Experimental benefits of GBL?
Posted April 3, 2012on:
I do not know any academic or well-informed individual who can read the claims of most experimental (or quasi-experimental) studies and agree wholeheartedly with them. Typically in these studies, one factor is withheld from one group but applied in another.
I am a proponent of game-based learning approaches, but I cannot help but process this study with some skepticism.
Here is a study in a nutshell:
The study examined student attention and engagement during 8 lesson cycles. CES is a K-5 public school with self-contained classrooms. The experimental group participated in a technology lesson that used digital games as its method of delivery or practice for the students. The control group participated in a technology lesson that utilized an alternative learning strategy. The alternative learning strategies selected for comparison with Digital Game-Based Learning (DGBL) included: a) carousel brainstorming, b) team webbing, c) concept attainment, d) jigsaw, e) learning stations, and f) roundtable. These alternative learning strategies were interactive, stimulating, and engaging for students. Research data was captured in the form of student observations and post-trial student surveys. Both groups were taught the same technology and content objectives utilizing varying instructional strategies.
The study concluded that:
Digital Game-Based Learning… groups showed more student engagement and time on task behavior than the alternative strategies. However, several trials did demonstrate that the alternative strategies produced more lesson engagement and a higher time-on-task group average than DGBL.
In terms of research design, the researcher did not indicate if intact classes were used. One is left to assume this was the case. The researcher also did not mention if the same teacher conducted both classes. If this factor was not controlled, then it could have impacted the results.
In most experimental studies where humans are subjects, there are confounding variables or other variables you cannot control like learner preferences, prior exposure to gaming, learner expectations, etc.
Classroom observations provide snapshot at best. Gaming is a lifestyle. Students learn with games even when they are out of class. So while the study opted to focus on what happened in class, the reality is that one cannot control for this factor.
The researcher might argue that it was what happened only in class that mattered and sought to determine if DGBL was a viable strategy. Staying on task and being engaged certainly contribute to learning, but he did not report if learning actually took place.
Instead, the focus seemed to be DGBL as an instructional strategy. But teachers teaching does not guarantee that learners are learning.
I would predict that if the results were analyzed statistically, there would be no or little difference between treatments. In other words, GBL is no less effective than other strategies, or better but not statistically so.
The question then is not whether to use GBL, but when to employ it.