An analysis of research trends in dissertations and theses studying blended learning Jeffery S. Drysdale ⁎, Charles R. Graham, Kristian J. Spring, Lisa R. Halverson Department of Instructional Psychology and Technology, Brigham Young University, Provo, UT, USA abstract article info Article history: Accepted 15 November 2012 Available online 23 November 2012 Keywords: Blended learning Hybrid learning Research Scholarship Graduate student research This article analyzes the research of 205 doctoral dissertations and masters' theses in the domain of blended learning. A summary of trends regarding the growth and context of blended learning research is presented. Methodological trends are described in terms of qualitative, inferential statistics, descriptive statistics, and combined approaches to data analysis. Research topics are divided into nine topics (learner outcomes, dispo- sitions, instructional design, interaction, comparison, demographics, technology, professional development, and other), each containing several sub-topics. Patterns in these topics are analyzed to identify gaps in research and to highlight opportunities for future research as the field of blended learning continues to grow. © 2012 Published by Elsevier Inc. 1. Introduction Thousands of corporate training programs, institutions of higher ed- ucation, and K-12 schools participate in blended learning (Picciano, Seaman, Shea, & Swan, 2012; Staker et al., 2011). Nearly half of four-year degree-granting postsecondary institutions and two thirds of two-year institutions in the U.S. were reported as offering “hybrid/ blended online learning” courses in 2007 (Parsad & Lewis, 2008). Blend- ed learning is also known to be on the rise in Australia (Eklund, Kay, & Lynch, 2003), Canada (Collaboration for Online Higher Education Research, 2011), and the UK (Sharpe, Benfield, Roberts, & Francis, 2006). Blended learning—in its simplest form, the thoughtful integra- tion of online and face-to-face-instruction (Garrison & Kanuka, 2004; Graham, 2006, 2013)—is being used with increased frequency around the world. Concurrent with this rise in implementation, research on blended learning has increased over the past decade, with much of the seminal work occurring in higher education contexts (Halverson, Graham, Spring, & Drysdale, 2012). An analysis of dissertations and theses can provide a window into the state of research in a particular domain and can be a powerful tool for understanding trends in theory develop- ment, methods, themes, and gaps within the domain (Davies, Howell, & Petrie, 2010). In this article we identify and analyze over 200 theses and disserta- tions written in the past decade in the domain of blended learning. Our analysis documents the growth of blended learning research and iden- tifies demographic, methodological, and topical trends in that body of research. 2. Literature review Over the past decade there have been many attempts to define blended learning (Graham, 2013; Mayadas & Picciano, 2007; Oliver & Trigwell, 2005; Picciano & Seaman, 2009; Vaughan, 2007). Graham (2013) reviewed the various definitions with their strengths and limita- tions. Additionally, Graham examined the literature related to learning effectiveness, learner satisfaction, faculty satisfaction, access and flexi- bility, and cost effectiveness. Graham identified the need for more theoretically grounded research. He also outlined opportunities for research exploring the link between satisfaction data and specific blended learning methods, accessibility, opportunity costs, cost effec- tiveness, and psycho-social relationships. Other efforts have recently been made to assess the state of blended learning research. Halverson et al. (2012) sought to identify the most impactful scholarship and research in blended learning. This study iden- tified the top 50 articles, 25 edited book chapters, 10 books, and 15 non-academic publications ranked by citation count. These seminal works indicate where the conversations on blended learning research are taking place. However, dissertations and theses related to blended learning were not considered part of that analysis. Additional actions to synthesize the results of Blended Learning research have also been undertaken. A meta-analysis conducted by the U.S. Department of Education reviewed 99 studies on online or blended learning (Means, Toyama, Murphy, Bakia, & Jones, 2009). Dissertations were included as part of the initial search, but most did not include enough data to calculate effect size and therefore did not meet the criteria for inclusion. Researchers found that students partici- pating in online or blended instruction produced stronger learning outcomes than those that participated only in face-to-face instruction with a mean effect size of s +0.20, p b .001. Only five instances of K12 instruction qualified for the analysis, leaving a need for further study on the subject. Internet and Higher Education 17 (2013) 90–100 ⁎ Corresponding author at: Brigham Young University, 150 MCKB, Provo, UT 84602, USA. Tel.: +1 801 709 4982. E-mail address: jeff.drysdale@byu.edu (J.S. Drysdale). 1096-7516/$ – see front matter © 2012 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.iheduc.2012.11.003 Contents lists available at SciVerse ScienceDirect Internet and Higher Education