The Impact of Game Type on Mathematical Learning: A meta-analysis and critical review

• In Event: Poster Session 04
  In Poster Session: PS 04 Topic - Technology, Media & Child Development

Abstract

Educational games have gained increased interest in mathematics education; however, numerous studies show the evidence supporting the effectiveness of math games is mixed (Clark et al., 2016, Fabian et al., 2016 Girard et al.,2013) and there is no clear explanation as to why (Dubé et al., 2019) The relation between a games’ pedagogical foundation and its ability to improve a specific learning outcome must be assessed to identify which game types are better suited to teach different mathematics content (Kiili, 2005). Specifically, math games can be considered effective in the extent to which they develop and facilitate the acquisition of factual, procedural, and conceptual knowledge. While studies have examined the ability of games to support math learning overall (Lee et al., 2019), no research looked at how game types contribute to these three foundational types of mathematical knowledge. Thus, the goal of the study is to systematically identify, categorize, and compare mathematical game studies to examine whether the different pedagogical approaches found in games facilitate different types of mathematical knowledge. The following research questions guided the investigation:
1) Which pedagogical approaches are used in mathematics games to support mathematical learning?
2) Which types of mathematics knowledge are promoted by mathematics games?
3) How effective is each pedagogical approach at improving each knowledge type?
Method: A systematic literature search identified 713 studies published 2010 to 2020 that included keywords related to “game”, “math”, and “K-12” students. After removing duplicates (n=86), screening abstracts (n=389), and applying strict pre-determined exclusion criteria (n=212), 26 studies remained. These studies were evaluated with respect to a) the pedagogical approach used in the games (i.e., Kebritchi & Hirumi’s 2008), b) the type of mathematics knowledge they facilitated (Bisanz & LeFevre, 1990; Rittle-Little Johnson, 2017), and c) their effect on math learning.
Results: R1) Only 23 out of 26 studies used games with a clear pedagogical approach. Most studies used direct instruction (38.40%) or experiential learning (26.92%), while fewer studies used discovery (11.52%) or constructivist games (11.52%). None of the studies took a situated cognition approach and some studies had games that could not be classified (11.52%). R2) 46 % of studies aimed to improve factual, 23% procedural, and 73% conceptual knowledge, with many studies (35%) aiming to improve more than one knowledge type. Of studies that focused on just one knowledge type, 19% focused solely on improving factual and 46% focused solely on conceptual knowledge, but no studies focused on procedural knowledge (Table 1). R3) Direct instruction games and experiential learning games produce a medium sized effect (g = 0.510, g = 0.46), whereas discovery and constructivist learning games produce a small sized effect (g = 0.236, g = 0.208) on overall math learning. Further, the effect of each pedagogical approach on each knowledge type is mixed (Table 2). Overall, behaviorally oriented pedagogies are still dominant in math games and the effectiveness of each pedagogical approach varies as a function of knowledge type.

Authors

• Gulsah Kacmaz, McGill University
  Presenting Author

• Adam Kenneth Dubé, McGill University
  Non-Presenting Author