goals, oftentimesthrough adaptive means (Bdker 1989). From its origins, activity theory considershuman activities to be directed at objects and mediated by artifacts (Vygotsky 1987).Similar to how constructivism acknowledges the uniqueness of the learner, activitytheory acknowledges, even encourages, unconventional activity, so long as they helpto accomplish specific goals. In educational environments, when instructors are ableto choose activities from both online and face-to-face mediums, they are also able toselect the activity that provides the best fit for any particular learning objective(Heckman and Annabi 2006; Mor et al. 2005). Social software, such as Twitter andfeatures of an ODB, once again, offer learners a variety of different artifacts, fromavatars, to profiles that supplement the specific activity of tweeting and/or discussing.

    For an OLC to be successful, it must become a vibrant space for learner interaction.Social presence theory looks at learners perception of the OLC. Research byGunawardena and Zittle (1997) found that high levels of social presence play a significantrole in improving instructional effectiveness and help construct a sense of online com-munity. In fact, how individuals perceive the OLC can directly affect his or her partici-pation within that OLC (Tu andMcIsaac 2002; Short et al. 1976). In an OLC designed foreducational purposes, activities are course assignments and are directly tied to coursegrades. We rely on Twitter to provide a quick and easy method to bring new informationinto the course and the discussion board to become a living artifact of interactions.

    Together, these three theories provide a well-rounded model that considers eachindividual learner within a larger course community and how different technologiescan influence and enhance learning and interaction.

    5 System design

    5.1 CMS versus OLC

    As mentioned, the CMS platform used at our university is Angel, which is owned andsupported by Blackboard Inc. Angel v7.4 contains a typical ODB. Illustrated inFig. 3, while functional, the discussion space lacks even basic social features suchas avatars and links to profiles. And the profile capabilities offered through Angel isalso rather limiting, with little space to share interests. Furthermore, the ODB lackseasy to use mechanisms for sharing other types of media, such as images or video.

    Even for hybrid-learning and blended learning courses, where students and in-structors meet face-to-face, having a more social online environment can be importantfor facilitating asynchronous information exchange and strengthening the classroom

    Educ Inf Technol

  • bonds. For these reasons, in addition to the fact that Angel is a closed-sourceenvironment, unable to be enhanced by our research team, we have chosen Elgg asthe OLC platform for this study.

    5.2 Elgg online learning community

    Today, there exist numerous sophisticated social platforms, which are available forinstructors to choose from. Elgg is one such environment and has been around since2006. Available through and, the social networking platformis bundled with blogging, file sharing, the ability to create unlimited sub-communitiesand peer-to-peer (P2P) networking capabilities. Additionally, Elgg provides for theability to restrict access to data across a number of levels, including individual-level,group-level, logged-in user-level and custom levels of restriction; a feature that makes itparticularly appealing to educators.

    Since 2006, our research has shown Elgg to function particularly well in highereducation and has helped compliment aspects where CMS systems have fallen short(Thoms et al. 2009, 2008). Illustrated in Fig. 4, the ability for any user to create a sub-community, known in Elgg as groups, allows both instructors to create as many groupsas they have classes. Additionally, students can create groups for class-based groupactivities. Consequently, each group has dedicated resources only available to assignedmembers of the sub-group.

    5.3 Accessing the twitter feed

    In our previous design iteration, we integrated Twitter into Elgg (Thoms 2012). Thisnew module allowed course instructors to retrieve relevant tweets from the Twitterfeed; specifically tweets that contained targeted keywords, such as BCS300, or

    Fig. 3 Angel discussion board component

    Educ Inf Technol

  • tweets targeted at or posted by a specific user, BCS300_Instructor. Illustrated in Fig. 5,students participating in course assignments would make tweets with the@BCS300_Instructor syntax. This syntax, @TwitterUser, is more commonly referredto as a mention. Consequently, in order for an individual to have tweets feed into theElgg system, they would need to mention @BCS300_Instructor for each of those posts.

    5.4 Discussion board integration

    Our previous design integrated Twitter with Elggs blogging engine (Thoms 2012). Whileresults were positive, a further review of our design showed that the integration was notoptimal. In this design cycle we integrated Twitter with Elggs ODB to allow individualsto continue a discussion thread initiated based on a students initial tweet. Since no suchmodule existed in Elgg, we constructed a new module to extract tweets that mentionBCS300_Instructor and display those tweet in chronological order on the OLC dashboard.

    Illustrated in Fig. 6 is a portion of the OLC dashboard and the view a studentwould see upon logging into the system. In the left column, students see a complete

    Fig. 4 Elgg v1.7.6 course communities

    Fig. 5 Twitter shout-at

    Educ Inf Technol

  • list of recent community tweets. The module presents users with the tweet, the OLCuser name and icon with a link to that users profile, Twitter account name, with a linkto that users external Twitter profile and an additional link, Rate It and Discuss!,which allows individuals to continue the conversation within the community ODB.

    Figure 7 provides a snapshot of the Elgg discussion board. In this particularscreenshot, a student has clicked on an existing discussion thread, which was initiatedwith a single tweet from a student. Additionally, keeping with our OLC design in priorversions of the system, we added modules to the OLC dashboard to present users with arandomized summary of popular tweets and discussion content from across the OLC,shown in the right hand side of Fig. 6. Detailed in the next section, students wererequired to participate in weekly Twitter discussions for course credit and the dashboardmodule helped facilitate these activities in a more seamless fashion.

    6 Research design

    To help support our OLC design, we targeted specific information systems courses atour university. Our experiment can be categorized as a two-group quasi-field exper-iment. Similar to the characteristics of a field experiment, we measure the effects ofour intervention on existing populations of college students where pre-existingbaselines exist for which to compare results. Our research questions, broken intoRQ1, RQ2 and RQ3 are as follows.

    RQ1. What impact will our new OLC design have on interaction (RQ1a), learning(RQ1b) and community (RQ1c) across online and face-to-face courses?

    Fig. 6 OLC dashboard

    Educ Inf Technol

  • RQ2. What impact will Twitter have on interaction (RQ2a), learning (RQ2b) andcommunity (RQ2c) across online and face-to-face courses?

    RQ3. What impact will a Twitter ODB have on interaction (RQ3a), learning(RQ3b) and community (RQ3c) across online and face-to-face courses?

    To explore these questions, Twitter was incorporated into the course syllabus andrequired for course credit. Assignments were divided into two parts and students wererequired to contribute weekly to Twitter and the ODB based on topics created by thecourse instructor. As one example, in Part 1 of Technology Assignment 11, studentswere required to tweet one article they found that raises ethical considerationssurrounding information systems. Students were instructed to provide a one linesummary and a link to the article. In Part 2, students were required to continue thediscussion across 23 discussion threads initiated by their peers.

    7 Results

    7.1 User population

    The OLC was implemented across two sections of BCS300, a capstone course ininformation systems required for all business and computer systems majors. The firstsection was a pure online course, while the other section met twice a week in a

    Fig. 7 OLC-twitter ODB

    Educ Inf Technol

  • traditional college setting. Pretest survey responses showed that 61 % of onlinestudents were aged between 17 and 22, with 35 % between 23 and 30 and 4 %between 30 and 49. Among our face-to-face population, 75 % were aged between 17and 22, with 25 % between 23 and 30. Regarding gender, 61 % of online studentswere male and 39 % were female, while 79 % of face-to-face students were male with21 % female.

    7.2 Site usage

    The software was utilized during the fall 2011 semester, from August 29 to December14. The site received 2,704 visitors averaging 25 visits a day from its initial popu-lation of 50 users. The online course concluded with 19 students receiving finalgrades and the face-to-face course concluded with 25 students. During this timeframe,and detailed in Fig. 8, 526 tweets were made generating 1,062 discussion posts and1,103 discussion ratings. 100 % of online users indicated using the OLC to tweet anddiscuss weekly. This number was 91 % for face-to-face users.

    7.3 Quantitative survey results

    Survey questionnaires were distributed to two sections of BCS300. A total of 23 onlinestudents (or a 96 % response rate) and 24 face-to-face students (96 %) participated in thepretest survey. For the posttest, the online course returned 19 surveys (100 %) and 23surveys from face-to-face students (92 %).

    7.3.1 Survey results: OLC

    Detailed fully in Table 1, responses related to overall perceptions of the OLC werefavorable. Both online and face-to-face students agreed that the OLC helped withinteraction (90 % and 92 % respectively). Regarding perceptions of learning, again, bothonline and face-to-face students agreed that the OLC helped with learning (74 % and69 % respectively) although a greater number of online students disagreed with thisstatement (16 % versus 4 %). Regarding community, both online and face-to-facestudents agreed that the OLC helped to increase it (90% and 87% respectively), althoughsome disagreement existed from online students (11 %). Overall, online and face-to-facestudents indicated a positive experience using the OLC (89 % and 96 % respectively).








    Tweets Discussion Ratings


    613 610


    449 493

    Online Face-to-Face

    Fig. 8 Content creation

    Educ Inf Technol

  • 7.3.2 Survey results: Twitter

    Detailed fully in Table 2, survey items asked specific questions regardingmicrobloggingand their use of Twitter throughout the semester. Overall, 76 % of online students and60 % of face-to-face students stated that Twitter increased interaction. When asked ifTwitter increased learning, 64 % of online students agreed, with 21 % expressingdisagreement compared to 70 % of n-to-face students agreeing and only 9 %disagreeing. When asked whether Twitter was an excellent tool for building coursecommunity, 84 % of online students agreed with 16% disagreeing, versus 74% of face-to-face students expressing agreement and only 4 % disagreeing. Overall, online

    Table 1 OLC: online vs. blended learning

    An OLC SA A N D SD


    (Pre) Will increase interaction. 39 % 29 % 35 % 46 % 22 % 21 % 4 % 4 %

    (Post) Increased interaction. 53 % 22 % 37 % 70 % 9 % 5 % 5 %

    (Pre) Will Increase learningfor this course.

    26 % 17 % 35 % 50 % 35 % 25 % 4 % 4 % 4 %

    (Post) Increased learningfor this course.

    37 % 17 % 37 % 52 % 11 % 26 % 16 % 4 %

    (Pre) Will help buildcourse community.

    17 % 25 % 57 % 42 % 22 % 29 % 4 % 4 %

    (Post) Helped buildcourse community.

    53 % 26 % 37 % 61 % 13 % 11 %

    SA strongly agree, A agree, N neither agree nor disagree, D disagree, D strongly disagree, ON online, F2Fface-to-face

    Table 2 Twitter: online vs. blended learning

    Twitter SA A N D SD


    (Pre) Will increase interaction. 4 % 21 % 61 % 46 % 22 % 33 % 9 % 4 %

    (Post) Increased interaction. 42 % 17 % 32 % 43 % 11 % 35 % 5 % 4 % 11 %

    (Pre) Will increase learningfor this course.

    4 % 8 % 22 % 42 % 48 % 42 % 17 % 8 % 9 %

    (Post) Increased learningfor this course.

    32 % 13 % 32 % 57 % 16 % 22 % 5 % 9 % 16 %

    (Pre) Will help buildcourse community.

    13 % 8 % 39 % 63 % 35 % 21 % 9 % 8 % 4 %

    (Post) Helped buildcourse community.

    42 % 13 % 42 % 61 % 22 % 5 % 4 % 11 %

    SA strongly agree, A agree, N neither agree nor disagree, D disagree, D strongly disagree, ON online, F2Fface-to-face

    Educ Inf Technol

  • students and face-to-face students indicated a positive experience using Twitter (84 %and 100 % respectively).

    7.3.3 Survey results: Discussion board

    Since the Twitter module and discussion board were largely separate components, wealso measured students perceptions of the Twitter ODB across our population. Fullydetailed in Table 3, 84 % of online students agreed and 87 % of face-to-face studentsagreed that the ODB increased interaction. When asked if the ODB increasedlearning, 69 % of online students agreed with 16 % disagreeing versus 91 % offace-to-face students agreeing and none disagreeing. Regarding course community,74 % of online respondents agreed with 10 % disagreeing, versus 78 % of face-to-facestudents agreeing and only 4 % disagreeing. Overall, online and face-to-face studentsindicated a positive experience using the ODB (90 % and 91 % respectively).

    7.4 Qualitative data

    We also collected qualitative feedback from students. As expected, open-endedresponses mirrored quantitative responses and were mixed. A student from the onlinesection stated, The use of Twitter allowed me to enjoy the course [while] workingwith my classmates at the same time. Another student stated, I have experiencedmore interaction with classmates in this 1 online course then I had in my previous 5online courses combined. However, not all students from the online course expressedsuch positive remarks. One student stated, I didnt fully understand why we had to useTwitter to make our initial posts. Couldnt we just create a new discussion topic? The[number of] characters is limiting. Many other students made remarks about thenavigation and OLC look and feel.

    From the face-to-face class, one student stated, SocialXYZ was a good alternativeto [traditional] assignments. Another stated, [SocialXYZ] was fun and interesting

    Table 3 Twitter discussion board: online vs. blended learning

    A twitter discussion board SA A N D SD


    (Pre) Will increase interaction. 22 % 21 % 57 % 46 % 17 % 33 % 4 %

    (Post) Increased interaction. 47 % 35 % 37 % 52 % 11 % 13 % 5 %

    (Pre) Will increase learningfor this course.

    17 % 13 % 48 % 50 % 30 % 33 % 4 % 4 %

    (Post) Increased learningfor this course.

    37 % 30 % 32 % 61 % 16 % 9 % 11 % 5 %

    (Pre) Will help buildcourse community.

    22 % 13 % 52 % 54 % 17 % 29 % 9 % 4 %

    (Post) Helped buildcourse community.

    37 % 35 % 37 % 43 % 16 % 17 % 5 % 4 % 5 %

    SA strongly agree, A agree, N neither agree nor disagree, D disagree, D strongly disagree, ON online, F2Fface-to-face

    Educ Inf Technol

  • and made it seem like it wasnt homework.Most of the critical comments regardingthe tool reflected site navigation and speed.

    8 Discussion and implications

    Since 2006 we have implemented social technologies within higher educationalsettings. In spring 2011 we outfitted a traditional blog with the Twitter microblogand discovered that the pairing was less than optimal. Consequently, for fall 2011 weadapted our design to integrate Twitter with an ODB and discovered stronger levelsof agreement across course interaction, learning and community.

    8.1 OLC: Online vs. blended learning

    Our first set of research questions, RQ1a, RQ1b and RQ1c explored the overallimpact of our OLC design on interaction, learning and community across pure onlineand blended learning courses. We discovered that although students who met face-to-face had higher levels of overall agreement on the OLCs impact on interaction (90 %versus 92 %), online students more strongly agreed with this statement (53 % versus22 %). This was also the case for community, which experienced stronger overalllevels of agreement (53 % versus 26 %). These numbers make sense when weconsider that for online students, the OLC comprised the entire course community,versus students who met face-to-face also benefited from in-class interaction, learningand community. Another possibility for the stronger levels of agreement considers theamount of activity online students generated versus their blended learning counter-parts. While both sections were required to participate across the same number ofassignments, the online course generated 25 % more content. Additionally, onlinestudents stated that they logged into the OLC more frequently, with 32 %indicating having logged in daily, versus only 4 % of blended learning students loggingin daily.

    8.2 Twitter: Online vs. blended learning

    Our second set of research questions, RQ2a, RQ2b and RQ2c, explored the overallimpact of Twitter in pure online courses versus a blended learningmodel.We discoveredthat a greater majority of online students agreed that Twitter increased interaction (74 %versus 60 %) and community (84 % versus 74 %). When we consider the fact that theaverage number of tweets per student by online students was 13, versus 11 from blendedlearning students, it is easy to attribute, to some degree, its success to the quantity oftweets generated.

    However, while interaction and community are important within an OLC, a keyaspect of an OLC, one that separates it from more popular social networking sites, isits emphasis on learning. Therefore, any new instructional tool should, in part,facilitate learning. To date, our results have indicated that Twitter provides a powerfultool for information exploration and information sharing, yet it falls short in increas-ing levels of perceived learning (43 % agreement and 35 % disagreement). As aresult, we labeled Twitter as a successful broadcast technology that does not directly

    Educ Inf Technol

  • contribute to learning (Thoms 2012). Thus, we returned to the drawing board andexplored new designs that might increase Twitters direct influence on learning.

    In our new design, the majority of online students (64 %) and blended learningstudents (70 %) agreed that Twitter increased learning, with online students morestrongly agreeing (32 % versus 13 %), which helped validate some for our systemenhancements. However, we still feel there is greater room for improvement, particu-larly for pure online courses, where we received the greatest amount of disagreement(21 %) compared to blended learning students (9 %). This remains an area we aim toexplore in future design iterations.

    8.3 Twitter discussion board: Online vs. blended learning

    Our last set of research questions, RQ3a, RQ3b and RQ3c, explored the impact of aTwitter discussion board across online and blended learning courses. We discoveredthat online students, while experiencing slightly lower levels of overall agreementthat the ODB increased interaction (84 % versus 87 %) and community (74 % versus78 %), more strongly agreed across these constructs. Once again, we attribute this tothe number of interactions that occurred across the Twitter ODB, versus their blendedlearning counterparts. The number of interactions that took place among face-to-facestudents compared to online students was considerably lower, with 37 % less activityper student occurring amongst face-to-face students. As discussed during the reviewof recent literature, a steady flow of activity remains a critical success factor for mostonline social networking software, which may not have been present in the blendedlearning environment.

    However, this trend did not hold true for learning. What we discovered was that anoverwhelming majority of blended learning students (91 %) experienced higherlevels of learning compared with only 69 % agreement from online students. Foronline students, the ODB is the heart of course dialogue, so it was somewhatdisconcerting to see such lower levels of agreement compared to face-to-face studentsin this respect. Additionally, online-students generated 44 % more discussion postsper student compared to face-to-face students. A further review of these discussionthreads might be more telling, but an initial review indicates a higher-level ofdiscourse occurring across online discussion threads versus those in the blendedlearning discussions. Therefore, it remains difficult to identify exactly what attributedto this higher level of learning, as experienced by face-to-face students. One possibleexplanation considers the combination of asynchronous discussion along with face-to-face instruction. Discussion posts often came up during course seminars, whichmet twice a week. And many times, the concepts and ideas expressed within the ODBwere validated by the course instructor, an aspect not afforded across the onlinecourse. And, surprisingly, the total number of discussion posts appears to have had noinfluence on the perceptions of learning.

    9 Next steps and future research

    The inclusion of Twitter into the course OLC seemed like a natural extension toenhance course interaction, learning and community. Furthermore, the extension of

    Educ Inf Technol

  • the Elgg ODB was also a positive addition. However, the Elgg ODB has limitations,primarily that it is not threaded. There are also limited navigation capabilities whenbrowsing discussions, as identified by numerous students. In future design iterationswe hope to introduce a more threaded approach to the Elgg discussion board, withsome possible anchoring capabilities.

    Looking into future research opportunities, we see the integration of Twitter acrossMassively Open Online Courses (MOOCS). While MOOCS offerings are relativelynew at our university, there has been a recent push for faculty to expand MOOCSofferings. Twitter is inherently massively-open and integrating it within the MOOCSenvironment could help foster higher levels of learning and interaction outside aMOOCS community.

    10 Conclusion

    In this research we perform a major overhaul on a recent design component for Elgg,an online learning community (OLC). In a prior study, we integrated Twitter with atraditional blog to discover that the impact Twitter had on interaction, learning andcommunity was limited. In our updated design, we integrated Twitter with an onlinediscussion board (ODB) and measured its impact across upper-division undergradu-ate online learning and blended learning courses. Our results showed strong levels ofagreement that the OLC, Twitter and Twitter ODB enhanced levels of interaction,learning and community across both online and blended learning environments.


    Balaji, M. S., & Chakrabarti, D. (2010). Student interactions in online discussion forum: Empirical researchfrom Media Richness Theory perspective. Journal of Interactive Online Learning, 9(1), 122.

    Betta,C. (2007). Social networking and academic life. Research assignment. Literature Report. DelftUniversity of Technology.

    Bdker, S. (1989). A human activity approach to user interfaces. Human-Computer Interaction, 4, 171195.Brescia, W., & Miller, M. (2006). Whats it worth? The perceived benefits of instructional blogging.

    Electronic Journal for the Integration of Technology in Education, 5(1), 4452.Chen, G., & Chiu, M. (2008). Online discussion processes: Effects of earlier messages evaluations,

    knowledge content, social cues and personal information on later messages. Computers & Education,50(3), 678692.

    Chen, J., & Pirolli, P. (2012). Why you are more engaged: factors influencing twitter engagement inoccupy wall street, Proceedings from the Sixth International AAAI Conference on Weblogs and SocialMedia, Jun. 57, 2012.

    De Wever, B., Schellens, T., Valcke, H., & Van Keer, H. (2006). Content analysis schemes to analyzetranscripts of online asynchronous discussion groups: A review. Computers & Education, 46(1), 628.

    Dunlap, J. C., & Lowenthal, P. R. (2009). Tweeting the night away: Using twitter to enhance socialpresence. Journal of Information Systems Education, 20(2).

    Ebner, M., Lienhardt, C., Rohs, M., & Meyer, I. (2010). Microblogs in higher education a chance tofacilitate informal and process-oriented learning? Computers & Education, 55(2010), 92100.

    Educational Marketer. (2003). Colleges increase use of course management systems, says MDR. Educa-tional Marketer, 34(8), 45.

    Grosseck, G., & Holotesco, C. (2008). Models good pedagogy responsive to students learning needs,The 4th International Scientific Conference eLSE: eLearning and Software for Education, Bucharest,Apr. 1718, 2008.

    Educ Inf Technol

  • Gunawardena, C. N., & Zittle, F. J. (1997). Social presence as a predictor of satisfaction within a computer-mediated conferencing environment. The American Journal of Distance Education, 11, 826.

    Heckman, R., & Annabi, H. (2006). Cultivating voluntary online learning communities in blendedenvironments. Journal of Asynchronous Learning Networks, 10(4).

    Hull, D., & Saxon, T. (2009). Negotiation of meaning and co-construction of knowledge: An experimentalanalysis of asynchronous online instruction. Computers & Education, 52(3), 624639.

    Jenks, J. (2011). The digital world of millennials. eMarketer.Johnson, C. (2001). A survey of current research on online communities of practice. Internet and Higher

    Education, 4(1), 4560.Kirkup, G. (2010). Academic blogging: academic practice and academic identity. London Review of

    Education, 8(1).La Pointe, K. D., & Gunawardena, C. (2004). Developing testing and refining of a model to understand the

    relationship between peer interaction and learning outcomes in computer-mediated conferencing.Distance Education, 25(1), 93106.

    Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge, England:Cambridge University Press.

    Liaw, S., & Huang, H. (2000). Enhancing interactivity in web-based instruction: A review of the literature.Educational Technology, 40(3), 4145.

    Mor, Y., Tholander, J., & Holmberg, J. (2005). Designing for constructionist web-based knowledgebuilding. Conference on computer support for collaborative learning: Learning 2005: the next 10years! (pp. 450459). Taipei: International Society of the Learning Sciences.

    Oblinger, D., & Oblinger, D. (2005). Is it age or IT: first steps toward understanding the net generation.In D. Oblinger, & J. Oblinger (Eds.), Educating the net generation, p. Online. Retrieved fromEducause.

    Official Twitter Blog. (2011). Retrieved online 6/11/2012 from, R., & Pratt, K. (1999). Building learning communities in cyberspace: Effective strategies for the

    online classroom. San Francisco: Jossey-Bass.Quan-Haase, A. (2005). Trends in online learning communities. SIGGROUP Bulletin, 25(1), 26.Redfern, S., & Naughton, N. (2002). Collaborative virtual environments to support communication and

    community in internet-based distance education. Journal of Information Technology Education, 1.Rhode, M., Reinecke, L., Pape, B., & Janneck, M. (2004). Community-building with web-based systems

    investigating a hybrid community of students. Computer Supported Cooperative Work (CSCW), 13(56),471499.

    Savery, J., & Duffy, T. (1996). Problem based learning: An instructional model and its constructivistframework. In B. Wilson (Ed.), Constructivist learning environments: Case studies in instructionaldesign. Englewood Cliffs: Educational Technology.

    Short, J. A., Williams, E., & Christie, B. (1976). The social psychology of telecommunications. New York:John Wiley & Sons.

    Solimeno, A., Mebane, M. E., Tomai, M., & Francescato, D. (2008). The influence of students and teacherscharacteristics on the efficacy of face-to-face and computer supported collaborative learning.Computers & Education, 51(1), 109128.

    Stacey, E. (2002). Social presence online: Networking learners at a distance. In Education and informationtechnologies. Kluwer Acad. Publishers, vol. 7, (287294).

    Thoms, B. (2011). A Dynamic Social Feedback System to Support Learning and Social Interaction inHigher Education. IEEE Transactions on Learning Technologies, 4(4), 340352.

    Thoms, B. (2012). Perceptions and Outcomes of Microblogging in Higher Education. Journal of Informa-tion Technology Education, 11(1), 179197.

    Thoms, B., Garrett, N., & Ryan, T. (2009). Online Learning Communities in the New U. InternationalJournal of Networking and Virtual Organisations, 6(5), 499517.

    Thoms, B., Garrett, N., Soffer, M., & Ryan, T. (2008). Resurrecting Graduate Conversation through anOnline Learning Community. International Journal of Information and Communication TechnologyEducation, 4(3), 341350.

    Tu, C. H., & McIsaac, M. (2002). The relationship of social presence and interaction in online classes. TheAmerican Journal of Distance Education, 16, 131150.

    U.S. Census Bureau. (2009). School enrollmentsocial and economic characteristics of students: Oct.2009. Retrieved 7/3/11, from

    Vygotsky, L. S. (1987). Mind in society: The development of higher psychological processes. Cambridge:Harvard University Press.

    Educ Inf Technol

  • Wakefield, J. S., Warren, S. J., & Alsobrook, M. (2011). Learning and teaching as communicative actions:A mixed-methods twitter study. Knowledge Management & E-Learning: An International Journal,3(4), 563584.

    Williamson, D. A. (2007). Social Network Marketing: Ad Spending and Usage, Social Network Marketing,Report by Debra Aho Williamson, Dec. 2007. Retrieved 6/2/08 from

    Zhao D. and Rosson, MB (2009). How and why people Twitter: The role that microblogging plays ininformal communication at work, Proceedings of the ACM 2009 International Conference onSupporting Group Work. New York: ACM, pp. 243252.

    Educ Inf Technol

    Introducing a twitter discussion board to support learning in online and blended learning environmentsAbstractIntroductionBackgroundSocial software across educationOnline discussion boards

    MicrobloggingTwitterTwitter used in education

    Theoretical modelSystem designCMS versus OLCElgg online learning communityAccessing the twitter feedDiscussion board integration

    Research designResultsUser populationSite usageQuantitative survey resultsSurvey results: OLCSurvey results: TwitterSurvey results: Discussion board

    Qualitative data

    Discussion and implicationsOLC: Online vs. blended learningTwitter: Online vs. blended learningTwitter discussion board: Online vs. blended learning

    Next steps and future researchConclusionReferences