Desktop version

Home arrow Philosophy

  • Increase font
  • Decrease font


<<   CONTENTS   >>

Online Courses

The rapid rise of distance-based learning and hybrid, online courses, as well as the general size and scale that these contexts afford, holds some promise for deep use of EDM/LA. Since the early 2000s, students have increasingly been exposed to and opted to enroll in online learning opportunities. Entire colleges have been “built” that offer the full complement of their academic programs online. Recent estimates indicate that nearly 34 percent of students take at least one online course during their undergraduate degree, a percentage that is growing every year (Allen & Seaman, 2014). Simultaneous with the explosion of online learning, research has been quick to investigate the role of learning strategies for successful achievement with online learning. For example, Wallace, Kupperman, Krajcik, and Soloway (2000) and his colleagues highlighted the complexity of online information seeking and the difficulties teachers and students have in negotiating these learning spaces. Ligorio (2001) and Tsai and Tsai (2003) indicated that this difficulty requires awareness and metacognitive strategies when engaging online inquiry learning beyond those required in off-line inquiry.

The rapid emergence of online options for students has been facilitated by the emergence of learning management systems (LMSs). Sixty-three percent of online courses use LMSs to serve the online learning needs of students and instructional goals of their teachers (Green & Hughes, 2013). The rise in computer-mediated and online education has opened up new approaches and avenues for collecting and processing data on students and course activities; every instructional transaction can be immediately recorded and added to a database. Many institutions use some form of basic learning analytics technology as part of their LMS platforms. Some of these data points are low level “count” metrics, such as click and engagement patterns and the timestamps associated with these actions (Jones, 2015). Other metrics offer a more robust window into strategic processing such as resource selection, help seeking, reading and writing habits, and multiple window browsing (Siemens, 2013). When these individual data are examined in isolation, their meaning and interpretation are quite limited, but when aggregated and examined through an EDM/LA framework, these rich data trails offer an opportunity to explore learning strategy and achievement from new and multiple angles (Macfadyen & Dawson, 2012). Toward this goal, EDM/LA provides a potentially powerful set of tools to aid instructors in iterative changes to their course materials to improve their effectiveness and supports for student success (Wang & Kelly, 2017).

Several online learning researchers have investigated the relationships between interaction sequences and learning outcomes using EDM/LA methods such as sequence mining to model learner behavior (Kock & Paramythis, 2011) and sequential pattern analysis (Perera, Kay, Koprinska, Yacef, & Zai'ane, 2009). Using these EDM/LA techniques, these researchers have been able to identify various sequences for high- and low-performing learners. Most notably, this research found that poor strategy choice, such as ignoring the intended path sequences designed by the course designers and jumping back and forth between resources as the need for information to complete assessment and activities required, led to lower achievement in the course. Similarly, in a study of more than 43,000 students participating on a Massively Online Open Course (MOOC), Mukala, Buijs, and Van Der Aalst (2015) found that better performing students moved through course materials in a very structured way, following a logical path through video-based material and course assessments and watching videos in “batches.”

It is important to note the study of online learning courses and MOOCs through EDM/LA does not just have a one-way function that focuses on student outcomes. It has also presented a robust methodology to examine how the design of learning environments can encourage or impede student learning. Aligning EDM/LA and the design of instructional environments allows researchers to examine if learners are engaging with course materials as the designer intended by examining students’ actual behaviors and strategies when engaged in the act of learning (Dalziel et al., 2016; Mor, Ferguson, & Wasson, 2015). This is an important area for research, as we know that reliance on student self-reports of learning behaviors or merely on student outcomes does not necessary paint an accurate picture of student learning. Nguyen, Huptych, and Rienties (2018) examined students’ time spend studying and engagement to an instructor’s intended learning behaviors, finding that at the macro level there was good alignment, but at the micro level, which was captured through an EDM/LA exploration of trace data, there were significant differences in the behaviors students exhibited and those intended by the learning design. They highlight that instructors that have access to this type of data could be better positioned to reflect on and adjust their teaching practices and instructional activities in the future. Further, they postulate that sharing this information regarding how fellow students are interacting with online course materials through the course, could also help learners to self-regulate their learning more efficiently. Unfortunately, most of the research on the use of EDM/ LA to facilitate automated changes instruction and scaffolding in MOOCS and online learning settings has fallen far short of the hype failing to implement robust models of instructional intervention (Baker, 2016).

Serious Games

Since the late 1990s, researchers have increasingly touted the capacity of video games as educational tools to cultivate situated activity, problem solving, and collaboration (Gee, 2003; Hamari et al., 2016; McGonigal, 2011; Squire, 2006). EMD/LA has been a popular methodology for examining learning within serious gaming environments and how students’ strategic processes correlate with positive outcomes (Liu, Kang, Liu, Zou, & Hodson, 2017). Within this pool of research, a number of studies have been able to isolate the strategies that students enact in the pursuit of goal attainment. For example, Kerr and Chung (2012), using click stream data, were able to differentiate between gamers who employed a systematic trial and error problem-solving strategy from those who simply guessed solutions randomly. Similarly, Snow, Allen, Jacovina, and McNamara (2015) examined the behavior traces learners left when navigating through an educational game, and found that students who demonstrated more controlled choice patterns generated higher quality learning outcomes compared to students who exhibited more disordered choice patterns.

In a more robust implementation of EDM/LA, Liu, Alexandrova, and Nakajima (2011) collected trace data from interactions within a gaming environment and paired it with survey data regarding students’ experience during game play. Their analysis resulted in the identification of five different problem-solving strategies: solution development, experimenting, solution review, solution reuse, and reading the tutorial, that were linked to variations in students’ motivation and experience of flow states. Sun, Wang, and Chan (2011) manipulated the amount and type of scaffolding provided in a game to understand the impact these various supports had on student strategy use. Through analysis of user behavior logs, they were able to isolate elimination techniques and trial-and-error approaches to problem solving. Interestingly, they also found that increasing levels of scaffolding within the game also led some learners to game the system. These learners became dependent on the tools, and while they ended up completing more problems, they also learned fewer problem-solving strategies. This finding is not insignificant as it points to the need for systems to detect patterns of strategy use within a game to adapt the learning space in ways that are most productive for different types of learners.

 
<<   CONTENTS   >>

Related topics