Asia-Pacific Edu Res

https://doi.org/10.1007/s40299-020-00523-z

REGULAR ARTICLE

Using Game-Based Learning to Support Learning Science:

A Study with Middle School Students
Meiqian Wang1 • Xudong Zheng2

Ó De La Salle University 2020

Abstract Game-based learning refers to establishing Introduction

learning environments that involve digital or non-digital
games to enhance students’ knowledge and skill acquisi- Game-based learning refers to a type of learning environ-
tion. Previous studies indicated that game-based learning is ment that involves game content and game play to enhance
superior to the conventional instructional method, but few knowledge and skill acquisition (Qian and Clark 2016).
compared the differences between different game types. Typically, game-based learning includes game activities
This study examined the effects of game-based learning on regarding problem solving and offers challenges to learners
Chinese middle school students’ learning of science and with a sense of achievement. Well-designed educational
their self-efficacy. We implemented an experiment to games show great potential to enhance teaching and
compare the effects between digital and non-digital game- learning. In science education, game-based learning is an
based learning. Our results showed that students in game- effective method to support students’ learning of scientific
based learning groups performed significantly better in the knowledge. Prior studies have reported that game-based
content knowledge assessment and reported higher self- learning not only have positive effects on students’ learn-
efficacy than the traditional lecture group. No significant ing but also can enhance their self-efficacy (Meluso et al.
difference was found between the digital and non-digital 2012; Sung and Hwang 2013; Hung et al. 2014). With
game groups when considering science learning perfor- these merits, more and more researchers and educators start
mance, but students of the digital game group showed to pay attention to integrating game-based learning into
significantly higher self-efficacy than those of the non- science classrooms.
digital game group. Implications of the findings for future In China, there is a growing interest in STEM education,
research and practice of game-based learning are discussed. especially in science education (Zhao and Thomas 2016;
Liang et al. 2017). Natural science is a required subject for
Keywords Science education
Digital game-based students in the 9-year compulsory education. Scientific
learning
Non-digital game-based learning
Self-efficacy
literacy is also considered a core literacy of all students
Chinese middle school students (Chen et al. 2009; Gao et al. 2016). However, in most
science classrooms, teacher-centered lecturing is still the
dominant instructional method. What’s more, Chinese
students have shown better response to traditional lecture
& Meiqian Wang styles relative to western students (Dahlin and Watkins
[email protected]
2000; Cheng and Wan 2016). Game-based learning is a
Xudong Zheng method that shows the potential to transform the teaching
[email protected]
and learning of science (Hamari et al. 2016; Khan et al.
1
National Engineering Research Center for E-Learning, 2017; Cardinot and Fairfield 2019), but is it applicable for
Central China Normal University, Wuhan, China Chinese students? Further study is needed to answer this
2
College of Educational Information Technology, Central question. Moreover, there exist at least two types of edu-
China Normal University, Wuhan, China cational games: digital and non-digital games. However,

123
 M. Wang, X. Zheng

few studies have compared the effects of game-based showed significant increases in science content knowledge
learning on students’ learning and self-efficacy between the in both collaborative and single game player conditions.
two types of games. Previous research is mostly based on Non-digital games refer to any board-, card-, or other
digital games, and non-digital games have received less type of games that can be played in physical environments
attention (Naik 2014). Some digital game-based learning (Uzun et al. 2013). Different from digital games, non-
proponents presumed that digital media had certain hidden digital games do not need electronic devices and can be
qualities compared to non-digital media as concerns implemented in different scenarios. Thus, non-digital
learning (Prensky 2001; Tapscott 1998). While non-digital games are widely used in STEM education and have been
game-based learning proponents believed that non-digital found to be effective in enhancing learning (Mustafa et al.
games had many advantages over digital games (von Gil- 2011; Liu and Chen 2013). For example, by integrating a
lem and Alaswad 2016). Echoing Berland et al. (2011) role-playing game into chemistry classes, Cook (2014)
argued that the transparency of the rules of non-digital concluded that game participation helped to facilitate stu-
games was an important element of their enjoyment. Thus, dents’ learning of science knowledge about the chemistry
investigating and comparing effects of digital and non- of plastics. In another study using Lego bricks to teach
digital games on students’ learning is important. The goal elementary school students science concepts, researchers
of this study was to examine the effects of game-based noted that playing games with Lego improved both stu-
learning on Chinese middle school students’ learning of dents’ knowledge and problem-solving ability (Li et al.
science and their self-efficacy in learning. We implemented 2016). Card games are another choice. Morris (2011)
an experiment to compare the effects between digital and conducted a study to use a card game to help students learn
non-digital game-based learning. The following research chemical formulas and found that the card game effectively
questions guided the study: promoted learning.
While previous research has investigated the effective-
1. Whether digital games have better effects on students’
ness of using educational games to support learning, few
learning of science than non-digital games?
studies have compared the effects on learning between
2. Whether students who play digital games show better
digital and non-digital games. Some researchers compared
self-efficacy than students who play non-digital games
game-based learning with conventional instructional
when learning science knowledge?
method (Brom et al. 2011; McLaren et al. 2017). Within
one game type such as digital games, prior studies also
compared the effects of the single game player condition
Literature Review with the collaborative condition (Chen et al. 2015; Chen
and Law 2016). However, little empirical research has been
Game-Based Learning conducted to examine the differences in effects on learning
between digital and non-digital games. One goal of our
The major characteristic of game-based learning is inte- study was to investigate whether digital games have better
grating games into instruction to promote knowledge and effects on students’ learning of science than non-digital
skill acquisition (Qian and Clark 2016). During the games (RQ1).
gameplay, students need to solve problems and complete
challenges that are relevant to the course content. Previous Self-efficacy and Game-Based Learning
studies have reported positive effects of using game-based
learning in science education (Yien et al. 2011; Lin et al. An effective educational game should include appropriate
2013). In general, researchers have examined two types of challenges that can make students experience a feeling of
games: digital and non-digital games. A digital game refers self-efficacy (Johnson and Mayer 2010). In other words,
to a game played using electronic devices such as com- successful game-based learning should consider learners’
puters, mobile devices, and so forth. Prior studies have self-efficacy. Self-efficacy refers to an individual’s belief in
found that digital games are effective to improve students’ his or her innate ability to achieve goals (Bandura
learning of science knowledge. For instance, Sung and 1982, 1994). It reflects students’ confidence in their capa-
Hwang (2013) conducted an experiment of integrating bilities to exert control over their own emotion, motivation,
educational computer games into elementary school sci- and behavior in the course of learning. These cognitive
ence courses and found that students’ learning achieve- self-evaluations influence various aspects of students’
ments were improved. In another study of fifth graders who learning experience, including the goals for which they
played a computer game called Crystal Island to learn strive, the amount of time and effort involved in achieving
science, Meluso et al. (2012) indicated that students the goals, and the likelihood of reaching specific levels of
learning performance. Studies have shown that students

123
Using Game-Based Learning to Support Learning Science: A Study with Middle School Students

with high self-efficacy achieved better learning results Methods

(Yusuf 2011; Safaria and Ahmad 2013).
Many researchers have explored the effects of game- Participants
based learning on students’ self-efficacy (Meluso et al.
2012; Leonard et al. 2016; Lu and Lien 2019). For The participants of this study were three groups of 7th
instance, by integrating computer games into elementary graders from a city in Central China. In this middle school,
school science classes, Sung and Hwang (2013) noted that the 7th graders were divided into 12 classes. In this study,
computer games improved students’ self-efficacy of using we randomly selected three classes as the research subjects,
computers to learn. In a study of nursing students using each class had about 31 students (see Table 1). The average
classroom simulation, Goldenberg et al. (2005) found that age of students was about 12.6 years old. They all had
role-playing games enhanced students’ self-efficacy in some prior experience of game-based learning (about once
performing health teaching. According to Hung et al. or twice per term). In this study, the scientific concept
(2014), game-based learning can effectively improve stu- taught to students was the reflection of light, which
dents’ self-efficacy in learning compared to traditional includes the four major parts: (1) objects that emit light are
teacher-directed lecture method. Hence, successful inte- called light sources, (2) light travels in a straight path in a
gration of educational games into instruction can benefit uniform medium, (3) light reflects from a smooth surface at
students in both learning performance and self-efficacy. the same angle as it hits the surface, (4) the path of light is
Prior studies provide evidence that game-based learning reversible.
increases students’ self-efficacy in learning, compared to
traditional lecture method. This can be great merit of using
educational games in instruction. However, few studies Research Design and Procedures
focused on comparing the effects of digital and non-digital
games on learners’ self-efficacy. While both digital and This study used an experimental design. Students of the
non-digital games can be used to implement game-based three classes used different instructional approaches (see
learning approach, gameplay and game experience of the Table 1). Students of class A used a digital game while
two types of games are very different. Thus, it is important students of class B used a non-digital game. Class C was
to examine students’ self-efficacy when playing different the control group and used the traditional lecture method.
types of games. This was our second research question All the three classes were taught by the same teacher.
(RQ2). The digital game used by class A was called Lazors. It is
a free Android app and can be downloaded from app stores
such as Google Play. In essence, Lazors is a puzzle game of
Summary lasers. To complete challenges in the game, players have to
use mirrors and move the blocks to reflect the laser to hit
Game-based learning is to establish learning environments the target. Figure 1 presents several screen shots of Lazors.
that involve digital or non-digital games to enhance stu- In this study, students of class A played this game in a
dents’ knowledge and skill acquisition. Previous research computer lab using the BlueStacks App Player, which is
reported that game-based learning showed positive effects designed to run Android games on computers (see Fig. 2).
on students’ learning performance. Researchers also found The non-digital game used by class B was Laser Maze
that game-based learning could improve students’ self-ef- from the American company ThinkFun. This game also
ficacy in learning. Majority of prior studies investigated the requires using mirrors and other game tokens to direct the
effects of using game-based learning compared to con- laser through a series of mind-challenging mazes and hit
ventional instructional method, either using digital games the target (see Fig. 3). Figure 4 shows a picture of a student
or non-digital games. Few have made the comparison who were playing Laser Maze. The key ideas of this game
between the two types of games. This study explored the are similar to the digital game Lazors. The major game
effects of game-based learning on Chinese middle school blocks of Lazors and the game tokens of Laser Maze are
students’ learning of science and their self-efficacy in
learning. We implemented an experiment to compare the
effects between digital and non-digital game-based Table 1 Groups and interventions
learning. Natural class Class A Class B Class C

Number of students 31 30 32
Type of game Digital Non-digital None

123
 M. Wang, X. Zheng

Fig. 1 Screen shots of lazors

test before the class. Second, students started to take the

class. For the two game-based learning groups, the teacher
first spent about 15 min introducing the topic, learning
objectives, learning tasks, and rules of playing the game.
Next, students started to play the game and investigate the
reflection of light. They had to complete specified learning
tasks of different difficulty levels. During the game period,
the teacher provided support as needed. After about 25-min
game play, the teacher discussed what they had learned
from the game with the students and summarized the
course content. For the control group (the traditional lec-
Fig. 2 Students playing lazors ture group), the teacher also spent about 15 min introduc-
ing the topic, learning objectives, learning tasks, and
almost identical. The two games both have light sources, experiments about the reflection of light. Next, the teacher
reflectors, obstacles, and so on. The main idea of the two did several experiments and provided some examples to
games is similar: the player needs to use game blocks/to- demonstrate the reflection of light. The experiments
kens (e.g., a mirror) to direct a laser to hit targets. They involved the rectilinear propagation of light, the law of
both are good games for teaching scientific knowledge reflection, and the principle of reversibility. During the
about optics such the reflection of light. The major dif- experiments, students had to observe the phenomena and
ference between the two games is the rule of choosing record the results. Finally, the teacher spent about 10 min
challenges. The non-digital game Laser Maze allows discussing the reflection phenomenon with students and
players to choose challenges freely, while the digital game summarized the course content. After the class, students
Lazors requires completing challenges from the easy ones took the post-test and the self-efficacy survey.
to the difficult ones.
There was also a control group who learned the reflec- Measurements
tion of light using the traditional lecture method. All the
three groups of students took a 50-min class about the The pre/post-test used in this study was an assessment tool
reflection of light taught by the same teacher. Figure 5 provided by the school’s assessment department. This
presents the overall procedures. First, students took the pre- assessment was based on the national assessment question

123
Using Game-Based Learning to Support Learning Science: A Study with Middle School Students

Fig. 3 The principle of laser

maze

pool for middle school physics. As this study only focused

on one science concept the reflection of light, the assess-
ment tool only consisted of the questions relevant to the
concept. It included eight multiple choice questions, five
Fill in the Blank questions, and ten True or False questions.
The total of possible points were 36.
To measure students’ self-efficacy, we used the New
General Self-Efficacy Scale (NGSE), which was designed
and developed by Chen et al. (2001). It is a validated
assessment of self-efficacy and has high validity and reli-
ability. NGSE has eight five-point Likert items. One is the
lowest score, and five is the highest. The higher score
indicates higher self-efficacy. To assess the internal con-
sistency, reliability analysis was conducted, and the Fig. 4 A student playing laser maze
Cronbach’s alpha was 0.91. In other words, this scale was
reliable.

Data Analysis

To answer RQ1, we analyzed students’ performance in the

content knowledge assessment. First, we used an analysis
of variance (ANOVA) to compare the three groups’ per-
formance in the pre-test to see whether significant differ-
ences existed among groups before the experiment. Next,
we used an analysis of covariance (ACNOVA) to analyze
the effects of using the proposed approach on students’
science learning achievements. Tukey’s honest significant
difference (HSD) post hoc tests were used for pair-wise
comparisons. To answer RQ2, we analyzed students’ self- Fig. 5 Experimental design for the learning activities
efficacy scores in the NGSE scale. We also used the
ANOVA technique to compare the differences between the Results
three groups. For the post hoc tests, we used Tukey’s HSD.
All the statistical analyses were done using SPSS V22.0. RQ1: Whether Digital Games have Better Effects
All statistically significant results were reported at p \ .05 on Learning Science Concepts than Non-digital
level, unless specified otherwise. Games?

Table 2 shows the descriptive statistics of students’ pre-test

and post-test scores. Students’ performance in the pre-test
was similar. Class A (the digital game group) had the

123
 M. Wang, X. Zheng

Table 2 Descriptive statistics of students’ pre-test scores and post- the Class A with the Class C score of 25.95 (p \ 0.01). In
test scores addition, scores of Class B (adjusted mean = 29.62) were
Mean (SD) also significantly higher than those of Class C (p \ 0.01).
No significant difference was found between Class A and
Group Class A (n = 31) Class B (n = 30) Class C (n = 32)
Class B.
Pre-test score 19.87 (4.32) 20.70 (4.13) 20.12 (4.39) Therefore, compared to the traditional lecture approach,
Post-test 28.48 (4.73) 29.83 (3.27) 25.90 (5.13) the game-based learning approaches showed better effects
score on students’ learning of science concepts. Our results also
imply that digital and non-digital games have similar
lowest mean score in the pre-test, and class B (the non- positive effects when students learn science concepts.
digital game group) had the highest mean score. The
ANOVA test results indicated that the difference of the RQ2: Whether Students Who Play Digital Games
pre-test scores among the three groups was not significant Show Better Self-efficacy than Students Who Play
(F(2, 90) = 0.299, p [ 0.05). In other words, students of Non-digital Games When Learning Science
the three groups had similar knowledge background about Knowledge?
the reflection of light before the experiment.
This study conducted an analysis of covariance Table 4 presents the descriptive statistics of students’ self-
(ANCOVA) using the students’ pre-test scores as the efficacy. Students of class A reported the highest self-ef-
covariate to exclude the impact of the pre-test on their ficacy, M = 4.55, SD = 0.48, while students of class C
science learning. After conducting the experiment, showed the lowest self-efficacy, M = 3.58, SD = 0.68. The
ANCOVA was performed on the post-test scores, in which results of the ANOVA test found that significant differ-
the pre-test was the covariant, the post-test scores were the ences in students’ self-efficacy existed among the three
dependent variable and the ‘‘different instructional meth- groups, F(2, 90) = 22.488, p \ 0.01. The results of the post
ods (three groups)’’ were the control variable, to examine hoc Tukey HSD tests indicated that students of class A
the relationships among the post-test results of the three (M = 4.55 SD = 0.48) showed a significantly higher self-
groups. As shown in Table 3, the ANCOVA results showed efficacy than students of other two classes. In addition,
that the variance between the three groups is significant students of class B (M = 4.24, SD = 0.61) showed a sig-
(F = 6.864, p \ 0.01). In other words, the post-test scores nificantly higher self-efficacy than students of class C
were significantly different due to the different instruc- (M = 3.58, SD = 0.68). Our results suggest that by using a
tional methods. Furthermore, post hoc analysis was per- game-based learning approach, students tend to have
formed to examine specific differences in achievement higher self-efficacy than those in a traditional lecture class
between the groups. A Tukey’s HSD post hoc test revealed when learning science. Our results also indicate that digital
that the scores of Class A were significantly higher than games may bring students higher self-efficacy than non-
those of Class C, comparing the adjusted mean of 28.64 for digital games.

Table 3 The ANCOVA result of students’ post-test scores for the three groups
Variable Group N Mean SD Adjusted mean F(2, 89) Post hoc
*
Post-test score Class A 31 28.48 4.73 28.64 6.864 Class A [ Class C
Class B 30 29.83 3.27 29.62 Class B [ Class C
Class C 32 25.90 5.13 25.95
*p \ 0.01

Table 4 Descriptive statistics and the ANOVA result of students’ self-efficacy

Variable Group N M SD F(2, 90) Post hoc
*
Self-efficacy Class A 31 4.55 0.48 22.488 Class A [ Class C
Class B 30 4.24 0.61 Class B [ Class C
Class C 32 3.58 0.64 Class A [ Class B
*p \ 0.01

123
Using Game-Based Learning to Support Learning Science: A Study with Middle School Students

Discussion challenges for learners so that they can experience a feeling

of self-efficacy. In our study, both digital and non-digital
Game-Based Learning and Learning Achievement games included game challenges with clear goals, instant
feedback, and proper difficulty levels. Thus, students may
Our results showed that the two game-based groups per- feel they are capable to complete the challenges. On the
formed better than the traditional lecture group. This is contrary, during the lecture, students may not clearly see
consistent with previous studies. Many empirical studies the learning goals and progress. Without challenges of
have found that students in a game-based learning class different levels, students could not get enough feedback
show better learning performance than those in a traditional especially positive feedback, which made it difficult for
lecturing class (McLaren et al. 2017; Wang et al. 2018). them to get a sense of achievement in science learning.
According to constructivism, effective learning requires Therefore, students may show lower self-efficacy.
learners’ active construction of new knowledge, rather than A key finding of our study is that students playing digital
passive receiving (Amineh and Asl 2015). From this point games reported higher self-efficacy than those playing non-
of view, game-based learning creates a learning environ- digital games. As the two games are similar in game
ment that allows students to actively construct the meaning content, rules, and blocks, the different effects on students’
of scientific knowledge. When playing games, students self-efficacy are surprising. However, during the experi-
have to solve problems and complete challenges using the ment, we noticed important differences between the two
knowledge they learn. In contrast, during a traditional games, which may explain our results. In the digital game,
lecturing class, students are usually passive learners and do students can easily identify the gap between their current
not have to apply the new knowledge to problem solving. status and the goal required by the challenge. For a digital
Therefore, it is not so surprising that the traditional lecture game, this is a built-in functionality and visible on the
group showed the lowest learning achievement. screen. However, for the non-digital game, students have to
Our results also showed that the digital game group and figure out this gap by themselves. In addition, when play-
the non-digital game group had no significant differences ing the non-digital game, students need to place those game
in students’ learning of science. The results suggest that blocks on the game board and adjust the blocks to complete
digital and non-digital games have similar positive effects the challenge by hands. In contrast, the actions to play the
when students learn science knowledge. The key difference digital game are easier. Students only need to use the
between digital and non-digital games is whether games mouse and sometimes the keyboard as well to play the
are played using electronic devices. The game content and game. To some extent, the digital game simplifies the
rules can be identical for a digital educational game and a difficulty of the gameplay and requires simpler skills. Thus,
non-digital educational game. In our study, the digital both the two features of the digital game might have
game Lazors and the non-digital game Laser Maze are very improved our students’ self-efficacy.
similar in game content, rules, blocks, and so on. Hence, it There is an interesting Dunning Kruger effect (Dunning
is possible that for learning the science concept the 2011) in this study. The students in Class A (digital game)
reflection of light, the two games are both sufficient to help had the highest self-efficacy score, but they actually had
students understand the concept and facilitate learning. In lower learning achievement than students in Class B (non-
this scenario, the effectiveness of using digital and non- digital game). It might be because non-digital game could
digital games is similar. better promote students’ reflection. Digital game might
give students a more explicit and more direct feedback than
Digital Games and Self-efficacy non-digital game, which made it easier for them to over-
come the difficulty in accomplishing tasks and achieving
For self-efficacy, the results of our study indicate that by goals. However, the quick hit of success also brought a
using a game-based learning approach, students tend to problem that students had no time to reflect on what kind of
have higher self-efficacy than those in a traditional lecture knowledge they had used during the task. While, non-
class. According to previous studies, game-based learning digital game’s feedback was more implicit because the
can effectively improve students’ self-efficacy, compared light beam was invisible, students needed to check their
to traditional instructional method (Hung et al. 2014; plans again and again, which helped them to establish
Yükseltürk et al. 2018). Our results confirm this argument. connections between knowledge and behaviors. It is
Among the three groups, students of the lecturing group important to provide appropriate feedback to help learners
showed the significant lower self-efficacy than the gaming self-explain their learning process (Tan et al. 2007), and
groups. Johnson and Mayer (2010) pointed out that game- studies showed that reflection combined with feedback
based learning environments provide appropriate enhanced performance better than feedback alone (Anseel
et al. 2009). The reflective observation of the feedback

123
 M. Wang, X. Zheng

might lead to the construction of schemata (Kiili 2005). To In the future, we will replicate our experiment in more
some extent, non-digital game had a better effect on schools with more students.
stimulating deeper learning. Second, our experiment only focused on one scientific
concept and did not explore the long-term effects of dif-
Implications ferent types of games. In this study, we investigated stu-
dents learning of the reflection of light and assessed their
Based on our findings, we make two recommendations for learning performance using pre- and post-tests. However, if
researchers and educators who are interested in game- game-based learning is used in teaching several scientific
based learning. First, because of the merits of digital concepts or used for a whole semester, the two types of
games, we recommend that researchers and developers of games may show different effects. Our next step is to
educational games design and develop more digital games design several units of a middle school science course
that can support teaching science knowledge. Nowadays, using game-based learning and to examine the effects of
digital devices are accessible to students and teachers. As digital and non-digital games in students’ learning of
digital natives, students are familiar with the use of digital science.
devices. Our findings indicate that digital games can benefit Finally, in this study, we only collected quantitative data
both learning performance and self-efficacy of students. to compare students’ learning performance and self-effi-
Therefore, we believe that developing more digital games cacy in different settings. However, with only quantitative
for teaching science can enhance the quality of science data, it is hard to understand why students reported higher
education. According to our results, we make two sug- self-efficacy when playing digital games. Future research
gestions for developing digital games for learning. First, it should use mixed methods collecting both quantitative and
is important to provide immediate feedback for students so qualitative data (e.g., interviews, class observations, etc.) to
that they can quickly reflect on their game play and study the effects of different types of game-based learning.
learning. Second, digital games should integrate automatic
data collection and analysis components to enable teachers
to track students’ learning progress in real time. We believe Conclusions
that with those features digital games can better support
teaching and learning of science classes. Game-based learning shows great potential to support
Second, we recommend developing professional devel- teaching science knowledge and facilitating students’
opment programs focusing on integrating educational learning. While previous studies reported that game-based
games into instruction. Compared to traditional lecturing, learning showed positive effects on students’ learning
game-based learning has its merits including enhancing performance and self-efficacy compared to conventional
students’ self-efficacy, improving learning performance, instructional method, few studies have investigated the
and so forth. However, teachers may not own the knowl- different effects between digital and non-digital games.
edge of implementing game-based learning in their This study examined the effects of game-based learning on
instruction. Integrating educational games into instruction Chinese middle school students’ learning of science and
is not easy, because it requires not only the content their self-efficacy in learning. We implemented an exper-
knowledge of subject matter but also the knowledge of iment to compare the effects between digital and non-
games and game integration. Therefore, we believe that digital game-based learning. Our results showed that stu-
professional development programs about game-based dents in game-based learning groups (both digital and non-
learning are vital to teachers who would like to use edu- digital game groups) performed significantly better in the
cational games. We recommend that more future profes- content knowledge assessment and reported higher self-
sional development programs focus on game-based efficacy than the traditional lecture group. No significant
learning to help teachers build their knowledge of and difference was found between the digital game group and
develop their skills for effectively integrating educational the non-digital game group when considering science
games into instruction. learning performance, but students of the digital game
group showed significantly higher self-efficacy than those
Limitations and Future Research Directions of the non-digital game group.
According to our findings, we recommend that
This study has several limitations. First of all, our sample researchers and developers of educational games design
size was relatively small. On average, we had 31 students and develop more digital games that can support teaching
in each group. While such a sample size was sufficient for science knowledge. Meanwhile, we suggest developing
our study, the generalizability of our results can be limited. professional development programs focusing on integrating
educational games into instruction. While game-based

123
Using Game-Based Learning to Support Learning Science: A Study with Middle School Students

learning has advantages over the conventional instructional Dahlin, B., & Watkins, D. (2000). The role of repetition in the
method when teaching science, without appropriate edu- processes of memorising and understanding: A comparison of
the views of German and Chinese secondary school students in
cational games and sufficient professional development Hong Kong. British Journal of Educational Psychology, 70(1),
about game-based learning, teachers may not able to 65–84. https://doi.org/10.1348/000709900157976.
effectively use educational games in science classroom. Dunning, D. (2011). The Dunning–Kruger effect: On being ignorant
We believe in the future, there will be more and more of one’s own ignorance. In Advances in experimental social
psychology (Vol. 44, pp. 247–296). New York: Academic Press.
quality teachers who can effectively implement game- Gao, H., He, W., Zhang, C., & Ren, L. (2016). Building scientific
based learning if we provide more effective educational literacy in China: Achievements and prospects. Science Bulletin,
games and relevant professional development programs. 61(11), 871–874. https://doi.org/10.1007/s11434-016-1076-0.
Goldenberg, D., Andrusyszyn, M.-A., & Iwasiw, C. (2005). The effect
of classroom simulation on nursing students’ self-efficacy related
to health teaching. Journal of Nursing Education, 44(7), 310.
Hamari, J., Shernoff, D. J., Rowe, E., Coller, B., Asbell-Clarke, J., &
References Edwards, T. (2016). Challenging games help students learn: An
empirical study on engagement, flow and immersion in game-
Amineh, R. J., & Asl, H. D. (2015). Review of constructivism and based learning. Computers in Human Behavior, 54, 170–179.
social constructivism. Journal of Social Sciences, Literature and https://doi.org/10.1016/j.chb.2015.07.045.
Languages, 1(1), 9–16. Hung, C. M., Huang, I., & Hwang, G. J. (2014). Effects of digital
Anseel, F., Lievens, F., & Schollaert, E. (2009). Reflection as a game-based learning on students’ self-efficacy, motivation,
strategy to enhance task performance after feedback. Organiza- anxiety, and achievements in learning mathematics. Journal of
tional Behavior and Human Decision Processes, 110(1), 23–35. Computers in Education, 1(2–3), 151–166. https://doi.org/10.
https://doi.org/10.1016/j.obhdp.2009.05.003. 1007/s40692-014-0008-8.
Bandura, A. (1982). Self-efficacy mechanism in human agency. Johnson, C. I., & Mayer, R. E. (2010). Applying the self-explanation
American Psychologist, 37, 122–147. https://doi.org/10. principle to multimedia learning in a computer-based game-like
1037/0003-066X.37.2.122. environment. Computers in Human Behavior, 26(6), 1246–1252.
Bandura, A. (1994). Self-efficacy. In V. S. Ramachandran (Ed.), https://doi.org/10.1016/j.chb.2010.03.025.
Encyclopedia of human behavior (Vol. 4, pp. 71–81). New York: Khan, A., Ahmad, F. H., & Malik, M. M. (2017). Use of digital game
Academic Press. based learning and gamification in secondary school science:
Berland, M. (2011). Understanding strategic boardgames as compu- The effect on student engagement, learning and gender differ-
tational-thinking training machines. Tabletop analog game ence. Education and Information Technologies, 22(6),
design (pp. 167–173). Pittsburgh: ETC Press. 2767–2804. https://doi.org/10.1007/s10639-017-9622-1.
Brom, C., Preuss, M., & Klement, D. (2011). Are educational Kiili, K. (2005). Digital game-based learning: Towards an experien-
computer micro-games engaging and effective for knowledge tial gaming model. The Internet and Higher Education, 8(1),
acquisition at high-schools? A quasi-experimental study. Com- 13–24. https://doi.org/10.1016/j.iheduc.2004.12.001.
puters & Education, 57(3), 1971–1988. https://doi.org/10. Leonard, J., Buss, A., Gamboa, R., Mitchell, M., Fashola, O. S.,
1016/j.compedu.2011.04.007. Hubert, T., et al. (2016). Using robotics and game design to
Cardinot, A., & Fairfield, J. A. (2019). Game-based learning to enhance children’s self-efficacy, STEM attitudes, and computa-
engage students with physics and astronomy using a board game. tional thinking skills. Journal of Science Education and Tech-
International Journal of Game-Based Learning, 9(1), 42–57. nology, 25(6), 860–876. https://doi.org/10.1007/
https://doi.org/10.4018/IJGBL.2019010104. s10956-016-9628-2.
Chen, G., Gully, S. M., & Eden, D. (2001). Validation of a new Li, Y., Huang, Z., Jiang, M., & Ting-Wen, C. (2016). The effect on
general self-efficacy scale. Organizational Research Methods, pupils’ science performance and problem-solving ability through
4(1), 62–83. https://doi.org/10.1177/109442810141004. Lego: An engineering design-based modeling approach. Journal
Chen, C. H., & Law, V. (2016). Scaffolding individual and of Educational Technology & Society, 19(3), 143–156.
collaborative game-based learning in learning performance and Liang, L. L., Liu, X., & Fulmer, G. W. (Eds.). (2017). Chinese science
intrinsic motivation. Computers in Human Behavior, 55, education in the 21st century: Policy, practice, and research.
1201–1212. https://doi.org/10.1016/j.chb.2015.03.010. Dordrecht: Springer. https://doi.org/10.1007/978-94-017-9864-8
Chen, F., Shi, Y., & Xu, F. (2009). An analysis of the Public Scientific Lin, C. H., Liu, E. Z. F., Chen, Y. L., Liou, P. Y., Chang, M., Wu, C.
Literacy study in China. Public Understanding of Science, 18(5), H., et al. (2013). Game-based remedial instruction in mastery
607–616. https://doi.org/10.1177/0963662508093089. learning for upper-primary school students. Journal of Educa-
Chen, C.-H., Wang, K.-C., & Lin, Y.-H. (2015). The comparison of tional Technology & Society, 16(2), 271–281.
solitary and collaborative modes of game-based learning on Liu, E. Z. F., & Chen, P. K. (2013). The effect of game-based learning
students’ science learning and motivation. Educational Technol- on students’ learning performance in science learning-A case of
ogy & Society, 18(2), 237–248. ‘‘conveyance go’’. Procedia-Social and Behavioral Sciences,
Cheng, M. H. M., & Wan, Z. H. (2016). Unpacking the paradox of 103, 1044–1051. https://doi.org/10.1016/j.sbspro.2013.10.430.
Chinese science learners: Insights from research into Asian Lu, Y. L., & Lien, C. J. (2019). Are they learning or playing?
Chinese school students’ attitudes towards learning science, Students’ perception traits and their learning self-efficacy in a
science learning strategies, and scientific epistemological views. game-based learning environment. Journal of Educational
Studies in Science Education, 52(1), 29–62. https://doi.org/10. Computing Research. https://doi.org/10.1177/073563311
1080/03057267.2015.1112471. 8820684.
Cook, D. H. (2014). Conflicts in chemistry: The case of plastics, a McLaren, B. M., Adams, D. M., Mayer, R. E., & Forlizzi, J. (2017). A
role-playing game for high school chemistry students. Journal of computer-based game that promotes mathematics learning more
Chemical Education, 91(10), 1580–1586. https://doi.org/10. than a conventional approach. International Journal of Game-
1021/ed4007277.

123
 M. Wang, X. Zheng

Based Learning (IJGBL), 7(1), 36–56. https://doi.org/10.4018/ Uzun, L., Çetinavci, U. R., Korkmaz, S., & Salihoglu, U. M. (2013).
IJGBL.2017010103. Developing and applying a foreign language vocabulary learning
Meluso, A., Zheng, M., Spires, H. A., & Lester, J. (2012). Enhancing and practicing game: The effect of VocaWord. Digital Culture &
5th graders’ science content knowledge and self-efficacy through Education, 5(1), 50–70.
game-based learning. Computers & Education, 59(2), 497–504. von Gillem, S., & Alaswad, Z. (2016). Games and game-based
https://doi.org/10.1016/j.compedu.2011.12.019. learning in instructional design. International Journal of Tech-
Morris, T. A. (2011). Go chemistry: A card game to help students nologies in Learning, 23(4), 1–7. https://doi.org/10.18848/
learn chemical formulas. Journal of Chemical Education, 2327-0144/CGP/v23i04/1-7.
88(10), 1397–1399. https://doi.org/10.1021/ed100661c. Wang, S. Y., Chang, S. C., Hwang, G. J., & Chen, P. Y. (2018). A
Mustafa, J., Khan, A., & Ullah, A. (2011). Investigating student’s microworld-based role-playing game development approach to
achievement in mathematics through non technological game engaging students in interactive, enjoyable, and effective math-
based teaching. International Journal of Scientific Research in ematics learning. Interactive Learning Environments, 26(3),
Education, 4(3–4), 151–164. 411–423. https://doi.org/10.1080/10494820.2017.1337038.
Naik, N. (2014). A comparative evaluation of game-based learning: Yien, J. M., Hung, C. M., Hwang, G. J., & Lin, Y. C. (2011). A game-
Digital or non-digital games?. In European Conference on based learning approach to improving students’ learning
Games Based Learning (Vol. 2, p. 437). Academic Conferences achievements in a nutrition course. Turkish Online Journal of
International Limited. Educational Technology-TOJET, 10(2), 1–10.
Prensky, M. (2001). Digital game-based learning. New York: Yükseltürk, E., Altıok, S., & Başer, Z. (2018). Using game-based
McGraw-Hill. https://doi.org/10.1145/950566.950567. learning with kinect technology in foreign language education
Qian, M., & Clark, K. R. (2016). Game-based learning and 21st course. Journal of Educational Technology & Society, 21(3),
century skills: A review of recent research. Computers in Human 159–173.
Behavior, 63, 50–58. https://doi.org/10.1016/j.chb.2016.05.023. Yusuf, M. (2011). The impact of self-efficacy, achievement motiva-
Safaria, T., & Ahmad, A. (2013). Effects of self-efficacy on students tion, and self-regulated learning strategies on students’ academic
academic performance. Journal of Educational, Health and achievement. Procedia-Social and Behavioral Sciences, 15,
Community Psychology, 2(1), 22–29. https://doi.org/10. 2623–2626. https://doi.org/10.1016/j.sbspro.2011.04.158.
12928/jehcp.v2i1.3740. Zhao, Z., & Thomas, G. P. (2016). Mainland Chinese students
Sung, H. Y., & Hwang, G. J. (2013). A collaborative game-based conceptions of learning science: A phenomenographic study in
learning approach to improving students’ learning performance Hebei and Shandong Provinces. International Journal of Edu-
in science courses. Computers & Education, 63, 43–51. cational Research, 75, 76–87. https://doi.org/10.1016/
https://doi.org/10.1016/j.compedu.2012.11.019. j.ijer.2015.11.008.
Tan, P. H., Ling, S. W., & Ting, C. Y. (2007). Adaptive digital game-
based learning framework. In Proceedings of the 2nd interna- Publisher’s Note Springer Nature remains neutral with regard to
tional conference on Digital interactive media in entertainment jurisdictional claims in published maps and institutional affiliations.
and arts (pp. 142–146). ACM.
Tapscott, D. (1998). Growing Up Digital: The Rise of the Net
Generation. New York: McGraw-Hill.

123