Elsevier

Computers & Education

Volume 159, December 2020, 104008
Computers & Education

Effects of solo vs. collaborative play in a digital learning game on geometry: Results from a K12 experiment

https://doi.org/10.1016/j.compedu.2020.104008Get rights and content

Highlights

  • We have performed an experiment in K12 classrooms on solo and collaborative gameplay.

  • Students in solo condition had higher in-game engagement, were more linear and solved more puzzles.

  • Students in collaborative condition less linear in their pathways and more exploratory with the game.

  • Teachers should expect different outcomes when implementing solo or collaborative games in the classroom.

Abstract

Digital games for learning are one of the most prominent examples of the use of technologies in the classroom, where numerous studies have presented promising results among children and adolescents. However, scarce evidence exists regarding different ways of implementing games within the classroom and how those affect students' learning and behaviors. In this study we explore the effect that collaboration can have in digital gameplay in a K12 context. More specifically, we have designed a 2 × 2 experimental study in which high school first year students participated in solo or collaborative gameplay in pairs, solving puzzles of diverse difficulty, using Shadowspect, a digital game on geometry. Our main results, computed by applying learning analytics on the trace data results, suggest that students playing solo had higher in-game engagement and solved more puzzles, while students collaborating were less linear in their pathways, skipping more tutorial levels and were more exploratory with Shadowspect features. These significant differences that we observe in solo and collaborative gameplay call for more experimentation around the effect of having K12 students collaborate on digital tasks, so that teachers can take better decisions about how to implement these practices in the classrooms of the future.

Introduction

Around the world, educational technology is being introduced slowly into the classroom and holds the potential to have a transformative effect on the educational landscape Baumöl and Bockshecker (2017). One of the most prominent examples is the use of digital games for learning De Freitas (2006). Numerous authors Prensky (2003); Shaffer (2006); Gee (2003) have expressed that well designed digital games represent outstanding opportunities for learning through an enjoyable experience. Playing video games is essential part of young generation's daily life Buckingham and Willett (2013). A recent report on media consumption Ofcom (2019), builds profile snapshots by age, estimating that 40% of kids 3–4 years old play games for nearly 6 h a week, 66% of kids 5–7 for 9 h a week, 81% of kids 8–11 for around 10 h a week and 77% of kids 12–15 for 12 h a week. Many studies provide ample evidence that using video games with children and adolescents can support various skills and life outcomes such as academic domain-specific learning including science Li (2013); Council et al. (2011) and math Divjak and Tomić (2011); Starkey (2013), executive functions Homer et al. (2018), involvement in real life sports Adachi and Willoughby (2016) or for acquiring health knowledge Baranowski et al. (2016).

While many teachers report a positive attitude towards games being used in K12 classrooms and believe that they can improve learning and curriculum, the actual number of teachers who are implementing digital games in their curriculum is contrarily low De Grove et al. (2012); Pastore and Falvo (2010). One of the main implementation barriers reported by teachers is uncertainty and limited literature about how to actually effectively implement games in the classroom Watson and Yang (2016); An and Cao (2017), and thus there is a necessity to support evidence-based decision making regarding how different teachers' decisions impact students' experiences when implementing games in classrooms. One strong attraction for teachers to use games in classroom is because of the belief that it is a powerful tool to foster collaboration among students and develop 21st Century skills such as collaborative problem solving Kim and Shute (2015). In this experimental case study, we explore what the effect of having students play a game solo vs. with others is, with the ultimate goal of supporting teachers' implementation decisions for game-based curriculum. A few studies compared solo vs. social play in commercial gaming contexts Arellano et al. (2016, pp. 184–193); Kaye and Bryce (2014), but little work is done in educational games, especially using in-game analytics. Given that learning games are frequently introduced as classroom activities Squire (2005), the existence of barriers to effectively implementing game-based learning Watson and Yang (2016); An and Cao (2017) and the importance of collaboration to prepare younger generations for the future UNESCO Bangkok (2016), educators would have more confidence in using games in classrooms if they knew what to expect from having students play these games alone or collaboratively and the potential implications for their learning.

We organize the rest of the manuscript in the following sections. Next Section 2 reviews related literature and Section 3 presents the current case study. Section 4 details the methods of the study, including an overview of Shadowspect, the context, experimental design, data and metrics. Section 5 presents the results and in Section 6 we discuss those results. Finally, in Section 7 we finalize with conclusions and future follow-up ideas.

Section snippets

Collaboration in game-based learning

There are two related yet subtly different concepts for collaboration in the context of game-based learning. First, collaborative game, refers to a specific type of game that has explicit game mechanics that require two or more players to work on joint tasks or quests Berland and Lee (2011). Second, collaborative gameplay, is a broader category that describes non-competitive gameplay where two or more players socially engaging while playing together with a common objective in mind. Note that

Current study

To address the gap identified in the literature review, we conducted a 2 × 2 factorial experiment to investigate the influence of gameplay style (solo or collaborative gameplay) when using the geometry game Shadowspect, while also varying puzzle complexity with the second factor. The gameplay style condition entangles either having students play solo or collaboratively in pairs. The puzzle complexity condition involves one set of puzzle levels slightly more difficult than the other. All

Shadowspect overview

Shadowspect1 is a 3D digital educational game. Shadowspect lies within the category of puzzle games (like the Witness or Bridge Builder). Shadowspect has clearly-defined goals, rules, obstacles for the players to overcome and provides only intrinsic rewards (satisfaction for getting the right answer). Each puzzle presents a number of silhouette views for a 3D figure where each figure is built by using a series of 3D geometric primitives such

Results

The results section presents comparison between groups and conditions to respond the research questions. In order to find if a difference on a quantitative variable is statistically significant we report the difference between means of each group, Student's t-Test with its p-value, and the effect size via Cohen's d.

Differences between gameplay styles

Our initial hypothesis was that a Collaborative play mode could make students slightly less engaged with the game environment and tasks, due to more talking, switching controls, exchange of ideas and social interactions, which in turn could be helpful to actually make students more engaged with the overall activity and less prone to disengage due to problems solving puzzles, but at the same time could make students slower to solve simple tasks.

Based on the data-driven engagement analytics, we

Conclusions

We have performed a cohort-based factorial experiment focusing on the effect of solo vs. collaborative gameplay using the digital geometry game Shadowspect with high school first year students. The main findings include that students playing Solo had a higher in-game engagement and solved more levels, while Collaborative students were less linear in their gameplay patterns (i.e., skipping more tutorial levels and demonstrating more exploratory behaviors with Shadowspect features). Our results

Credit author statement

José A. Ruipérez-Valiente: Conceptualization, Methodology, Formal analysis, Investigation, Data Curation, Writing - Original Draft, Visualization. Yoon Jeon Kim: Conceptualization, Methodology, Investigation, Writing - Review & Editing, Project administration, Funding acquisition.

Acknowledgements

We want to acknowledge support from the MIT-SPAIN ''la Caixa'' Foundation SEED FUND and the Spanish Ministry of Economy and Competitiveness through the Juan de la Cierva Formaci\'on program (FJCI-2017-34926).

References (47)

  • D.G. Arellano et al.

    Measuring affective, physiological and behavioural differences in solo, competitive and collaborative games

  • M.E. Arterberry et al.

    Collaborative problem solving in five-year-old children: Evidence of social facilitation and social loafing

    Educational Psychology

    (2007)
  • T. Baranowski et al.

    Games for health for children-current status and needed research

    Games for Health Journal

    (2016)
  • U. Baumöl et al.

    Evolutionary change of higher education driven by digitalization

  • M. Berland et al.

    Collaborative strategic board games as a site for distributed computational thinking

    International Journal of Game-Based Learning (IJGBL)

    (2011)
  • D.M. Bressler et al.

    Inside, outside, and off-site: Social constructivism in mobile games

  • D. Buckingham et al.

    Digital generations: Children, young people, and the new media

    (2013)
  • P.M. Chapman

    Models of engagement: Intrinsically motivated interaction with multimedia learning software

    (1997)
  • N.R. Council

    Learning science through computer games and simulations

    (2011)
  • S. De Freitas

    Learning in immersive worlds: A review of game-based learning

    (2006)
  • B. Divjak et al.

    The impact of game-based learning on the achievement of learning goals and motivation for learning mathematics-literature review

    Journal of Information and Organizational Sciences

    (2011)
  • M. Freire et al.

    Game learning analytics: Learning analytics for serious games

  • J.P. Gee

    What video games have to teach us about learning and literacy

    Computers in Entertainment (CIE)

    (2003)
  • Cited by (17)

    • Examining the effects of students' self-efficacy and prior knowledge on learning and visual behavior in a physics game

      2022, Computers and Education
      Citation Excerpt :

      Educational game is a game serving to create engaging learning experience for meeting certain educational goals and outcomes (Connolly et al., 2012; Hainey et al., 2016). Educational games have been designed to teach a variety of subjects (e.g., mathematics--Ruipérez-Valiente & Kim, 2020; Vanbecelaere et al., 2020; science--Clark et al., 2011; Lester et al., 2014; Martin et al., 2019) or to assist learners in developing cognitive skills (e.g., executive function--Homer et al., 2018; visuospatial skills--Wauck et al., 2017). There is a general consensus that educational games, if used appropriately, can contribute to positive cognitive and affective outcomes (Clark et al., 2016; Connolly et al., 2012; Hainey et al., 2016; Karakoç et al., 2020).

    • The use of a personalized learning approach to implementing self-regulated online learning

      2022, Computers and Education: Artificial Intelligence
      Citation Excerpt :

      Scholars have recognized the benefits of using computers and communication technologies to support students’ learning activities (Brezovszky et al., 2019; Hwang et al., 2012; McLaren et al., 2022; Ruipérez-Valiente & Kim, 2020).

    View all citing articles on Scopus
    View full text