Investigating Classroom Implementations of Game-Based Learning

• In Event: Designing and Implementing Educational Games: Exploring How to Increase the Effectiveness of Games for Learning

Abstract

There is growing evidence that narrative-centered learning environments, a class of
game-based learning environments that embed educational content in interactive story scenarios, are an effective medium for fostering student learning, problem solving, and engagement (Johnson, 2010; Nelson & Ketelhut, 2008; Rowe, Shores, Mott, & Lester, 2011). A key benefit of narrative-centered learning environments is their capacity to discreetly support students’ learning processes by tightly integrating educational and narrative elements, leveraging the motivational benefits of narrative and digital games. Care must be taken in designing narrative-centered learning environments to ensure that story elements do not distract from educational goals (Adams, Mayer, MacNamara, Koenig, & Wainess, 2012; Rowe, McQuiggan, Robison, & Lester, 2009). Yet, interactive narrative features—such as compelling plots, believable characters, and fantasy settings—can provide naturalistic mechanisms for guiding inquiry and scaffolding learning.

Over the past several years, our lab has been developing CRYSTAL ISLAND, a narrative-centered learning environment for middle school microbiology (Rowe, Shores, Mott, & Lester, 2011). Designed as a supplement to classroom science instruction, CRYSTAL ISLAND’s curricular focus has recently been expanded to include literacy education based on Common Core State Standards for reading informational texts. The latest edition of CRYSTAL ISLAND was developed with the Unity game engine, which provides 3D rendering, audio, and input device capabilities, and enables deployments in schools through web browsers.

To better understand how game-based learning can be utilized in classrooms, we have been investigating school-based implementations of CRYSTAL ISLAND. During school-based implementations of CRYSTAL ISLAND, learning occurs both within the game and outside the game in classroom activities. In the game, students solve a science mystery over several sessions, and these sessions are interleaved with class periods dedicated to complementary paper-based activities that are completed individually or in small teams. In both the game and classroom, students read texts and complete “concept matrices” with varying degrees of pre-specified structure, and in some cases they generate concept matrices from scratch.

We recently completed a classroom study of CRYSTAL ISLAND in 14 schools involving over 1,200 students. Of these, 864 students completed the study’s pre- and post-study assessments within the allotted time period. Paired samples t-tests were conducted to determine if students improved their reading skills as a result of the CRYSTAL ISLAND unit. There was a significant increase observed between students’ scores before (M = .60, SD = .26) and after (M = .74, SD = .28) the intervention; t(863) = 16.21, p < .01; Cohen’s d = .52, r = .25. We did not find evidence of significant differences between females and males on three out of five dimensions of the Intrinsic Motivation Inventory (Interest/Enjoyment, Pressure/Tension, or Value/Usefulness). However, results indicated that females reported expending greater Effort/Importance than males. Conversely, males reported greater perceptions of Competence than females. While game-based learning environments show considerable potential for classroom use, individual differences in students’ intrinsic motivation for game-based learning point toward the need for further study of design principles and technologies for effectively diagnosing and scaffolding student learning, motivation and engagement.

Authors

• James Lester, North Carolina State University

• Eleni V. Lobene, North Carolina State University

• Bradford Mott, North Carolina State University

• Jonathan Rowe, North Carolina State University