Emerging Practice and Research in Blended Learning. Charles R. Graham

Transcription

1 Running Head: RESEARCH IN BLENDED LEARNING The following is a draft version from a chapter soon to be printed in the Handbook of Distance Education. Graham, C. R. (2013). Emerging practice and research in blended learning. In M. G. Moore (Ed.), Handbook of distance education (3rd ed., pp ). New York, NY: Routledge. Emerging Practice and Research in Blended Learning Charles R. Graham Department of Instructional Psychology & Technology Brigham Young University (801) charles.graham@byu.edu

2 2 Abstract This chapter introduces the concept of blended learning as the combining of online and face-toface instruction. Background and analysis related to variations in institutional definitions of blended learning as well as rationales for adopting blended learning are presented. Prominent models of blended learning in higher education, k-12, and corporate training contexts are highlighted. A synthesis of current research related to blended learning is organized around the themes of theory, learning effectiveness, learner satisfaction, faculty satisfaction, access and flexibility, and cost effectiveness. The chapter ends with recommendations for future research related to blended learning environments. Keywords: blended learning, hybrid course, mixed mode, online and face-to-face modalities, learning environments, learning effectiveness, access, flexibility, learner satisfaction, cost effectiveness

3 3 1. Background and Definition of Blended Learning Discussion of blending learning (BL) is appearing with increased frequency in both the scholarly literature and the popular press. A 2011 literature search by the author found close to 200 dissertations and hundreds of journal articles on the topic. Additionally, the American Society for Training and Development identified BL as a top trend in the knowledge delivery industry (Rooney, 2003). In higher education the use of BL has grown rapidly, with predictions that it will become the new traditional model (Ross & Gage, 2006, p. 167) or the new normal in course delivery (Norberg, Dziuban, & Moskal, 2011, p. 207). Historically, blended learning was predominantly found in corporate and higher education contexts, but its use is now increasingly found in in K-12 education (Staker, Chan, Clayton et al., 2011; Picciano, Seaman et al., 2011). A 2008 report sponsored by the North American Council for Online Learning (NACOL) projected, Blended learning is likely to emerge as the predominant model of the future (Watson, 2008, p. 3). Despite current popularity of the term blended learning, it is defined with considerable variation across institutional contexts. While researchers have expressed frustration over the unclear definitional contours of this new BL ecosystem (see Oliver & Trigwell, 2005; Teng, Bonk, & Kim, 2009; Picciano, 2009), the fact is that the landscape of BL is still evolving rapidly. Much of the current research in BL has focused on attempting to describe and chart its boundaries. Operational definitions and taxonomies developed for BL environments currently lack the widespread consensus or maturity that researchers desire. This section will highlight four central issues related to definitions of BL. 1. What is being blended? 2. Should reduced seat time be part of the definition?

4 4 3. Should the quantity of online instruction be part of the definition? 4. Should quality factors be part of the definition? 1.1 What is being blended? The primary issue in defining BL is the question of what is being blended? Several analyses of existing definitions have been published. Graham (2006) identified the literature s three most common answers to the question: (1) blending online and face-to-face instruction, (2) blending instructional modalities (or delivery media), and (3) blending instructional methods. Oliver & Triggwell (2005) identified and critiqued six different mixes that could be involved in a BL approach. In order to avoid another insufficient definition of BL, Sharpe et al. (2006, p. 18) identified eight dimensions on which blending might occur. The most common use of the term BL denotes a combination of traditional face-to-face and online instruction. Other proposed definitions do not distinguish BL course experiences from either distance learning or traditional face-to-face instruction. For example, mixing content delivery media or pedagogies is common across both distance and face-to-face learning. Using either of these blends to define BL would conflate the definitions, essentially making everything BL. 1.2 Is Less Seat Time Required? A second issue focuses on how institutions operationalize the distinction between traditional face-to-face courses and BL courses. Operational definitions are important in clarifying the options they offer their students. At the core of this issue is the point at which a face-to-face course becomes a BL course? Do significant online components constitute BL if students meet for the same amount of time in class?

5 5 As traditional learning environments increase use of information and communication technologies (ICTs), critical distinctions between face-to-face and distance learning are difficult to base solely on technology use. Some researchers express concern that BL should go beyond merely bolting technology onto a traditional course, using technology as an add-on to teach a difficult concept or adding supplemental information (Vaughan, 2007, p. 82). Thus some definitions of BL include a reduction in face-to-face contact or seat time (Vaughan, 2007; Picciano, 2009; Mayadas & Picciano, 2007). For example, Picciano s (2009) definition requires that a portion (institutionally defined) of face-to-face time [be] replaced by online activity (p. 10). 1.3 How Much Online Learning is Required? A third issue commonly raised in defining BL is how much online learning is required to define the experience as BL? Does a face-to-face course with one brief online experience qualify? Conversely, should an online course with a face-to-face orientation be considered BL? Several authors have acknowledged this issue by defining the boundary between BL and other modalities as a proportion of content delivered online. For example, Allen & Seaman (2007) categorized traditional as having 0% of content delivered online, web facilitated as 1-29% online, blended as 30-79% online, and online as 80% or more. Similarly, Watson et al. (2010) set a threshold of 30% online delivery of content for an environment to be considered blended. A challenge with percentage thresholds is the difficulty in measuring something that is not easily or accurately quantifiable. Additionally, even if a percentage could be accurately determined, what practical difference would exist between courses with 29% versus 30% of content delivered online?

6 6 1.4 Should quality be included in the definition? A fourth issue prominent in definitions of BL is whether or not quality should be included in the definition. The inclusion of quality in the definition is primarily supported by those who desire to use BL as a tool for transformational change. For example, Garrison and Kanuka (2004) defined BL as the thoughtful integration of classroom face-to-face learning experiences with online learning experiences (p. 96, emphasis added). Similarly, in 2005 participants in the Sloan-C Blended workshop included in their definition courses that integrate online with traditional face-to-face class activities in a planned, pedagogically valuable manner (Picciano, 2005, p. 97, emphasis added). These subjective qualifiers to BL definitions are well intentioned but more aspirational than practical. They stem from a fear that the transformational potential of BL course redesign could be compromised by those not concerned with improving pedagogical quality (Graham & Robison, 2007; Vaughan, 2007) a fear that is not unfounded. Salomon (2002) warned of education systems tendency to preserve themselves and their practices: A most powerful and innovative technology is taken and is domesticated, or if you want trivialized, such that it does more or less what its predecessors have done, only it does it a bit faster and a bit nicer. (p. 72). 1.5 Can a Vague Definition Actually be Useful? It may be helpful to think of the term blended learning as a boundary object (Norberg, Dziuban, & Moskal, 2011): an element shared across communities of practice, plastic enough to adapt to local needs and constraints of the several parties employing them, yet robust enough to maintain a common identity across sites....weakly structured in common use... strongly structured in individual site-use (Star & Griesemer, 1989, p. 393).

7 7 Agreement seems widespread that BL involves a combination of face-to-face and online learning. Beyond that, individual researchers and institutions differ in how they further limit definition boundaries. Some require a reduction in face-to-face seat time, while others may specify levels of online or face-to-face instruction for BL. In a review of the BL literature and implementation practices in the United Kingdom, researchers recognized benefits to a broadly structured definition of BL: We noted from the interviews that some institutions have developed their own language, definitions or typologies to describe their blended practices. We suggest that this poor definition [of blended learning] may be a strength and part of the reason why the term is being accepted. The lack of definition allows institutions to adapt and use the term as they see fit, and to develop ownership of it. (Sharpe, et al., 2006, p. 17) In this chapter, BL will simply be defined as learning experiences that combine face-toface and online instruction. 2. Models of Blended Learning in Practice Current emphasis has focused on distinguishing BL from traditional and online coursework. However, future learning systems may be differentiated less on whether they blend than on how they blend (Ross & Gage, 2006). A time may come when the term blended learning becomes obsolete because blending has become the new normal (Cross, 2006; Graham, 2006; Norberg, Dziuban, & Moskal, 2011). However, currently a label is helpful to refer to these structurally different environments. The contemporary BL ecosystem is rich with innovation and creativity. The diversity of blends reflect the range of possibilities for transforming students learning experience. With

8 8 time, taxonomies and widely accepted models will provide additional stability for researchers and practitioners. This section of the chapter highlights several prominent examples of emerging BL models. Current models tend to focus on physical dimensions of the learning environments and very general high-level pedagogical approaches. 2.1 K-12 Education Models BL has been steadily increasing in K-12 contexts (Picciano et al., 2011), partially as an alternative to purely online models and in response to community expectations and funding models that require schools to physically supervise children during the day (Wicks, 2010). Only recently have efforts been made to systematically characterize the K-12 blends as different from prior traditional and distance learning options. Watson (2008) analyzed ten charter/virtual schools along a spectrum containing seven levels from fully online to traditional face-to-face, varying in the amount of face-to-face versus online instruction as well as optional or required components of both systems. Two years later, Watson participated in expanding the model to include seven defining dimensions: 1. Level of online instruction (unit/lesson, single course, entire curriculum) 2. Time (fixed daily schedule, modified schedule, open entry/open exit) 3. Role of online components (enhance traditional instruction, transform traditional instruction) 4. Teacher role (leads instruction, supports instruction, not involved) 5. Student role (teacher-driven learning, teacher-guided learning, independent learning) 6. Student support (little or none, school-based mentoring, school and home mentoring)

9 9 7. Student to teacher ratio (traditional classroom ratio, 2-3 times traditional classroom ratio, instructional helpdesk model) In 2011, the Innosite Institute released a research report documenting 40 case studies of BL in K-12 contexts across the United States (Staker et al., 2011). The report classified BL models along two dimensions in a 2D graph: the horizontal dimension representing the physical location of the student (brick and mortar to remote) and the vertical dimension representing the location of the course content (offline to online). Within this framework six models were identified (see Table 1), and each of the 40 programs was classified as one of the models. Most recently the six models were further collapsed into four models (Staker & Horn, 2012) Insert Table 1 here Higher Education Models Many institutions of higher education have developed an institution-specific BL categorization. Typically these models vary along the dimension of how much of a course is taught face-to-face versus mediated by technology (see Table 2 for three examples). Picciano (2009) created a two-dimensional representation of blended models with a face-to-face versus online dimension on the horizontal axis and a minimal to infused technology dimension on the vertical axis Insert Table 2 here

10 10 Some of the earliest models of higher education BL resulted from a study of 30 course redesign projects funded by the National Center for Academic Transformation and the PEW Charitable Trusts (Twigg, 2003). The three-year grant program involving universities from across the United States was designed to enhance quality and save funding through redesigning large enrollment courses. Of the 30 course redesigns, 26 used BL approaches to achieve cost savings and quality gains, while 4 adopted a purely online approach (Graham & Allen, 2009). Table 3 outlines the 5 different models that emerged from the redesign Insert Table 3 here Corporate Models Rossett & Frazee (2006) outlined three general models (see Table 4) describing common blends in the corporate environment. The anchor and bookend blend represents a mixture of three common environments: instructor-led classroom instruction, independent online learning, and instructor-guided online learning. The bookend and field blends also highlight a trend supported by BL that integrates learning directly into workplace activities Insert Table 4 here Many corporate training models emphasize integration of informal and formal learning environments, embodied in a move from instructor-led classroom learning to real-time workplace learning supported by online learning tools. Collis (2006) outlined a multi-national company s model that centers each course around a business need or competence gap and

11 11 focuses on completing a complex work-based activity with face-to-face support from supervisors and online connection to remote experts and peers. Lewis & Orton (2006) described a threephased BL model at IBM: 1. Phase 1: 26 weeks of self-paced online learning in the workplace environment 2. Phase 2: 5 days of face-to-face in-class instruction in the company learning lab 3. Phase 3: 25 weeks of online learning focused on applying skills in the workplace 3. Blended Learning Research Research related to BL is relatively undeveloped compared to research in distance and traditional learning environments. In fact, much of the early work in BL has occurred as an outgrowth of distance learning research. As developed in Section 1, the issues and foci of BL research and ways they compliment and are distinct from traditional distance learning research are still being defined. This section of the chapter first explores theoretical frameworks that researchers are using to guide research in BL contexts. Then the Sloan Consortium s five pillars (learning effectiveness, learner satisfaction, faculty satisfaction, access, and cost effectiveness) are used as an organizing framework for reviewing the current BL research. 3.1 Theory Theory provides a common language and focus for creating and discussing knowledge in scholarly communities (Dubin, 1978). Behavioral and social science groups have explored extensively what constitutes theory and what role theory plays in the knowledge creation process. Burkhardt & Schoenfeld (2003) noted that a lack of attention to coherent theory building leaves us looking balkanized and incoherent, the whole being less than the sum of its parts (p. 13).

12 12 As BL research increases, theoretical frameworks should be developed to address the issues unique to BL environments. Shea (2007) suggested several existing conceptual frameworks that could be used, including the how people learn (HPL) framework that focuses on the development of learner-centered, knowledge-centered, assessment-centered, and communitycentered learning environments (Bransford, Brown, & Cocking, 1999). Bunderson (2003) recommended (among other possibilities) the theory of engaged collaborative discourse (Xin, 2002) as a lens for analyzing BL environments. A few researchers have attempted to connect or extend distance education theories to address BL issues. The most comprehensive of these is the work of Garrison and Vaughan (2008) using the community of inquiry framework with constructs of social, teaching, and cognitive presence to situate and analyze the design of university-level BL (Akyol, Garrison, & Ozden, 2009; Borup, Graham, & Velasquez, 2011; Borup, West, & Graham, 2012). Other researchers have used the theory of transactional distance with concepts of autonomy, dialog, and structure to explain findings in BL contexts (Dron, Seidel, & Litten, 2004; Wheeler, 2007; Lim, Fadzil, & Mansor, 2011). Other prominent theories applied to the design of BL environments include Keller s (1983) theory of motivation, Giddens (1984) structuration theory and Laurillard s (1993) conversation theory (see Keller, 2008; Stubbs, Martin, & Endlar, 2006; Heinze, Procter, & Scott, 2007). Structuration theory, which explains the relationship between social structure and individual agency, has been used heavily in information systems research. Conversation theory outlines a multi-stage process describing the communication between instructor and student. Finally, several researchers interested in the adoption of BL as an educational innovation have used Rogers (1983) diffusion of innovations theory to frame their analyses (Intharaksa, 2009;

13 13 Grgurovic, 2010; Fetters & Duby, 2011), explaining both the stages of adoption and the factors that affect the rate of adoption of innovations. While some of the research in BL is solidly grounded in theory, most of the existing research has sought to describe or solve localized challenges without contributing to coherent development of theory. Many studies consider theory only as background information or as a lens to describe findings or outcomes; few attempt to contribute substantively to the conversation about theory. Just as distance learning required theory to focus researchers on psychological rather than physical distance, BL needs theories to focus researchers on the substantive psychosocial issues that make it distinct. 3.2 Learning Effectiveness The first important question addressed by the BL research regards its effectiveness as an environment for helping students learn. A challenge inherent in this question deals with which characteristics of learning environments have potential to directly impact learning. Media studies resulting in no significant difference have taught researchers that while the physical characteristics (affordances) of the environment enable and constrain particular pedagogical methods, the active ingredient in learning is the pedagogy rather than the medium (Clark, 1983). Although the physical characteristics of the learning environments (e.g., online or face-to-face) are not causal factors, they may represent classes of pedagogies distinct enough to enable differences to be measured in meta-analyses where researchers have not yet identified the actual causal factors. At least five recent meta-analyses have looked specifically at BL as a moderating variable (Paul, 2001; Zhao et al., 2005; Sitzmann et al., 2006; Bernard et al., 2009; Means et al., 2009). A dissertation by Paul (2001) compared the effectiveness of Web-based training (WBT)

14 14 environments (80 studies) with blended environments combining face-to-face and WBT (15 studies). The study reported a minimal difference in favor of blended over purely online courses, with the mean effect size of the blended condition at and the WBT condition at In 2005 Zhao et al. conducted a meta-analysis considering a dozen moderating factors, which found that of all the factors, instructor involvement (76 studies) had the most significant impact. They concluded that largely because of instructor involvement studies that used a combination of technology and face-to-face education resulted in the most positive outcomes (p. 1863). A meta-analysis in 2006 by Sitzmann et al. compared the effectiveness of web-based instruction (WBI) and BL with classroom instruction (CI). Results were analyzed separately for outcomes in declarative (104 studies) and procedural knowledge (18 studies); BL was determined to be more effective than CI, with effect sizes of for declarative knowledge and for procedural knowledge outcomes. The effects of BL on the outcomes were much larger than the effect sizes for pure WBI compared with CI, which were for declarative knowledge and for procedural knowledge. Additionally, Bernard et al. (2009) conducted a meta-analysis of distance education (DE) courses that compared synchronous DE (5 studies), asynchronous DE (37 studies), and mixed DE (i.e., blended; 7 studies). This study found no significant difference among the three DE modalities, but concluded that the low number of studies available in the non-asynchronous DE categories suggested a need for further research attention. Most recently, in 2009 the U.S. Department of Education sponsored a meta-analysis looking at contrasts between online and traditional face-to-face learning (Means et al., 2009). The analysis used 50 different contrasts from 45 published studies comparing online and face-toface instruction. In 21 of the 50 cases, online learners had opportunities for face-to-face contact with an instructor and were therefore considered BL. The primary findings of the study claimed

15 15 that classes with online learning (whether taught completely online or blended) on average produced stronger student learning outcomes than did classes with solely face-to-face instruction. The mean effect size for all 50 contrasts was +0.20, p <.001. (p. 18). When the data were disaggregated to compare the effects from the purely online contrasts (29 cases) with the BL contrasts (21 cases), researchers found BL to be superior, with a mean effect size of compared to a mean effect size of for purely online. These meta-analyses provided evidence of differences in outcomes of online, blended, and face-to-face learning that are worth studying. However, the causal factors that lead to the improved outcomes are not understood. The analysis of Zhao et al. suggested that instructor involvement may be a critical factor. The Means et al. outcomes similarly attributed the largest pedagogical effect to instructor-directed learning environments. However, these studies did not identify aspects of instructor involvement that are important: e.g., Is something qualitatively different about the kinds of instructor interactions in a face-to-face or synchronous high-fidelity environment versus interactions in an asynchronous text-based environment? Or is instructor involvement more critical for certain kinds of learning (declarative versus procedural)? The Means et al. study reported opportunity for face-to-face time with the instructor during instruction to be one of the significant moderating variables for online learning. Results from the attempt of the Bernard study to differentiate between learning declarative knowledge and procedural knowledge seemed to imply that human interaction does not have a significant moderating effect on acquiring declarative knowledge. Others have argued that improved outcomes may have more to do with increased learner time on task in the BL environment (Sitzmann et al., 2006; Means et al., 2009).

16 16 While the meta-analyses give a broad view of the impact of BL in experimental studies, many non-experimental studies have also looked at learning effectiveness in BL contexts. The Research Initiative for Teaching Effectiveness at the University of Central Florida (UCF), an early institutional adopter of BL, conducted a multi-year study involving tens of thousands of students examining success rates (defined as C- grade or above) of their online, blended, and face-to-face course offerings. The analysis accounted for college, gender, and modality. They found that while college was the best predictor of success rates, within each college the success rates for BL were higher than both face-to-face and entirely online courses for both genders (Dziuban et al., 2004). Another study looking at success rates of BL offerings at the Rochester Institute of Technology (with a high number of students with hearing impairment) reported 95% of students completing courses with a C- grade or above (Starenko, Vignare, & Humbert, 2007). Many smaller studies have also found positive learning outcomes for BL courses (Reasons et al., 2005; Boyle et al., 2003; Cottrell & Robinson, 2003; Dowling et al., 2003; O Toole & Absalom, 2003; Riffell & Sibley, 2004). Larson and Sung (2009) went beyond academic performance and found that 52% of BL students had an increased interest in the subject matter, more than either the online or face-to-face students. As mentioned at the beginning of this section, more research is needed to identify the quantity and quality factors of the blended designs that impact achievement and success rates. Shea and Bidjerano (2011) recently completed a study using the community of inquiry framework to analyze levels of teaching presence and social presence in blended and online learning environments and the relative impact of these constructs on cognitive presence. The BL students reported higher levels of all three components of the teaching presence construct (instructional design, facilitation of discourse, and direct instruction) and two of the three social

17 17 presence components (affective communication and open communication with instructor and other students). Researchers have reasoned that this could explain why meta-analyses show students in blended courses outperforming students in purely online courses. 3.3 Learner Satisfaction Research has also attempted to identify what leads to learner satisfaction in BL courses. Learner satisfaction is complicated, as it is influenced by the expectations, goals, and preferences of the learners as well as the course design and implementation. For example, a learner who expects no instructor interaction and experiences a moderate interaction level might be very satisfied, while a student expecting high interaction might be dissatisfied with the same moderate level. Vignare (2007) suggested a need to identify and benchmark the elements that lead to satisfaction in BL environments. A couple of studies have attempted to do this. Dziuban, Moskal, & Hartman (2005) used factor analysis to identify two dimensions of satisfaction, which they identify as learning engagement and interaction value. They later identified eight elements that contribute to learner satisfaction in online and blended courses (Moskal, Dziuban, & Hartman, 2010). Rothmund (2008) also found a correlation between learner interaction and satisfaction in blended courses. Akyol, Garrison, & Ozden (2009) found that students valued both social presence and teaching presence in their BL experiences. Because student satisfaction is connected to learner dispositions as well as course design, some researchers have hypothesized that student satisfaction with BL will be moderated by the learner characteristics. For example, the course satisfaction of non-traditional adult learners may be significantly influenced by the convenience, flexibility, and reduced opportunity costs of BL (Moskal, Dziuban, & Hartman, 2010). Researchers at the University of Central Florida sought to

18 18 understand how satisfaction might be influenced by generational markers (e.g., millennials, boomers, genxers) (Dziuban, Moskal, & Hartman, 2005; Dziuban, Moskal, Brophy-Ellison, & Shea, 2007; Dziuban, Moskal, & Futch, 2007). They found significant differences, with millenials feeling the least favorable towards blended environments. Another study used Kolb s (1984) Learning Style Inventory to determine if students with different learning styles had different satisfaction profiles in a blended course (Akkoyunlu & Soylu, 2008). The researchers hypothesized that assimilators (oriented towards information & ideas) would be more satisfied by online components, while divergers (oriented towards people & feelings) would be drawn to the face-to-face components of the blend. The researches found significant differences between the views of the two groups on all six measured elements of the blended environment. Assimilators scored the online and face-to-face environments as a close 1st and 2nd place, while divergers placed the greatest value on the face-to-face environment and least value on the online environment. Graff (2003) also looked at learning styles within blended courses and found a significant difference in how learners experienced a sense of community based on learning style. Students with intuitive learning approaches (judgments based on feelings) experienced less sense of community than other students in the blended courses. Rovai & Jordan (2004) conducted a study looking at the sense of community developed in fully online, blended, and traditional face-to-face courses. The findings suggest that blended learning experiences build a stronger sense of community among students than either fully online or traditional courses. Available evidence shows that many learners value both the richness of interactions in a face-to-face environment and the flexibility, convenience, and reduced opportunity costs associated with online learning. Perhaps this combination is why most research finds high levels of student satisfaction with BL options. However, researchers must be cautious because, as with

19 19 learning effectiveness, it is the pedagogical possibilities (like interaction levels or learner choice) enabled by the modality that can lead to satisfaction. In fact, recent research looking at over a million course evaluations across different course formats showed that modality does not impact the dimensionality by which students evaluate their course experiences (Dziuban & Moskal, 2011, p. #abstract in press page # not available yet). 3.4 Faculty Satisfaction The University of Central Florida found that 88% of instructors were satisfied with teaching blended courses and that 81% were definitely willing and 13% probably willing to teach another blended course (Dziuban et al., 2004). This compares to the 87% faculty satisfaction with teaching purely online courses, with only 67% definitely willing to teach purely online again (Dziuban et al., 2004). All faculty involved in a BL pilot program at the University of Wisconsin were happy with their first blended teaching experience and were willing to recommend the approach to others (Aycock, Garnham, & Kaleta, 2002). However, in contrast a BL pilot project at the Rochester Institute of Technology found only 41% were willing to teach a blended course again (Vignare & Starenko, 2005). Faculty satisfaction with course redesign efforts involving online learning was influenced by student-related factors (satisfaction, interaction levels, performance, etc.), instructor-related factors (recognition, reliable technology, collaboration opportunities, professional development, etc.), and institution-related factors (workload, compensation, promotion and tenure issues, etc.) (Bolliger & Wasilik, 2009). Student-related factors were of first concern because faculty are not likely to be satisfied if their students are not having a positive learning experience (Moskal, Dziuban, & Hartman, 2010). Larsen & Sung (2009) used student ratings of faculty competence as an indirect measure of faculty satisfaction in blended and online contexts; they reported

20 20 student ratings of exceptional faculty competence for purely online, blended, and face-to-face modalities at 76%, 44%, and 37% respectively. Many teachers thrive on the relationships that they build with their students, and those relationships motivate and renew them in their teaching. Certain forms of online learning that involve minimal interaction between teacher and student or in which the instructor feels reduced to being an assignment grader can be unsatisfying for faculty (Hawkins, Barbour, & Graham, 2011; 2012). BL can provide added satisfaction for faculty (particularly faculty used to traditional modes of teaching) because they are able to have some face-to-face relationship building with the students. As with online learning, researchers recognize that workloads for faculty in BL contexts may increase, especially in the first few years of teaching (Colwell, 2006). Such an increase is partially due to the expectation of increased communication with the instructor via and other ICT channels, as well as increased time required to learn new technologies and teaching strategies and time required to create and maintain online materials. Faculty who are not compelled to teach blended courses and receive proper preparation are more likely to be satisfied with the outcomes (Vignare, 2007). While sustained professional development may increase faculty workload, it may also facilitate faculty collaboration and teaching success that increase faculty satisfaction. Effective programs specifically created for supporting faculty in BL course redesign can be found at the University of Central Florida (Dziuban, Harman, Moskal, et al., 2004), the University of Wisconsin-Madison (Garnham & Kaleta, 2002; Aycock, Garnham & Kaleta, 2002), Rochester Institute of Technology (Vignare & Starenko, 2005; Starenko, Vignare, & Humbert, 2007), and the University of Calgary (Vaughan & Garrison, 2005, 2006).

21 Access & Flexibility Access issues are driving much of the growth of BL around the world. However, little research has specifically investigated access and flexibility. A 2007 survey of 366 U.S. school districts, representing approximately 3,632 schools and 2 million students, identified five primary reasons for online and BL options: 1. Offering courses not otherwise available at the school 2. Meeting the needs of specific groups of students 3. Offering Advanced Placement or college-level courses 4. Reducing scheduling conflicts for students 5. Permitting students who failed a course to take it again (Picciano & Seaman, 2007, p. 9) All of these reasons are related in some way to providing learners with access to educational opportunities that would be difficult or impossible to provide in the traditional way due to costs. In particular, the online and blended options can address small and rural school needs as well as diverse student needs like advanced placement and credit recovery courses. A BL initiative at Rochester Institute of Technology (RIT) targeted students who were deaf or hard of hearing, as well as English language learners (ELLs) (Long et al., 2007). Those subgroups had greater access to classroom participation as discussions were moved from the face-to-face context to asynchronous discussion forums. As a result, satisfaction levels for both subgroups were higher than for the normal student population. In higher education the opportunity costs for participating in traditional learning options are too high for many potential students who have work, children at home, or other commitments that would make a rigid school schedule inaccessible. Flexible online options reduce the

22 22 opportunity costs associated with time and place scheduling, but often come at the cost of losing access to a high-touch, relationship-oriented environment with instructors and peers. Data from an international study investigating the use of information and communication technologies in higher education found that change towards more flexible student options is happening at a slow pace (Collis & Wende, 2002). BL options provide students with opportunities for benefit/ cost tradeoffs relevant to their own circumstances and preferences. Blended options can be of particular interest to institutions that reach out to non-traditional learners in the local communities where they are already well known and trusted; in the literature this idea, referred to as localness, has been an emphasis of Sloan Consortium grant funding (Mayadas & Picciano, 2007; Moloney, Hickey, Bergin, Boccia, Polley, & Riley, 2007; Sachs, 2007). 3.6 Cost Effectiveness Cost effectiveness drives institutions towards adopting BL approaches in higher education (Betts, Hartman, & Oxholm, 2009). The University of Central Florida, for example, has been able to reduce costs with improved scheduling efficiencies and reduced need for physical infrastructure for BL courses (Dziuban et al., 2004, 2011). The National Center for Academic Transformation supported early research to examine whether universities could engage in large-scale course redesign that would simultaneously decrease cost and improve learning outcomes ( They offered $6 million in grants to 30 institutions to engage in course redesign projects and follow detailed cost analysis procedures. Half of the course redesigns involved BL with reduced classroom seat time, seven involved significant technological enhancements with no reduction in seat-time, and eight involved moving completely online with some optional face-to-face class sessions (Graham & Allen, 2009). Twigg (2003) reported collective cost savings across all the projects of $3.6 million each year,

23 23 with institutional cost reductions ranging from 20% to 84% with an average savings of 40%. Additionally, the projects reported quality improvements such as higher course completion rates, student satisfaction, retention rates, and attitudes towards the subject matter. Historically corporations have been more effective at reporting figures related to cost effectiveness than institutions of higher education. However, Blain (2010) reported that only 3% of global companies and 26% of European companies measure return on investment (ROI). Bersin and Associates conducted a large-scale study to identify what works in corporate BL (Bersin, 2003), looking closely at BL training programs in 16 large corporations, with an emphasis on determining ROI. Almost all of the BL programs studied generated an ROI of 100% or better. Factors related to cost reduction for the blended programs included reduction in wait time for training, reduction in training hours and associated salary expenses, and reduction in facilities and other training costs. Institutions that have reported significant ROI due to BL training include Avaya (Chute, Williams, & Hancock, 2006), Microsoft (Ziob & Mosher, 2006), IBM (Lewis & Orton, 2006), and Intel (Mahesh & Woll, 2007). 4. Recommendations for Future Research With increasing use of blended learning in higher education, corporate training, and K-12 learning contexts, more theoretically grounded research is needed to guide practice. The foundation for creating scholarly knowledge lies in theoretical development, including frameworks that can address any of the three important activities of knowledge creation: explore, explain, or design (Gibbons & Bunderson, 2005). Exploring is the process of identifying, describing, and categorizing, which must include documenting the wide range of different blended learning models and developing taxonomies that enable meaningful systemization of the models for analysis of both physical and pedagogical characteristics of the blends. In addition,

24 24 theoretical frameworks must explain the how and why behind BL outcomes, which may include extending prominent distance learning theories to the BL context as well as developing new theories specifically created for understanding BL issues. Also, the development of design-based instructional theories will aid practitioners in tailoring BL to their specific needs. Research shows some general patterns across the three learning modalities (face-to-face, online only, and blended) but still needs to uncover the root causes for improved learning outcomes in BL contexts. Productive areas for exploration include the impact of interaction quantity and quality on outcomes, the relationship between face-to-face and online presence in BL community building, and the correlation between time on task and BL outcomes. Garrison and Vaughan (2008) identified establishing and maintaining cognitive presence in blended communities as the top BL research need related to the CoI framework. Other researchers have suggested exploring the relationship between learner characteristics and success with different blends (Dziuban et al., 2007). This will be particularly important as adolescent learners, who lack such adult learning characteristics as high self-regulation, are introduced to BL. Although a number of studies point to student and faculty satisfaction with BL, more is needed to link the satisfaction data with specific BL design features. Such connections enable better understanding of how. BL designs impact not only performance and satisfaction outcomes but also learners dispositions towards the subject matter, which are likely to significantly impact learner persistence beyond the boundaries of a course. While access and flexibility are among the most frequently cited purposes for adopting blended learning, very little research has sought to quantify the impact of BL on accessibility as well as on opportunity costs associated with increased flexibility. Similarly, empirical research related to the cost effectiveness of BL is limited, particularly in K-12 and higher education.

25 25 Research related to both access and cost issues could help institutions, designers, and learners develop and select the models of BL best suited to their needs. Finally, although BL is often described in terms of blending the physical attributes of online and face-to-face instruction, the psycho-social relationships are the issues at the core of blended learning research and design. Because education is a design-oriented field like engineering or architecture, it must be concerned with tradeoffs involving cost, efficiency, and effectiveness. The foundational challenge of BL research is seeking to understand (1) what humans do very well and (2) what machines do very well, so that the strengths of both can be maximized as they are blended in the service of learning. References Akkoyunlu, B., & Soylu, M.Y. (2008). A study of student s perceptions in a blended learning environment based on different learning styles. Educational Technology & Society, 11(1), Akyol, Z., Garrison, D.R., & Ozden, M.Y. (2009). Online and blended communities of inquiry: Exploring the developmental and perceptional differences. International Review of Research in Open and Distance Learning, 10(6), Allen, I. E., & Seaman, J. (2007). Online nation: Five years of growth in online learning. Sloan Consortium report. Aycock, A., Garnham, C., & Kaleta, R. (2002). Lessons learned from the hybrid course project. Teaching with Technology Today, 8(6), Bernard, R.M., Abrami, P.C., Borokhovski, E., Wade, C.A., Tamim, R.M., Surkes, M.A, & Bethel, E.C. (2009). A meta-analysis of three types of interaction treatments in distance education. Review of Educational Research, 79(3), doi: / Bersin Betts, K., Hartman, K., & Oxholm, C. (2009). Re-examining & repositioning higher education: twenty economic and demographic factors driving online and blended program enrollments. Journal of Asynchronous Learning Networks, 13(4), Blain, J. (2010). Current learning trends in Europe and the United States. Cegos Group white paper. Bolliger, D.U., & Wasilik, O. (2009). Factors influencing faculty satisfaction with online teaching and learning in higher education. Distance Education, 30(1), doi: /

26 26 Boyle, T., Bradley, C., Chalk, P., Jones, R., & Pickard, P. (2003). Using blended learning to improve student success rates in learning to program. Journal of Educational Media, 28(2), doi: / Borup, J., Graham, C. R., & Velasquez, A. (2011). The use of asynchronous video communication to improve instructor immediacy and social presence in a blended learning environment. In A. Kitchenham (Ed.), Blended learning across disciplines: Models for implementation (pp ). Hershey, PA: IGI Global. Borup, J., West, R. E., & Graham, C. R. (2012 in press). Improving online social presence through asynchronous video. Internet and Higher Education. doi: /j.iheduc Bransford, J.D., Brown, A.L., & Cocking, R.R. (1999). How people learn: Brain, mind, experience, and school. Washington, D.C.: National Academy Press. Bunderson, C.V. (2003). Four frameworks for viewing blended learning cases: Comments and critique. Quarterly Review of Distance Education, 4(3), Burkhardt, H., & Schoenfeld, A.H. (2003). Improving educational research:toward a more useful, more influential, and better-funded enterprise. Educational Researcher, 32(9), doi: / x Chute, A.G., Williams, J.O.D., & Hancock, B.W. (2006). Transformation of sales skills through knowledge management and blended learning. In C.J. Bonk & C.R. Graham (Eds.), Handbook of blended learning: Global perspectives, local designs (pp ). San Francisco, CA: Pfeiffer. Clark, R.E. (1983). Reconsidering research on learning from media. Review of Educational Research, 53(4), Collis, B. (2006). Putting blended learning to work. In C.J. Bonk & C.R. Graham (Eds.), Handbook of blended learning: Global perspectives, local designs (pp ). San Francisco, CA: Pfeiffer. Collis, B., & van der Wende, M. (2002). Models of technology and change in higher education: An international comparative survey on the current and future use of ICT in higher education. New Economy. Center for Higher Education Policy Studies report. Colwell, J.L. (2006). Experiences With A Hybrid Class : College Teaching, 2(2), Cottrell, D.M., & Robison, R.A. (2003). Case 4: Blended learning in an accounting course. Quarterly Review of Distance Education, 4(3), Cross, J. (2006). Forward. In C.J. Bonk & C.R. Graham (Eds.), Handbook of blended learning: Global perspectives, local designs (p. xvii-xxiii). San Francisco, CA: Pfeiffer. Dowling, C., Godfrey, J., & Gyles, N. (2003). Do hybrid flexible delivery teaching methods improve accounting students learning outcomes? Accounting Education, 12(4), doi: / Dron, J., Seidel, C., & Litten, G. (2004). Transactional distance in a blended learning environment. Alt-J: Research in Learning Technology, 12(2), doi: / Dubin, R. (1978). Theory building. New York, NY: The Free Press. Dziuban, C.D., Hartman, J. L., Cavanagh, T. B., & Moskal, P.D. (2011). Blended courses as drivers of institutional transformation. In A. Kitchenham (Ed.), Blended learning across disciplines: Models for implementation (pp ). Hershey, PA: IGI Global.

27 27 Dziuban, C., & Moskal, P. (2011, in press). A course is a course is a course: Factor invariance in student evaluation of online, blended and face-to-face learning environments. The Internet and Higher Education. doi: /j.iheduc Dziuban, C., Hartman, J., Juge, F., Moskal, P., & Sorg, S. (2006). Blended learning enters the mainstream. In C.J. Bonk & C.R. Graham (Eds.), Handbook of blended learning: Global perspectives, local designs (pp ). San Francisco, CA: Pfeiffer. Dziuban, C., Hartman, J., Moskal, P., Sorg, S., & Truman, B. (2004). Three ALN modalities: An institutional perspective. Elements of Quality Online Education: Into the Mainstream (pp ). Needham, MA: Sloan Consortium. Dziuban, C., Moskal, P., & Futch, L.S. (2007). Reactive behavior, ambivalence, and the generations: Emerging patterns in student evaluation of blended learning. In A.G. Picciano & C.D. Dziuban (Eds.), Blended Learning: Research Perspectives (pp ).Needham, MA: Sloan Consortium. Dziuban, C., Moskal, P., & Hartman, J. (2005). Higher education, blended learning and the generations: Knowledge is power-no more. In J. Bourne & J. C. Moore (Eds.), Elements of quality online education: Engaging communities (pp ). Needham, MA: Sloan Consortium. Dziuban, C., Moskal, P., Brophy-Ellison, J., & Shea, P. (2007). Technology-enhanced education and millennial students in higher education. Metropolitan Universities, 18(3), Fetters, M. L., & Duby, T. G. (2011). Faculty development: A stage model matched to blended learning maturation. Journal of Asynchronous Learning Networks, 15(1), Garnham, C., & Kaleta, R. (2002). Introduction to hybrid courses. Teaching with Technology Today, 8(6). Retrieved from Garrison, R. D., & Vaughan, N. D. (2008). Blended learning in higher education: Framework, principles, and guidelines. San Francisco, CA: Jossey-Bass. Gibbons, A. S., & Bunderson, V. (2005). Explore, explain, design. Encyclopedia of Social Measurement (Vol. 1, pp ). San Diego, CA: Elsevier. Giddens, A. (1984). The constitution of society: Elements of the theory of structuration. Elements. Berkeley, CA: University of California Press. Graff, M. (2003). Individual differences in sense of classroom community in a blended learning environment. Journal of Educational Media, 28(2), doi: / Graham, C.R. (2006). Blended learning systems: Definition, current trends, and future directions. In C.J. Bonk & C.R. Graham (Eds.), Handbook of blended learning: Global perspectives, local designs (pp. 3-21). San Francisco, CA: Pfeiffer. Graham, C.R., & Allen, S. (2009). Designing blended learning environments. In P. L. Rogers, G. A. Berg, J.V. Boettecher, C.Howard, L. Justice, & K. Schenk (Eds.), Encyclopedia of Distance Learning (Vol. 2, pp ). Hershey, PA: IGI Global. Graham, C.R., & Robison, R. (2007). Realizing the transformational potential of blended learning: Comparing cases of transforming blends and enhancing blends in higher education. In A.G. Picciano & C.D. Dziuban (Eds.), Blended Learning: Research Perspectives (pp ). Needham, MA: Sloan Consortium. Grgurovic, M. (2010). Technology-enhanced blended language learning in an ESL class: A description of a model and an application of the Diffusion of Innovations theory. (Unpublished Doctoral Dissertation). Iowa State University, Ames, IA.