Running head: COI AND LEARNING PRESENCE Learning Presence as a Moderator of Cognitive Presence in the Community of Inquiry Model

This study of over 2000 US college students examines the Community of Inquiry framework (CoI) in its capacity to describe and explain differences in learning outcomes in hybrid and fully online learning environments. We hypothesize that the CoI model’s theoretical constructs of presence reflect educational effectiveness in a variety of environments, and that online learner self-regulation, a construct that we label “learning presence” moderates relationships of the other components within the CoI model. Consistent with previous research (e.g., Means, Toyama, Murphy, Bakia & Jones, 2009; Shea & Bidjerano, 2011) we found evidence that students in online and blended courses rank the modalities differently with regard to quality of teaching, social, and cognitive presence. Differences in help seeking behavior, an important component of self-regulated learning, were found as well. In addition, results suggest teaching presence and social presence have a differential effect on cognitive presence, depending upon learner’s online self-regulatory cognitions and behaviors, i.e. their learning presence. These results also suggest a compensation effect in which greater self regulation is required to attain cognitive presence in the absence of sufficient teaching and social presence. Recommendations for future research and practice are included. COI AND LEARNING PRESENCE 3 Cognitive Presence in The community of Inquiry Model as a Function of Self-regulated Learning Between the fall 2008 and fall 2009 academic years more than one million new college students in the United States took an online course for the first time. Adding this number of students to online education in a single year, an increase of more than 20%, represents a milestone of sorts. The addition of these new learners is the largest single-year numerical increase since such data have been collected, bringing the total number of online students to more than 5.5 million (Allen & Seaman, 2010). The current growth rate among online students is more than ten times the growth rate in higher education generally, which saw an increase of less than 2% overall. The data suggest that online education continues to grow phenomenally with about 30% of all US college students taking at least one online course (Allen & Seaman, 2010). Other research suggests that this growth is likely to continue and that we may see as many as 50% of college students in online courses by 2014 (Christensen, Horn, Caldera & Soares, 2011). In addition to the millions of students in fully online courses, we are now seeing growth in the numbers of students enrolled in blended or hybrid courses in which part of the instruction is carried out online with a reduced portion remaining in the classroom. Some forecast (Watson, 2008) that this sector will grow even faster than fully online education, especially in pre-college settings. The present study seeks to understand the experiences of these many millions of students, and the millions more predicted to follow them, from both conceptual and empirical perspectives. Given the ongoing growth of online and blended education, it is crucial that we gain insights into successful learners engaged in these modalities and begin to develop a profile of how people learn online. We know that online learning is similar to and yet distinct from COI AND LEARNING PRESENCE 4 classroom learning. Clearly the absence of the dynamic of face-to-face interaction represents a change and we know that for most students online education is carried out in an asynchronous internet-based format. The US Department of Education reports, for example, that more than 11,200 college level programs have been designed for delivery fully at a distance and that 98% of these use asynchronous internet-based technologies to some extent (Parsad & Lewis, 2008). There is a longstanding belief that distance education requires a greater degree of selfdirectedness and self-reliance and it seems probable that learners in asynchronous, largely textbased online courses face challenges requiring considerable persistence and determination. This belief is supported by data that has implications for the continued expansion of online learning. For example in a recent national study of online education (Allan & Seaman, 2010) a majority of institutions in the US reported that online learning requires more discipline on the part of the student and that this represents a barrier to continued growth of online education. The issues of online student discipline, self-directedness, and self-reliance are important factors and in this paper we will look at possible distinctions confronting fully online learners and learner who enroll in blended courses. If we are to understand how people learn online, how it may be improved, and how and whether it should continue to grow in the now prevailing method (asynchronous, internet based) it is important to frame these issues within a model that allows for more coherent description and explanation. We posit that two existing theories provide this framing, the Community of Inquiry (CoI) model and self-regulated learning (SRL) theory. The community of inquiry framework The Community of Inquiry model outlines the behaviors and processes required to enable knowledge construction in asynchronous online environments through the development of COI AND LEARNING PRESENCE 5 various forms of “presence”. The CoI framework discusses three forms of presence: teaching-, social-, and cognitive presence (Garrison, Anderson, & Archer, 2001). The model articulates these conceptual components considered foundational to successful knowledge building in collaborative online environments. The framework theorizes online learning as a result of collaborative work among active participants in learning communities characterized by instructional orchestration suitable to the online environments (teaching presence) and a supportive mutually respectful online setting (social presence). The teaching presence construct describes online instructional skills such as organization, design, discourse facilitation, and direct instruction (Anderson, Rourke, Garrison, & Archer, 2001; Garrison & Arbaugh, 2007) and articulates the specific behaviors likely to result in a productive community of inquiry (e.g., Shea, Li, Swan, & Pickett, 2005). Social presence reflects online discourse that promotes positive affect, interaction, and cohesion (Rourke, Anderson, Garrison, & Archer, 1999) and supports a collegial, collaborative environment. The model also references cognitive presence, a multivariate measure of significant learning that results from the cyclical process of practical inquiry within such a community of learners. The specific form of interaction within the cognitive presence construct thus reflects a pragmatic view of learning (Dewey, 1933; Lipmann, 2003; Pierce, 1955). However the model as a whole can be seen to articulate the knowledge construction features of Larreamendy-Joerns and Leinhardt’s (2006) “epistemic engagement” model with teaching presence serving the overarching instructional function and social presence supporting productive participation. In a recent review of the research on CoI, Garrison and Arbaugh (2007) concluded that there was a need for more quantitatively-oriented studies, more cross-disciplinary studies, and opportunities for identifying factors that moderate or extend the relationship between the COI AND LEARNING PRESENCE 6 framework's components and course outcomes. The study presented here attempts to build on these recommendations with a quantitative approach to an analysis of thousands of students studying across many different disciplines. Seeking to add to the literature on CoI in the context of the most recent, rigorous, and comprehensive review of online learning (Means, et. al, 2009), results presented here extend the framework’s components with an additional focus on learners and their differential capacity to self-regulate in online and blended settings. Self-regulated learning Despite the increasing interest in the application of the CoI framework as an overarching conceptual model explaining the nature of asynchronous distance learning, with some exceptions (Shea & Bidjerano, 2010; Shea et al., in press) very few studies have taken into account individual-level determinants, that can play an equally important role in students’ perceptions of cognitive engagement and gains. The notion that individual difference characteristics interact with instructional environment in intricate ways to produce specific learning outcomes is not new. The aptitude-treatment interaction framework, described by Cronbach and Snow (1977), suggests that “one size does not fit all” for the effectiveness of any instructional method or strategy depends largely on the individual attributes of the learner. Neither instruction in isolation, nor learners’ characteristics by themselves predict learning, rather the combination of the two results in optimal learning outcomes (Snow, 1989). Although the ATI framework has been originally developed to explain the complex interplay between instruction, referred as to treatment, and student ability, it is currently agreed upon that the paradigm can be extended to include a broad range of interactions between individual traits (intelligence, learning style, personality, motivation, effort, self-efficacy, meta-cognition, self-regulation) and the quality of COI AND LEARNING PRESENCE 7 learning environment (teacher, teaching style, instructional methods, strategies, social milieu, instructional mediums) with their effect on learning and educational outcomes. Of the aforementioned person-level characteristics, individual differences in selfregulated learning have emerged as a powerful explanatory mechanism of the relationship between instruction and educational outcome. Hence, the second conceptual framework that may provide guidance into the experience of online learners is the SRL theory. Several socialcognitive models of SRL have been proposed and extensively studied in research (Zimmerman, 2001). Despite the subtle nuances in definitions and operationalization of the construct, there is a consensus in the literature that self-regulation is a cyclical, recursive process that encompasses goal setting, planning, executing actions, monitoring, self-reflection and self-assessment. Selfregulated learners set proximal attainable goals based on refined self-knowledge and analysis of the complexity of the learning task; structure their own learning environment in ways conducive to learning; chose appropriate learning strategies, constantly monitor the progress made towards the task, and evaluate the extent to which goals have been achieved (Zimmerman, 2000; Pintrich, 2000). Therefore, SRL is necessarily a multidimensional construct that “involves cognitive, affective, motivational and behavioral components that provide the individual with the capacity to adjust his or her actions and goals to achieve the desired results in light of changing environmental conditions” (Zeidner, Boekaerts, & Pintrich, 2000, p. 751). SRL is therefore more than what has been conventionally referred to as volition, self-discipline or self-directedness. Metacognition or knowledge about oneself as a learner (Boekaerts, 1995), motivation and selfefficacy beliefs (Schunk& Zimmerman, 1998), awareness and use of appropriate learning strategies that optimize performance (Garcia & Pintrich, 1994; Pintrich, Smith, Garcia, & COI AND LEARNING PRESENCE 8 McKeache, 1993; Pressley, Beard El-Dinary, Wharton-Mcdonald, & Brown,1998) are all important components of SRL. Researchers have consistently demonstrated the beneficial effects of self-regulatory behaviors on students' academic performance in traditional classrooms (Pintrich et. al., 1993; Winne, 2005; Zimmerman, 1998; Zimmerman, 2008). Given the relatively recent advent and uniqueness of online education, the research of the role of learner self-regulation in online courses is only in its inception. However, because learning in virtual environments removes well-known conventions of direct student-teacher contact and immediacy that is familiar in the face-to-face format, the ability of learners to engage in SRL behaviors is construed as a crucial factor for success online (Barnard, Lan, To, Paton, Lai, 2009). Findings from the few studies that have examined aspects of SRL (i.e., self-efficacy, strategy use, effort and persistence) in online courses mirror what has been known about the factors and processes operating in the context of the traditional classroom. Learners’ performance in a distance education course is facilitated to a great extent by the level of confidence in their skills to acquire course specific content and technology skills, use of their time wisely (Gibson, 1998; Lynch &Dembo, 2004; Wang & Newlin, 2002), and ability to set specific learning goals geared towards mastery (Curry, Haderlie, & Ku, 1999; Schrum and Hong, 2002). Online learners who are better able to manage their study time effectively, structure their learning environment, and seek assistance when faced with setbacks are more likely to be successful as well (Hsu, Ching, Mathews & Carr-Chellman, 2009; Puzziferro,2008; Wang & Newlin, 2002). Self-regulated learners tend to have more positive perceptions of online courses (Howland & Moore, 2002) and report higher satisfaction with the latter if they engage in cognitive learning strategies such as rehearsal, and elaboration (Puzzifierro, 2008). In addition, COI AND LEARNING PRESENCE 9 student achievement in online courses depends on learners’ perceptions of course communication and collaboration partly because of the affordances for self-regulatory learning behaviors (Barnard, Paton, & Lan, 2008). More recently, in an attempt to incorporate self-efficacy and effort regulation, both important aspects of self-regulated learning, in a more comprehensive model combining elements of the CoI and SRL frameworks, Shea and Bidjerano (2010) suggested that SRL represents an important mediator of the links between teaching presence, social presence, and cognitive presence. Capitalizing upon the idea that CoI cannot be considered apart from SRL, in another mixed-methods study, Shea and colleagues (in press) reported that online students, monitor their time and cognitive strategies, regulate their study environment, and exercise control over their interactions with technology, peers, and faculty to maximize learning. Even though we have gained significant insight about the learners’ SRL as a critical factor in both traditional and online learning, our literature review suggests that a common denominator of previous research is its exclusive attention to linear prediction models in which SRL contributes either singularly or incrementally to cognitive, affective or behavioral outcomes [Of all of the reviewed work, only one study (e.g., McManus, 2000) has explored the role of SRL as a potential moderator of the link between treatment and outcome.] Although useful, such additive models often do not reflect the nature and complexities of the individual-environment interactions. Guided by the understanding that person-environment interactions are omnipresent (Snow, 1989), we sought to delineate the impact of SRL as a potential modifying variable that shapes how learners respond to teaching and social aspects of online instruction, as defined by