Abstract:

Recent reviews of blended online instruction have raised significant concerns about its effectiveness. While a comprehensive review of over 1000 studies (up to 2006), showed a modest positive impact for blended over face-to-face instruction (in K–12 education, career technology, medical and higher education, as well as corporate and military training), this effect was due as much to elements of learning time and instructional as it was to the media itself (US Department of Education, 2010). Recent experience with massive open online courses (MOOCs) and online university programs have produced very mixed results. Studies at both Stanford and Harvard (e.g., Ho et al., 2014) show that even for well-prepared students simply providing online tools rarely enhances instruction or completion rates. For struggling students, a recent study (Bettinger, Loeb, Fox & Taylor, 2015) compared in-person courses at DeVry University against DeVry university courses taught online. The study disturbingly found that online courses could significantly depress learning for both current and future courses. According to the authors, taking a course online reduces student achievement by about one-quarter to one-third of a standard deviation, as measured by course grades, and reduces the probability of remaining enrolled by three to ten percentage points (over a base of 68 percent). Taking a course online also reduces student grade point average in the next term by more than one-tenth of a standard deviation.The primary take-away from these studies is that designers of online learning must not rush to use technology, and thereby lose focus on decades of research of how people learn. In this paper, I will examine recent findings on the marginal performance of online instruction, and point to key aspects of learning research that must be considered in the design of online instruction. For example, research shows a clear advantage of active learning over lecturing (Freeman et al., 2014), and the importance of attention to cognitive science for generative learning (e.g., Fiorella & Mayer, 2015). The paper will also stress the need for greater attention the use of formative and summative measurement to inform the design and deployment of online and blended instruction (e.g., Pellegrino, 2012, 2013).

Keywords:

Online, learning, technology.