Available online at www.sciencedirect.

com

ScienceDirect
Procedia - Social and Behavioral Sciences 93 (2013) 1994 1999

3rd World Conference on Learning, Teaching and Educational Leadership (WCLTA-2012)

The effects of peer scaffolding in problem-based gaming on

the frequency of double-loop learning and performance
in integrated science process skills
Raheleh Rahmani a *, Merza Abbas a, Ghasem Alahyarizadeh b
a

Centre for Instructional Technology & Multimedia, Universiti Sains Malaysia, 11800 Penang, Malaysia
b

NOR Lab, School of Physics, Universiti Sains Malaysia, 11800 Penang, Malaysia

Abstract
This study investigated the effects of peer scaffolding in inquiry-based games on the tendency to engage in double-loop learning
and performance in integrated science process skills. Double-loop learning involves deep levels of engagement and exploration
was hypothesized to enhance integrated science process skills among the players. Also, peer scaffolding was hypothesized to
promote double-loop learning. A physics-based video game was chosen and a quasi-experimental study involving two intact fifth
grade classes was employed. The findings showed that peer scaffolding significantly increased the engagement in double-loop
learning and significantly improved performance in integrated science process skills.

Authors.by
Published
Elsevier
Ltd. Open
under CCunder
BY-NC-ND
license.
2013
2012The
Published
ElsevierbyLtd.
Selection
and access
peer review
the responsibility
of Prof. Dr. Ferhan Odaba
Selection and peer review under responsibility of Prof. Dr. Ferhan Odaba
Keywords: Peer scaffolding; problem-based gaming; double-loop learning; integrated science process skills

1. Introduction
An important component of learning science is mastering science process skills (Martin et al, 2004; Mei et al.,
2007). Padilla (1990) defines science process skills as a set of broadly transferable abilities, appropriate to many
science disciplines and reflective of the behavior of scientists. Process skills are defined at two levels: basic skills
that comprise observation, communication, classification, measuring using numbers, inference, prediction, using
space and time relationship and integrated skills that include interpreting data, controlling variables, defining
operationally, formulating hypothesis and experimenting. The teaching of these skills is founded on the
interpretation of the Piagetian theory (Adey & Harlen, 1986) and employs the concepts of assimilation,
accommodation and schemata construction. Basic process skills are connected to assimilation where an event or
object fits into an existing scheme, while integrated process skills are related to accommodation in which the
existing scheme needs to be modified or recreated in order to account for the object or event. Based on Piagets
stages of cognitive development, basic skills can be fostered in the early stages of cognitive development, while
integrated skills must be introduced in the stage of formal operational development (King, 2011). Therefore, the

* Raheleh Rahmani. Tel.: +6-017-413-1412

E-mail address: rahelrahmani@gmail.com

1877-0428 2013 The Authors. Published by Elsevier Ltd. Open access under CC BY-NC-ND license.
Selection and peer review under responsibility of Prof. Dr. Ferhan Odaba
doi:10.1016/j.sbspro.2013.10.154

Raheleh Rahmani et al. / Procedia - Social and Behavioral Sciences 93 (2013) 1994 1999

1995

critical age to teach integrated process skills is around 11 years old, when the ability of higher order thinking begins
to emerge.
In Malaysia these skills are incorporated into both the primary and secondary school curricula (Lan et al., 2007)
and students are exposed to various aspects of science and inquiry-based learning in the classroom and the science
laboratories. However, follow-up studies in Malaysia and some other developing countries consistently reveal low
performance in tasks that require integrated science process skills (Gonzales et al., 2004, 2009; Martin et al, 2004).
These findings indicate the need to improve the teaching and learning of integrated process skills.
2. Problem-based gaming
Lawson (1995) proposes the use of the teacher-led learning cycle method to teach scientific thinking. Students
begin by exploring a new phenomenon to create disequilibria in their schema, and while the students are in a
cognitively conflicted state, the teacher offers tentative answers or procedures that students improve on by
generating and testing alternative solutions or new arguments to reestablish equilibrium. Collins and Stevens (1983)
offer a set of teacher-led inquiry-based learning strategies to force students into deep thinking and continuous review
of their knowledge and beliefs when learning science. These models and many others like them are founded on the
belief that scientific thinking skills are better taught by an expert and in a formal setting instead of acquired
independently at home. However, Padilla (1990) clarifies that science process skills cannot be developed by students
unless through inquiry in different contexts and content areas. Though, all aspects of inquiry from posing a
question to designing an investigation and experimenting are difficult for students and schools to conduct
because of practical and logistical constraints such as the lack of laboratory facilities and supplies to adopt inquiry
approaches (Honey & Hilton, 2011).Computer games provide an opportunity to achieve student-centered inquirybased learning with overcoming curricular and logistical obstacles (Honey & Hilton, 2011). Complex inquiry-based
computer games provide a new method to acquire knowledge and skills in a constructivist manner in which players
can examine their ideas and receive feedback on their hypotheses and strategies (Jong et al., 2010). A number of
studies have shown that computer games enhance childrens cognitive development (Buchanan, 2003; Plowman,
2005) and in particular improve their higher-order cognitive processes (Pillay, 2003; Ko, 2003). Thus, the US
Committee on Science Learning of the National Research Council (2011) acknowledged digital games as a worthy
resource that deserves future investment and investigation to improve science process skills (Honey & Hilton,
2011). Kiili (2007) proposes the problem-based gaming model (PBG) to illustrate the learning mechanisms involved
in playing inquiry-based games. In PBG, learning occurs in the game world as a cyclical process of direct
experience. The model suggests that the games can be played at the surface level through single-loop learning (SLL)
or at the deeper level of exploration and engagement through double-loop learning (DLL), with an important
conjunction of reflection. In this study, the PBG model is modified to reintegrate the pedagogical elements
suggested by Piaget (i.e., assimilation, accommodation, and schema construction) and Vygotsky (i.e., peer
scaffolding) to better explain and illustrate the mechanisms of cognitive engagement in playing complex games
(Figure 1.1).
In the modified PBG model, the player starts with exploring the challenge with an early set of playing strategies
through a simple active experiment or trials. Based on the results of the initial strategies the player performs an
assessment or reflection and modifies his schema through assimilation. In this way, the player can use the science
process skills that he/she already possesses to conquer the obstacles or challenges in the game and to form an early
schema for his/her knowledge about the game. The feedback and reflection processes determine the modification of
the schemata and consequently the players behavior. The reflection stage enables the player to decide whether to
continue the earlier scheme at the SLL level or to construct a new scheme through accommodation by changing the
strategies and move into the DLL level (Kiili, 2007).

1996

Raheleh Rahmani et al. / Procedia - Social and Behavioral Sciences 93 (2013) 1994 1999

Games

Exploration of the
challenge

Active
experiment
SLL

Schemata
construction

Reflection

DLL
Formulating and testing
of advanced strategies

Figure 1. Modified PBG model (from Kiili, 2007)

3. Peer scaffolding and Double Loop Learning in PBG

Peer scaffolding has been shown to enhance classroom learning. The term of peer scaffolding was introduced
by Wood et al. (1976) to describe the learning scenario in which a peer assists another learner to complete a task or
solve a problem that was unable to be accomplished alone. In instructional studies, peer scaffolding has been shown
to be a catalyst for driving deep processing (Wertsch & Resnick, 1991, Jeris, 1997). In the socio-cultural approach
of cognitive development suggested by Vygotsky (1989), interacting with others enables more efficient knowledge
construction by the learners (Shen & ONeil, 2008). Whitton (2009) argues that playing in pairs fosters peer learning
and promotes better understanding of the games interface. In addition, Schrier (2007) reported that playing in pairs
facilitates social interaction, information exchange, debate, discussion and sharing many resources among students.
Natriello & McDill (1986) also reported that peers influence educational expectations of individuals of the group.
Kiili (2005) postulated higher frequencies for reflection among members of groups or pairs. Thus, playing with
peers has the potential to facilitate deep or double-loop learning.
Currently no study has investigated whether students engaged in PBG with scaffolding in games of various
levels of difficulty are more likely to engage in DLL and subsequently enhance their integrated skills better than
students in PBG without scaffolding. Thus, following Wertsch & Resnick (1991) and Jeris (1997) it was
hypothesized that students in PBG with peer scaffolding would perform more DLL cycles and deeper reflection as
they advance through the various levels of the game and in turn improve their performance in integrated science
process skills while students in PBG without scaffolding would perform more SLL and score lower in integrated
science process skills. Also, following research of Rebetez et al. (2005), Schrier (2007), Kiili (2007) and Payne
(2010), we hypothesized that the DLL users in both peer scaffolding and individual groups would perform better in
integrated science process skills than the SLL users.
4. Method
A quasi-experimental study involving two intact classes was employed. The sample comprised 60 male and
female fifth grade pupils. One class was assigned to play the game individually and the other in pairs, with pupils
having the choice of partner. Intact classes and voluntary pairing were maintained to retain the normal functioning
of classes and prevent inadvertent introduction of other variables that could confound the outcomes of the treatment.
A verification of the students achievement in science revealed significant differences between the classes. Thus, the
students science achievement scores were used as covariate. The independent variable of this study was peer

Raheleh Rahmani et al. / Procedia - Social and Behavioral Sciences 93 (2013) 1994 1999

1997

scaffolding, and the learners intensity of reflection, levels of engagement in the game in the form of SLL and DLL,
and performance in integrated science process skills were the dependent variables.
The instruments consisted of the Ohio achievement assessments of science (2007) for grade 5 for evaluating
student performance in integrated process skills, a game-playing log to record reward points for each level, and a
reflection questionnaire. The Ohio test had a reported reliability index of = 0.86 (AIR Technical Team, 2008).
Some articles in a few items such as types of animals and plants were changed to fit the Malaysian context with the
assistance of a science teacher. The students were required to complete 35 levels of a physics-based game called
Crazy Machines Elements without the teacher imposing on how deeply each level was to be completed. 12 of the
levels required SLL processing while 23 levels required DLL processing with a total of 229 nuts or reward points to
be collected. These levels were independent of each other, so students could complete each of them in full or
partially before moving on to any of the next levels. As they played the game, the students were required to log their
reward points collected for each level in the form provided (Game-playing Log). Reward points of 175 or lower
would indicate an overall SLL level of processing while scores of over 175 would indicate DLL level of processing.
The reflection questionnaire consisted of Likert-scale response items for the pupils to report on the intensity of
reflection while engaging in the game and had a reliability of = 0.71. The students were also given all the time
they needed to play the game and completed the required levels in 10.5 hours. The level of significance was set at p
= 0.05.
5. Results and discussion
5.1. ANCOVA was employed to test the effects of the treatment on performance in integrated science process skills
and ANOVA tests were employed to test the effects of the treatment on the reward points earned and reflection. A set
of primary analyses was carried out to check the assumptions of ANOVA and ANCOVA, namely, normality,
linearity, homogeneity of regression slopes and independence of covariate and treatment, and all these assumptions
were met.Performance by treatment
Table 1 presents the mean scores, standard deviations, and results of the tests for hypotheses 1, 2, and 3 concerning
reward points earned, reflection and integrated science process skills by treatment groups.
Table 1. Mean scores, standard deviations and p values for reward points, reflection and integrated process skills by treatment
Reward points
Treatment
N
M
SD
ANOVA
Peer scaffolding 34 187.26 15.89 F(1,59) = 4.078
p = .048
Individual
26 178.58 17.30

N
31
26

M
32.23
30.35

Reflection
SD
ANOVA
2.46 F(1,56) =5.593
p = .022
3.52

Integrated science process skills

N
M
SD
ANCOVA
34 15.09 3.93 F(1,59) = 1.318
p = .256
26 16.65 4.19

Hypothesis 1: Students in the peer scaffolding group will report significantly higher engagement in DLL
compared to students in the individual group. The One-way ANOVA test reported F(1, 59) = 4.078 at p = .048. As p
< 0.05, the hypothesis was accepted. Both groups reported means greater than 175, indicating that both groups were
engaged at the DLL level of processing but the peer scaffolding group reported significantly higher means in DLL.
This finding is consistent with previous research that reports that peer scaffolding enhances deep processing
(Wertsch & Resnick, 1991) and better understanding of the games interface (Whitton, 2009). The finding of this
study shows that peer scaffolding in PBG enhances double-loop learning.
Hypothesis 2: Students in the peer scaffolding group will report significantly higher reflection scores compared
to students in the individual group. The ANOVA test results shown in Table 1 reports F(1, 56) = 5.593 at p = .022.
As p < 0.05, the hypothesis was accepted. This finding is in good agreement with other studies. Schrier (2007)
reported that playing in pairs or groups encourages the players to reflect deeply on the information gathered during
the game and the content of the game is understood better. Also, Payne (2010) reported that children in cooperative

1998

Raheleh Rahmani et al. / Procedia - Social and Behavioral Sciences 93 (2013) 1994 1999

learning groups reflect better on what they did. The finding of this study confirms the positive effect of peer
scaffolding on reflection among the players.
Hypothesis 3: Students in the peer scaffolding group will report significantly better performance in integrated
science process skills compared to students in the individual group. The results of the ANCOVA test revealed that
F(1, 59) = 1.318 at p = .256 (Table 1). As p > 0.05, the hypothesis was rejected indicating that there was no
significant difference between the peer scaffolding and the individual groups on performance in integrated science
process skills. This finding was obtained because the duration for this variable to take effect was too short, i.e. only
10.5 hours. A longer time would undoubtedly produce different results.
5.2. Performance by levels of engagement
Hypothesis 4: For both the peer scaffolding and individual groups, DLL users will perform significantly better than
SLL users in integrated process skills. A further analysis was conducted to investigate the effects of engaging in
SLL and DLL on the integrated science process skills scores. The data was split by treatment groups and the reward
point scores were recoded into SLL or DLL levels of engagement. Table 2 shows the mean scores, standard
deviations, and results of ANCOVA for integrated process skills by the levels of engagement in the game for both
the peer scaffolding and individual groups. The data demonstrate that more students were engaged in DLL in the
peer scaffolding group (79.4%) than in the individual group (65.4%). The performance in integrated skills between
students employing DLL and SLL was also different for each group. For the peer scaffolding group, students
engaged in DLL reported a significantly higher mean in integrated process skills than their counterparts engaged in
SLL (F(1, 33) = 8.698 at p = .006) but for the individual group there was no significant difference between those
engaging in DLL and SLL (F(1, 25) = 1.229 at p = .279). These results can be explained from the view that peer
scaffolding increases the likelihood of engaging in DLL also results in deeper reflection and processing when
engaging in the tasks. Students in the individual group who were engaged in DLL did not process deep enough to
differentiate them from students in SLL. These findings are consistent with Wertsch (1991), Jeris (1997), and
Rebetez, et al. who reported that peer scaffolding fosters deeper levels of cognitive processing. These findings show
that that peer scaffolding fosters deeper engagement in DLL and as a result improved student performance in the
integrated science process skills.
Table 2. Mean Scores, standard deviations and p values on integrated skills for two groups by level of engagement in DLL and SLL

Level
DLL
SLL

Integrated skills for peer scaffolding group

N
M
SD
ANCOVA
27
16.07
3.38
F(1,33) =8.698
p = .006
7
11.29
3.77

N
17
9

Integrated skills for individual group

M
SD
ANCOVA
17.53
3.61
F(1,25) =1.229
p = .279
15.00
4.92

6. Conclusion
The study found that in the modified PBG model, peer scaffolding increases the likelihood of engaging in doubleloop learning and enhances deeper reflection and processing that in turn improves performance in integrated science
process skills. From these findings it can also be concluded that the modified PBG model is a suitable framework to
explain the processes of learning involving student-centered inquiry-based learning through complex video games.
However, this study was conducted using a very small sample, thus caution is in order in generalizing the findings
and more studies are recommended.

Raheleh Rahmani et al. / Procedia - Social and Behavioral Sciences 93 (2013) 1994 1999

1999

R eferences
Adey, P. S., Harlen, W. (1986). A Piagetian analysis of process skill test items. Journal of Research in Science Teaching, 23, 707726.
Buchanan, K. (2003). Opportunity knocking: Co-opting & games. ALT-N, 43, 10-11.
AIR Technical Team. (2008). Validity evidence based on internal structure: Examination of the factor structure of the Ohio Achievement Tests.
Retrieved October 18, 2011, from http://www.ode.state.oh.us.
Collins, A., & Stevens, A. L. (1983). A Cognitive Theory of Inquiry Teaching In , C. M. Reigeluth (Ed.).(1983). Instructional-design theories
and models: An overview of their current status. Hillsdale: Lawrence Erlbaum Associates.
Gonzales, P., Guzmn, J.C., Partelow, L., Pahlke, E., Jocelyn, L., Kastberg, D., & Williams, T. (2004). Highlights From the Trends in
International Mathematics and Science Study (TIMSS) 2003 (NCES 2005005). U.S. Department of Education, National Center for
Education Statistics. Washington, DC: U.S. Government Printing Office.
Gonzales, P., Williams, T., Jocelyn, L., Roey, S., Kastberg, D., & Brenwald, S. (2009). Highlights from TIMSS 2007. Science, (September), 112.
U.S. Department of Education.
Honey, M. A., and Hilton, M. (2011). Learning Science Through Computer Games and Simulations. M. A. Honey and M. Hilton, eds. National
Academies Press Available at: http://www.nap.edu/catalog.php?record_id=13078.
Jeris L. (1997). An Empirical study of Relationship Between Team Process Interventions and Double-Loop Learning. Doctoral dissertation,
Department of Leadership and Educational Policy Studies, Northen Illinois University.
Jong, M. S., Shang, J., & Lee, F. (2010). Constructivist Learning Through Computer Gaming. In M. Syed (Ed.), Technologies Shaping
Instruction and Distance Education: New Studies and Utilizations (pp. 207-222).
Kiili, K. (2005). Educational Game Design: Experiential Gaming model revised. Tampere University of Technology, Pori, Research report 3, 3-6.
Kiili, K. (2005).
Kiili, K. (2007). Foundation for problem-based gaming. British Journal of Educational Technology, 38(3), 394-404.
King K.P. (2011). Working with Young People: Intellectual Development over Time. Retrieved on 2012 February 1 from
sites.roosevelt.edu/kking/files/2011/11/Development-and-STEM-Education.pdf
Ko, S. (2002). An empirical analysis of childrens thinking and learning in a computer game context. Educational Psychology, 22(2), 219-233.
Lan, O.S, Ismail Z., Fook, F.S. (2007). Assessing Competency In Integrated Science Process Skill And Its Relation With Science Achievement.
Project Report. Universiti Sains Malaysia.
Lawson, A. E. (1995). Science teaching and the development of thinking. Belmont: California.
Martin, M., Mullis, I., Gonzales, E., and Chrostowski, S. (2004). TIMSS 2003 International Science Report. Boston: IEA and Insternational
Study Center, Boston College.
Mei G.T.Y., Kaling Ch., Xinyi Ch. S., Sing J.S.K., Khoon K.N.S. (2007). Promoting science process skills and the relevance of science through
science alive. Proceedings of the Redesigning Pedagogy: Culture, Knowledge and Understanding Conference, Singapore, May 2007.
Natriello, G., & McDill, E. L. (1986). Performance standards, student effort on homework, and academic achievement. Sociology of Education,
59(1), 1831. American Sociological Association. Retrieved from http://www.jstor.org/ stable/2112483.
Ohio Department of Education. (2007). Ohio Achievement Tests Administration Statistical Summary. Retrieved
from
http://education.ohio.gov/GD/Templates/ Pages/ODE/ODEDetail.aspx?page=3&TopicRelationID=285&ContentID=9479&Content=117721
Padilla, M. J. (1990). The science process skills. Research Matters... To the Science Teacher. National Association for Research in Science
Teaching, No. 9004.
Payne, C.R. (2010). Information technology and constructivism in higher education: Progressive learning frameworks. Hersey, PA: IGI Global.
Pillay, H. (2003). An investigation of cognitive processes engaged in by recreational computer game players: Implications for skills of the future.
Journal of Research on Computing in Education, 34(3).
Plowman, L., & Stephen, C. (2005). Children, play, and computers in pre-school education. British Journal of Educational Technology, 36(2),
145-157.
Rebetez C., Btrancourt M., Sangin M., Dillenbourg P. (2005). Collaborer pour mieux apprendre d'une animation, Proceedings of the 17th
international conference on Francophone sur l'Interaction Homme-Machine, p.171-176, September 27-30, 2005, Toulouse, France
Schrier, K. (2007). Reliving the Revolution: Designing Augmented Reality Games to Teach Critical Thinking. In D. Gibson, C. Aldrich, & M.
Prensky (Eds.), Games and Simulations in Online Learning: Research and Development Frameworks (pp. 250-270). Hershey, PA:
Information Science Publishing.
Shen Ch.Y. & ONeil H. F. (2008). Role of worked examples to stimulate learning in a game. In H. F. ONeil & R. S. Perez (Eds.), Computer
games and team and individual learning (pp. 185-204). Oxford, UK: Elsevier.
Vygotsky, L. S. (1989). Concrete human psychology. Soviet Psychology, 27(2), 5377.
Wertsch, J., Resnick, L. B., Levine, J. M., & Teasley, S. D. (1991). A sociocultural approach to socially shared cognition. In L B Resnick, J M
Levine, & S D Teasley (Eds.), Perspectives on socially shared cognition (pp. 85-100). American Psychological Association.
Whitton, N. (2009). Learning with Digital Games: A Practical Guide to Engage Students in Higher Education (The Open and Flexible Learning
Series). New York: Routledge.
Wood, D., Bruner, J., & Ross, G. (1976). The role of tutoring in problem solving. Journal of Child Psychology and Psychiatry, 17, 89-100.