Computers & Education 82 (2015) 84e95

Contents lists available at ScienceDirect

Computers & Education

journal homepage: www.elsevier.com/locate/compedu

Social gaming is inSIDE: Impact of anonymity and group identity on

performance in a team game-based learning environment

naff, Nicolas Michinov*, Olivier Le Bohec, Marine Delaval
Benjamin Le He
Universit

e Rennes 2, CRPCC EA n
1285, France

a r t i c l e i n f o a b s t r a c t

Article history: The present experiment aimed to determine how quiz performance in a team game-based learning
Received 7 July 2014 environment can be predicted from the Social Identity model of Deindividuation Effects (SIDE). Ac-
Received in revised form cording to this model, anonymity influences social behavior by accentuating the salience of group
30 October 2014
identity and reducing interpersonal differences, leading to greater group identification and motivation to
Accepted 3 November 2014
work for one's own group. As these effects could lead to higher cognitive performance, the goal of the
Available online 14 November 2014
present research was to extend predictions based on the SIDE model on performance in online game-
based learning environments. After measuring their prior computing knowledge, 343 Master Degree
Keywords:
Interactive learning environments students were placed in virtual teams on a trivial criterion to perform a series of online quizzes about
Distance education and telelearning computing and the Internet. An anonymous (or individuated) username was attributed to each team
Anonymity member to connect to the online learning environment, and information about comparison between
Intergroup comparison teams was used to manipulate the degree of salience of group identity (high versus low). As predicted by
Social gaming the SIDE model, anonymity boosted performance when group identity was salient, but only for students
with low prior knowledge. Unexpectedly, it was also found that anonymity boosted the performance of
students with high prior knowledge when group identity was not salient. A similar pattern was found for
perceived mastery of computing and the Internet. Theoretical and practical implications of the SIDE
model are discussed, and specifically its application to social gaming to optimize online learning.
© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

One of the challenges of the 21st century is the growing need to help students develop skills and knowledge in computing and the
Internet. This need is also crucial to prepare students to become teachers and use computers and other technologies efficiently inside and
outside the classroom. However, students preparing to be future teachers, as well as teachers themselves, often have a lack of knowledge
and skills in computing and the Internet. Indeed, a National Report on innovation in teaching practices using digital environments and their
implications for teacher training revealed that only 37% of French teachers indicate being comfortable with information and communication
technologies (ICT) (Fourgous, 2012). An IPSOS survey (2012) with a sample of 404 French teachers, questioned via the Internet about their
teaching practices, revealed that 27% rarely or never used ICT in the classroom, and 27% of those reported a lack of training in this domain.
Similarly, a study showed that only 28% of Canadian pre-service teachers felt comfortable using ICT after a Teacher Education Program
(Martinovic & Zhang, 2012). A previous European survey conducted among teachers in 27 European countries found that 66% of re-
spondents said they had used computers with their students in the classroom during the previous year, but 62% of them said they used them
rarely, i.e., in less than 25% of their classes (Empirica, 2006). It is therefore important to find solutions to engage future teachers, and students
in general, to acquire computing knowledge, and to develop skills to use ICT efficiently in the classroom. As our purpose was to increase
performance and motivation to acquire computing knowledge, we examined the role of social and psychological processes, such as social
comparison theory (Festinger, 1954) and its extensions to social identity theory (e.g., Abrams & Hogg, 1990; Tajfel, 1974, 1978; Tajfel & Turner,
1986) and self-categorization theory (e.g., Oakes, 1987; Oakes, Haslam, & Turner, 1994; Turner, 1982, 1987; Turner & Haslam, 2001; Turner,

Rennes 2, CRPCC EA 1285, Place du Recteur Henri Le Moal, CS 24307, 35043 Rennes Cedex, France. Tel.: þ33 2 99 14 19 55.
* Corresponding author. Universite

naff), nicolas.michinov@univ-rennes2.fr (N. Michinov).
E-mail addresses: benjamin.lehenaff@univ-rennes2.fr (B. Le He

http://dx.doi.org/10.1016/j.compedu.2014.11.002
0360-1315/© 2014 Elsevier Ltd. All rights reserved.
 B. Le H
enaff et al. / Computers & Education 82 (2015) 84e95 85

Oakes, Haslam, & McGarty, 1994; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987). We postulated that these processes could be used as
powerful social incentives for acquiring computing and Internet skills and knowledge.
More specifically, we based our study on a model derived from social identity theories: the Social Identity model of Deindividuation
Effects (SIDE) (Reicher, Spears, & Postmes, 1995; Spears & Lea, 1992, 1994), which uses social identity and anonymity to bolster the iden-
tification to team, leading to higher implication for the sake of one's team. A common aspect of these social identity theories is that
intergroup comparisons have a crucial impact on various attitudes and behaviors in social life, but only a few studies have examined their
impact on cognitive performance and, to our knowledge, no study have been conducted in game-based learning environments.
Despite the great number of video games and serious games that have been developed for educational purposes (e.g., Arena & Schwartz,
2013; Coffman & Klinger, 2008; Eschenbrenner, Nah, & Siau, 2008; Squire & Barab, 2004; Warburton, 2009; see Young et al., 2012, for a
review), only a small number are based on social psychology theories (Baldwin & Dandeneau, 2009). Instead, they are often designed
intuitively from social processes such as competition to reinforce their attractiveness, and some use quizzes where teams compete against
each other to attain learning objectives (Goldberg & Dintzis, 2007). More recently, it has been shown that giving students quizzes with
formative feedback may increase self-evaluation, engagement, and exam performance (Balter, Enstrom, & Lingenberg, 2013). Similarly,
delivering performance feedback based on social comparison processes may have similar beneficial effects for self-evaluation in learning
environments. Such feedback can be delivered automatically and anonymously in the classroom using clicker devices (e.g., Fies & Marshal,
2006; Lantz & Stawiski, 2014), or remotely in digital game-based learning environments (e.g., Cameron & Dwyer, 2005; Erhel & Jamet, 2013;
Jackson & McNamara, 2013; Ma, Oikonomou, & Jain, 2011; Wouters, van Nimwegen, van Oostendorp, & van der Spek, 2013). Consequently,
performance feedback were used in the present study to test their impact on quiz performance in a team game-based learning environment
aiming to motivate students to acquire computing and Internet knowledge.

1.1. Theoretical foundations of social identity theories

According to social identity theory (Tajfel, 1974, 1978; Tajfel & Turner, 1986; see Postmes & Branscombe, 2010, for a review), an in-
dividual's social identity is determined by the group to which he/she belongs. When individuals feel that they belong to a given group, they
tend to establish comparisons between “them” and “us”, in order to maintain an advantage of their group over other groups. It has been
demonstrated that the mere categorization of people into distinct groups, even when based on trivial criteria, is sufficient to create a social
identity for them (e.g., Tajfel, Billig, Bundy, & Flament, 1971).
Self-categorization theory (Turner, 1982; Turner et al.,1987) expands the idea that social identity processes are fundamental to understand
group behaviors by distinguishing between social identity (self-definition in terms of social category membership) and personal identity (self-
definition in terms of personal or idiosyncratic attributes). According to this theory, group identification leads individuals to perceive
themselves in terms of the characteristics they share with other members of their in-group (their group or social identity) rather than in terms
of the idiosyncratic characteristics that differentiate them from other individuals (their personal identity; Turner et al., 1987).
The Social Identity model of Deindividuation Effects (Reicher et al., 1995; Spears & Lea, 1992, 1994), which lies at the heart of the present
study, extends previous theories and models of Computer-Mediated Communication (see Footnote1), such as media richness theory (Daft &
Lengel, 1984) and reduced social cues model (Kiesler, Siegel, & McGuire, 1984), by demonstrating that certain features of Internet
communication, such as (visual) anonymity, may accentuate (and not reduce) the salience of possible social identities. Consequently, less
attention is devoted to differences within the group and more attention is paid to similarities. In other words, according to the SIDE model,
anonymity enhances the salience of social identity originating from intergroup comparisons and may have an influence on group members'
attitudes and behaviors. Unfortunately, only a few studies to date have examined the impact of social identity and intergroup comparisons
on cognitive performance (see Michinov, Michinov, & Toczek-Capelle, 2004; Tanis & Postmes, 2008).

1.2. Effects of intergroup comparison on performance

Social identity theories have always focused mainly on intergroup relations and favoritism toward one's group (e.g., Abrams & Hogg,
1990; Tajfel & Turner, 1986). Indeed, intergroup comparisons have been identified as having a crucial influence on various attitudes and
behaviors, and more particularly on self-esteem (Lemyre & Smith, 1985; Marmarosh & Corazzini, 1997; Redersdorff & Martinot, 2009;
Wright & Forsyth, 1997) and discriminatory behaviors (e.g., Cadinu & Reggiori, 2002; Locksley, Ortiz, & Hepburn, 1980; Tarrant, 2002).
However, to our knowledge, few studies have examined the impact of intergroup comparison on performance (e.g., James & Greenberg,
1989; van Knippenberg, 2000; Ouwerkerk, Ellemers, & de Gilder, 1999; Pilegge & Holtz, 1997; Turner & Haslam, 2001; Worchel,
Rothgerber, Day, Hart, & Butemeyer, 1998), even though it has been recognized that intergroup comparison increases people's motiva-
tion to work on behalf of their in-group (James & Greenberg, 1989) and reduces social loafing, i.e. the tendency to reduce one's effort in a
group when one's own contribution is not identifiable (Huguet, Charbonnier, & Monteil, 1999; Karau & Williams, 1993). For example,
Worchel et al. (1998) observed that various conditions enhancing commitment to group work (future interaction, interdependence, and
wearing a uniform in the presence of an out-group member) can eliminate group members' tendency to loaf. They demonstrated that when
group salience is reinforced by contextual cues such as wearing a uniform or having a group name, intergroup comparison tends to increase
manual and cognitive task performance. In a gaming context, Farzan, Dabbish, Kraut, and Postmes (2011) manipulated the online visual
representation of teams and individuals in a non-collaborative game such as Tetris®. They showed that the mere presence of others, and
probably comparison with other teams and/or individuals, increased the number of game play sessions. In other words, persistence in the
game was higher when participants were assigned to teams than when they played individually. This effect was obtained irrespective of
experimentally induced social conditions: “within” (i.e. gamers tried to score higher than their teammates), “between” (i.e. gamers tried to
help their team win), or “within/between” (i.e. gamers tried to score higher than their teammates and to help their team win). Although

1
Computer-Mediated Communication is defined as any exchange of information (either factual, intellectual, or emotional) occurring through the use of two or more
networked computers (McQuail, 2005).
86 B. Le H
enaff et al. / Computers & Education 82 (2015) 84e95

social comparison processes were not mentioned in their paper, it is reasonable to suppose that they were at work during the gaming
sessions. Other studies based on the SIDE model found that anonymity of group members may reinforce the impact of intergroup com-
parisons, although very few studies to date have examined its impact on performance.

1.3. Effects of anonymity versus individuation on performance

According to the SIDE model (Reicher et al., 1995; Spears & Lea, 1992, 1994), when individuals are anonymous within their group, they
tend not to focus their attention on the differences between group members, preferring to center their attention on the ingroup as a whole in
relation to other groups. In other words, as interpersonal clues are rendered less accessible by an anonymous intergroup context, ingroup
members tend to increase their identification with the group. This model has been tested over the last two decades, several experiments
showing that the joint effect of intergroup comparison and anonymity leads to attraction to the ingroup (e.g., Kim, 2009; Kim & Park, 2011;
Lea & Spears, 1991, 1992; Lea, Spears, & de Groot, 2001; Lea, Spears, Watt, & Rogers, 2000; Michinov et al., 2004; Michinov & Monteil, 2003;
Postmes, 1997; Postmes, Spears, & Lea, 1998, 2002; Rogers & Lea, 2005), to stereotypic perception of the exogroup (e.g., Lea et al., 2001;
Postmes, 1997), to extremization of opinions (e.g., Postmes, 1997; Spears, Lea, & Lee, 1990), to cheating behavior in online games (Chen
& Wub, 2013), and so forth. These findings indicate that group norm adhesion and social influence processes are reinforced in anony-
mous intergroup comparison settings (e.g., Lea & Spears, 1991; Postmes, 1997; Postmes, Spears, & Lea, 2000; Postmes, Spears, Sakhel, & de
Groot, 2001), including the isolation and visual anonymity of Computer-Mediated Communication (e.g., Spears, Lea, Postmes, & Wolbert,
2011; Spears & Postmes, 2014). Thus, it is not surprising that a great number of studies based on the SIDE model have been conducted
in online environments where people may remain visually anonymous (e.g., Gue
gan & Michinov, 2011; Postmes, 1997; Postmes et al., 2002).
In Computer-Mediated Communication, it has generally been shown that anonymous participants identify more strongly with their group
than individuated participants.
More important for our purpose, other studies have tested the SIDE model in educational settings, including online collaborative
(learning) environments. For example, Michinov et al. (2004) showed that group identification of pupils collaborating remotely and syn-
chronously in online three-member groups was stronger when they were anonymous than when they were individuated. Similarly, Ren,
Kraut, and Kiesler (2007) and Ren et al. (2012) customized online communities so that users would either be individually identifiable, or
would be identifiable only as team members. They showed that individuals belonging to online communities with an identity as group
members developed a stronger sense of community and commitment despite the virtual dispersion of the members. However, the effects of
anonymity on performance remain little investigated to date (see Tanis & Postmes, 2008). In a study where students had to complete a
project using a computer-supported collaborative learning system (Lea, Rogers, & Postmes, 2002), it was found that collaboration between
remote students was facilitated under an anonymity condition, in other words, when individuating features of group members were
concealed in the user interface. Another study demonstrated that collaboration between students was more successful when group
members were anonymous than when they were individuated in a dyad (Tanis & Postmes, 2008). More specifically, the authors found that
the absence of personal identity cues led to better subjective (study 1) and objective (study 2) performance in dyads than when these cues
were present, and this effect was explained by greater group identification.
Despite a large number of studies verifying the predictions about the cognitive and affective effects of the SIDE model, including in online
environments, there have been few studies about the joint impact of anonymity and salience of group identity on cognitive task perfor-
mance. Because cognitive performance may be influenced by participants' prior knowledge (see Dochy, Segers, & Buehl, 1999, for a review),
it needs to be controlled to examine the impact on performance of the two variables at the core of the SIDE model (anonymity and salience of
group identity).

1.4. The role of students' prior knowledge

As far as we know, no study based on the SIDE model has examined its impact on performance taking into account students' prior
knowledge. Nevertheless, studies conducted in classroom settings before the development of the SIDE model may be relevant to support the
present study (Monteil, 1988, 1993; Monteil & Huguet, 1993, 2001). Although these studies were not based on the SIDE model per se, they
examined the joint effects of anonymity and intergroup comparison on academic performance among children. For example, Monteil (1988,
study 1) manipulated the intergroup comparison and anonymity of pupils during a standardized biology lesson before their prior
knowledge had been assessed from their previous performance. Under the intergroup comparison condition, information was given about
the level of knowledge of each group of pupils, whereas the academic level of each group was not mentioned in the other experimental
condition. Half of the pupils in each group were also informed that no-one would be questioned during the lesson (anonymity condition),
whereas the other half were informed that each of them would be questioned during the lesson (individuation condition). In fact, no pupil
was questioned during the lesson in either condition, and performance was assessed at the end of the lesson. Results found that when the
intergroup comparison was salient, the performance of pupils with low prior knowledge was better under the anonymity than the indi-
viduation condition, and similar to students with high prior knowledge in the anonymity condition. No difference was observed between
the anonymity and individuation conditions when intergroup comparison was not salient: pupils with high prior knowledge performed
better than those with low prior knowledge, in line with their usual performance.
The aim of the present study was to extend those findings, examining the effects of both anonymity and salience of group identity on
performance in a team game-based learning environment designed to teach computing and Internet knowledge, taking into account the
students' prior knowledge.

1.5. Overview and hypotheses

Participants agreed to test an online learning environment based on gaming, working in virtual teams. After assessing their prior
knowledge of computing and the Internet through a preliminary standardized online test, a fictitious task was administered to randomly
assign them to one of the two virtual teams (Hashtag or Arobase) using the same procedure as the minimal group paradigm (Tajfel et al.,
 B. Le H
enaff et al. / Computers & Education 82 (2015) 84e95 87

1971). They were randomly distributed to four conditions, involving combinations of two variables: salience of group identity (high vs. low)
and anonymity (anonymity vs. individuation). The first variable was manipulated using intergroup comparison feedback given (or not)
during a quiz task. The second variable was manipulated by the username attributed to the participant. After performing the task over a 12-
day period, participants were invited to complete a web-questionnaire incorporating manipulation checks, and an additional measure of
perceived mastery in computing and the Internet.
Based on the Social Identity model of Deindividuation Effects (Reicher et al., 1995; Spears & Lea, 1992, 1994), as well as previous studies
performed in classroom settings (Monteil, 1988; Monteil & Huguet, 1993), it was expected that when group identity was rendered salient by
intergroup comparison feedback, students with low prior knowledge would achieve higher scores on the quiz when they were anonymous
than when they were individuated. Because anonymity helps enhance the salience of group identity, participants may be more engaged in
their group, leading them to make greater efforts to help their group succeed, without any risk for their self-esteem. Conversely, students
with high prior knowledge were expected to perform better when they were individuated than when they were anonymous.

2. Method

2.1. Participants and experimental design

Participants were 343 Master Degree students (257 women and 86 men) aged 22e47 years (M ¼ 24.62; SD ¼ 2.78) at the University of
Rennes (Brittany, France). All came from various disciplines (literature, arts, humanities, and social sciences), and were training to become
teachers in primary and secondary schools. They volunteered to test a new online learning environment based on social gaming for 12 days in
order to acquire the computing and Internet knowledge that they would need in their teaching practice. A web-questionnaire was completed
by 59.47% of the participants (N ¼ 204) at the end of the session. Before the study, participants were divided into 40 teams of 15 students, with
the same proportion of males and females (one-third and two-thirds respectively). Each team was programmed to play against one other team
throughout the experiment, thus forming 20 sessions. Participants were randomly assigned to one of the four experimental conditions with a
2 (Anonymity versus Individuation)  2 (High versus Low Salience of group identity) between-participants design.

2.2. Apparatus and materials

A team game-based learning environment (UL-TE@MS/Q2i) was set up for the purpose of the present study. This entirely computerized
environment was accessible online from the web, and offered students the opportunity to acquire computing and Internet knowledge by
enabling them to “play” a series of quizzes in virtual teams. The interface was divided into three distinct parts (see Appendix): a chart
showing the results of each team; a list of all participants' usernames with their individual scores presented from top down (leaderboard
with the highest score at the top); and a quiz with radio buttons to select the correct statement among three alternatives. Only one
statement was correct, the other was false, and the third offered the option of not answering.

2.3. Procedure

On the pretext of testing a new online learning system based on social gaming, participants were informed by e-mail that they had the
opportunity to use a team game-based learning environment for a limited period. They were informed that the online learning environment
was designed to help them acquire computing and Internet knowledge. There was an attachment with a hyperlink to a web page, together
with a code number randomly assigned to each participant (from 0001 to 0500). Each code number was matched in a database with a
predetermined anonymous or individuated username corresponding to one of the experimental conditions to which participants were
assigned. Once participants had agreed to participate by clicking on a button after reading a consent form, they entered their code number,
and had to perform two preliminary tasks at their first connection. The first task was used to allocate the participant to a virtual team and to
give them a username. The second task was administered to evaluate their prior knowledge in the computing and Internet domain.

2.3.1. Assignation task: team membership and attribution of usernames

To allocate participants to a virtual team, a bogus estimation task was administered following the “minimal group” paradigm procedure
(Tajfel et al., 1971). It involved estimating the number of two computer symbols in six images which were successively displayed on the
screen. The choice of the symbols (@, Arobase, and #, Hashtag) resulted from a pretest (see Footnote2). A great number of @ and # symbols
were presented on each image during 4 s, and it was impossible to decide with certainty how many symbols of each type there were. After
each image, participants had to indicate their answer on a multiple-choice questionnaire. At the end of this task, all participants were
instructed that the computer would analyze their responses and allocate their group membership and username. In reality, all participants

2
The pretest was administered to 134 undergraduate students (95 females and 39 males) at the beginning of a social psychology lecture in several disciplines of hu-
manities and social sciences, language, and literature, but not psychology students. On the pretext of a study about the impressions formed by the usernames that individuals
use on the Internet, participants had to indicate their impression of a person whose username began with one of four symbols (@, #, $, or x), by indicating on a 7-point Likert
scale from 1 (not at all) to 7 (totally) to what extent the username defined this person on 18 statements organized around three factors: Sociability (e.g., this person has team
spirit, this person helps others), Computing/Science/Technology (e.g., this person has computing skills), and Social Networks (e.g., this person frequently uses social networks
such as Facebook). Four versions of the questionnaire were drawn up, varying only the symbol at the beginning of the same username (@_OHL.451, #_OHL.451, $_OHL.451, or
§_OHL.451). A KruskaleWallis non-parametric test was conducted, showing that the same anonymous username was perceived as significantly different depending on the
symbol used on each factor: Sociability, Computing/Science/Technology, and Social networks (c2 ¼ 8.67, df ¼ 3, p < .034, c2 ¼ 11.06, df ¼ 3, p < .011, and c2 ¼ 9.57, df ¼ 3, p < .
023, respectively). More specifically, results showed that usernames with @ and # were not perceived as significantly different on the three dimensions. For Sociability,
usernames with @ (M ¼ 2.70/Mean rank ¼ 30.75) did not differ from the same username with # (M ¼ 3.00/Mean rank ¼ 37.77; ManneWhitney test, Z ¼ 1.47, N1 ¼ 36,
N2 ¼ 31, p ¼ 0.14), and similarly for Computing/Science/Technology (@, M ¼ 5.75/Mean rank ¼ 33.00 and #, M ¼ 5.92/Mean rank ¼ 35.16; ManneWhitney test, Z ¼ 0.45,
N1 ¼ 36, N2 ¼ 31, p ¼ 0.65) and for Social networks (@, M ¼ 4.39/Mean rank ¼ 32.53 and #, M ¼ 4.61/Mean rank ¼ 35.71; ManneWhitney test, Z ¼ 0.67, N1 ¼ 36, N2 ¼ 31,
p ¼ 0.50). Because the symbols @ and # did not give rise to a difference in perception, they were chosen to assign participants to distinctive teams.
88 B. Le H
enaff et al. / Computers & Education 82 (2015) 84e95

received bogus feedback to assign them to a minimalistic group. After a 5-s delay, the computer informed participants that they were a
member of either the “@ team” or the “# team”, on the pretext that they were better estimators for a given symbol. Half the participants
were randomly assigned to one team, and the other half to the other using a round-robin procedure.
Simultaneously, they received a username that allowed them to connect to the online learning environment and which was used to
manipulate anonymity or identification following a method similar to the one used by Postmes et al. (2002). Usernames were manually
generated in a database by the experimenter before the experiment and matched with the code number that participants previously
received by email. For example, a participant, whose real name was known by the experimenters, given the code number 001 was matched
with an individuated username; the next participant in the alphabetical list given the code number 002 was matched with an anonymous
username, and so forth. In the anonymity condition, the username was composed of the team symbol (@ or #) followed by an underscore,
three random characters, a point, and three random numbers (e.g., #_tgb.322, @_gfh.564). The team symbol was included in the username in
order to generate an apparent uniformization, reinforcing the notion of belonging to a given group (see Kim, 2009). In the individuation
condition, random characters and numbers were replaced by the first three letters of the participant's first name and the first three letters of
his/her last name (e.g., @_fre.pun, @_flo.bal).

2.3.2. Assessment of students' prior knowledge

To determine each participant's level of computing and Internet knowledge, they had to complete a pretested 20-item web question-
naire. It was composed of a series of twenty statements (e.g., “The keyboard shortcut for Cut is Ctrl þ X”), and the participant indicated
whether it was true, false, or if they did not know. Scores gathered from this questionnaire were used later to divide students into two
equally sized groups, low-prior knowledge students and high-prior knowledge students, using a median-split procedure. The pretest was
administered to 134 undergraduate students who were invited to participate to the first pretest. Analyses of the data revealed that the scores
of correct responses on the questionnaire ranged from 0 to 16 out of a total of 20 (M ¼ 8.1; SD ¼ 3.1). The KolmogoroveSmirnov goodness-of-
fit test yielded no significant results for the distribution, indicating that scores followed a normal distribution, KeS Z ¼ 1.17, ns.

2.3.3. Quizzing task and manipulation of the salience of group identity

Once the two preliminary tasks had been accomplished, participants entered the system with their username to perform online quizzes
alongside both their team members and those of another team. They were exposed to a series of quizzes of varying difficulty, which were
randomly and automatically selected from a large database containing 800 quizzes. Their responses were used to deliver social comparison
feedback according to experimental condition. A “bonus” procedure was chosen to stimulate participation, giving þ2 for a correct answer, 1
for an incorrect answer, and 0 for no response. The responses of each participant were added to produce a single score, which was averaged
with those of other team members for intergroup comparison. The salience of group identity was directly manipulated in the system by means
of a chart giving either the mean score of the two teams (high salience of group identity) or only the mean score of their own team (low salience
of group identity). Scores were calculated in real time for each team over the 12-day period during which quizzes were performed, and they
were automatically updated when a participant completed a quiz. The system was closed at the end of the 12-day period, and participants
were sent an email asking them to answer a web-questionnaire including manipulation checks and a measure of perceived mastery.

2.4. Measures

2.4.1. Manipulation checks

Two measures classically used in experimental social psychology, self-evaluation from social comparison information and perception of
anonymity, were incorporated in the post-experimental questionnaire to test the effectiveness of the manipulations. Participants were asked
to estimate the easiness of evaluating the performance of their own team relative to another team (“I could easily estimate the level of my team
in comparison to another team”) on a 7-point scale ranging from 1 (Totally disagree) to 7 (Totally agree). The perception of anonymity (“My
teammates and I were totally anonymous”) was also measured on a 7-point scale ranging from 1 (Totally disagree) to 7 (Totally agree).

2.4.2. Prior knowledge

Based on the participants' responses on the 20-item web questionnaire, a score was calculated to assess their prior knowledge in the
computing and Internet domain. The score was simply determined by adding the number of correct responses for the 20 items for each
participant. The scores varied from 2 to 20 (M ¼ 11.38; SD ¼ 3.15) with a median value of 12.

2.4.3. Performance
Performance on the online quizzes was measured by summing the scores of each participant over the 12-day period using the “bonus”
metrics described above (correct answer ¼ þ2; incorrect answer ¼ 1; no answer ¼ 0). The number of quizzes performed by participants
and the number of days they did quizzes were also measured. These measures strongly correlated with quiz scores (r ¼ .98, p < .001, and
r ¼ .62, p < .001, respectively). Because the scores were used as metrics to deliver comparison feedback to both participants and virtual
teams, it was decided to use it as measure of performance.

Table 1
Descriptive data of the participants' distribution in teams and conditions (only active participants in the online learning environment are mentioned).

Salience of group identity Means of group size Number of participants Number (and percent) of males Number (and percent) of females
Anonymity Low (N ¼ 10 teams) 8.40 [6e12] 84 20 (23.81%) 64 (76.19%)
High (N ¼ 10 teams) 9.50 [7e12] 95 25 (26.31%) 70 (73.68%)
Individuation Low (N ¼ 10 teams) 8.10 [5e10] 81 17 (26.56%) 64 (79.01%)
High (N ¼ 10 teams) 8.30 [6e11] 83 24 (28.91%) 59 (71.08%)
 B. Le H
enaff et al. / Computers & Education 82 (2015) 84e95 89

2.4.4. Additional measure: perceived mastery

In the post-experiment online questionnaire, participants had to estimate their computing and Internet mastery by entering a per-
centage from 0% (no mastery) to 100% (complete mastery) in a single text box. The percentages varied from 10 to 100 (M ¼ 62.09; SD ¼ 16.06).

3. Results

3.1. Descriptive analyses

Each team was initially composed of 15 participants randomly distributed to experimental conditions by the experimenter, but some did
not connect to the online environment, affecting the effective distribution of participants in teams and conditions. As shown in Table 1, the
distribution across teams and conditions remained balanced.

3.2. Manipulation checks

3.2.1. Perception of the salience of group identity

A 2 (Anonymity versus Individuation)  2 (High versus Low salience of group identity) between-participants ANCOVA was performed. As
it can have potential effects, gender was controlled and treated as a covariate in the statistical analyses (see Footnote3). Statistical analyses
showed a significant effect of the salience of group identity on the perceived easiness of evaluating the performance of their team in
comparison with another team. It appeared that participants thought they were more able to evaluate the performance of their own team
when the salience of group identity was high (M ¼ 5.70 and SE ¼ .15) than when it was low (M ¼ 4.36 and SE ¼ .17), R2 ¼ .15, F (1,
203) ¼ 34.61, MSE ¼ 2.59, p < .001.

3.2.2. Perception of anonymity

The same ANCOVA was performed on the perception of anonymity, and results yielded a significant effect of the anonymity manipu-
lation, with participants feeling more anonymous in the anonymity condition (M ¼ 5.91 and SE ¼ .19) than in the individuation condition
(M ¼ 3.62 and SE ¼ .20), R2 ¼ .26, F (1, 203) ¼ 68.81, MSE ¼ 3.83, p < .001.

3.3. Performance

Because participants were in teams, and therefore not entirely independent of each other, a mixed model analysis was performed in
order to account for the ‘team’ factor varying in intercepts across participants. Three independent variables and a covariate were integrated
in the two-level random intercepts model. The three variables were anonymity (Anonymity versus Individuation), salience of group identity
(Low versus High), and prior knowledge (High versus Low) of participants whose scores on the initial computing knowledge questionnaire
were divided into two equal groups from the median (Me ¼ 12); gender was treated as a covariate.
At level 1 (individual), the independent variables and the covariate were integrated in the fixed part of the model to look for an asso-
ciation with quiz performance. At level 2 (group), the ‘team’ variable was integrated in the random part of the model to determine whether
belonging to a team influenced performance. With regard to the random part of the model, the inclusion of ‘team’ resulted in a non-
significant between-team variance when the (unrestricted) Maximum Likelihood estimation method was used (c2 ¼ 0.185, df ¼ 1,
p > .05), indicating that belonging to a team did not have an impact on performance. Thus, introducing the ‘team’ factor in the random part of
the model did not make a significant contribution to the effects of the fixed variables on performance. By contrast, the independent variables
included in the fixed part of the model revealed both main and interaction effects on performance.
Although not expected, results revealed that participants had higher scores in the anonymity (M ¼ 519.07 and SE ¼ 30.31) than in the
individuation condition (M ¼ 428.32 and SE ¼ 31.43), F (1, 342) ¼ 4.32, MSE ¼ 161848.42, p ¼ .038, s2 ¼ .013. Not surprisingly, participants
with high prior knowledge had higher scores (M ¼ 518.07 and SE ¼ 31.17) than those with low prior knowledge (M ¼ 429.31 and SE ¼ 31.11),
F (1, 342) ¼ 3.99, MSE ¼ 161848.42, p ¼ .046, s2 ¼ .012. More interesting, the predicted significant three-way interaction was found, F (1,
342) ¼ 11.41, MSE ¼ 161848.42, p < .001, s2 ¼ .033 (see Fig. 1). The interaction was decomposed between the two conditions of salience of
group identity in order to provide a better test of this effect and improve its interpretation.

3.3.1. High salience of group identity

As predicted, results showed that low prior knowledge participants had better scores under the anonymity than individuation condition,
t (81) ¼ 2.914, p < .01. By contrast, no significant difference was found for high prior knowledge participants, t (93) ¼ .977, ns. Likewise, no
significant difference was observed between participants with low and high prior knowledge under the anonymity condition, t (93) ¼ 1.452,
ns, while high prior knowledge participants scored higher than those with low prior knowledge under the individuation condition, t
(81) ¼ 2.339, p < .05.

3.3.2. Low salience of group identity

Contrary to the predictions, results showed that high prior knowledge participants obtained better scores under the anonymity than
individuation condition, t (75) ¼ 2.255, p < .05. By contrast, no significant difference was found with low prior knowledge participants, t
(86) ¼ .288, ns. Moreover, high prior knowledge participants had higher quiz scores than those with low prior knowledge under the

3
Gender was treated as a covariate for three main reasons. First, gender was very unevenly distributed in our sample, with a large majority of women. Second, prior studies
show that gender has an impact on the use of online tools (Bråten & Strømsø, 2006) and attitudes towards computing (Ong & Lai, 2006). It appears that women have lower
computing self-efficacy and behavioral intention to use e-learning than men. Finally, other studies on stereotype threat in the computer domain (Cooper, 2006; Koch, Müller,
& Sieverding, 2008) have revealed that women have lower performance than men when their gender identity is salient than when it is not.
90 B. Le H
enaff et al. / Computers & Education 82 (2015) 84e95

Fig. 1. Performance on quizzes according to anonymity and salience of group identity for low and high prior knowledge students.

Table 2
Means and Standard Deviations by condition as a function of Anonymity/Individuation, Salience of Group Identity, and Prior Knowledge.

Salience of group identity Prior knowledge Performance on quiz Perceived mastery

N M SD N M SD
Anonymity Low Low 47 438.15 378.68 30 56.03 15.38
High 37 629.03 462.37 20 68.40 14.55
High Low 43 570.51 446.84 27 64.26 11.66
High 52 439.83 428.00 33 66.30 14.01
Individuation Low Low 41 461.32 373.93 22 62.23 13.70
High 40 400.90 425.29 21 65.19 19.11
High Low 40 324.57 315.00 27 50.65 16.63
High 43 529.14 462.09 23 66.39 16.83

anonymity condition, t (82) ¼ 2.080, p < .05, while no significant difference was found under the individuation condition, t (79) ¼ .679, ns
(see Table 2).
These results suggest that students with low prior knowledge may succeed as well as those with high prior knowledge particularly when
anonymity accentuates the effects of the saliency of group identity on performance.

3.4. Perceived mastery

Unsurprisingly, a similar analysis performed on the perceived mastery scores revealed that greater mastery was perceived by partici-
pants with high prior knowledge (M ¼ 66.31 and SE ¼ 1.58) than those with low prior knowledge (M ¼ 58.56 and SE ¼ 1.49), F (1,
202) ¼ 12.52, MSE ¼ 230.98, p < .001, s2 ¼ .06. More importantly, a significant three-way interaction was found on perceived mastery, F (1,
202) ¼ 7.21, MSE ¼ 230.98, p ¼ .008, s2 ¼ .04 (see Fig. 2). As previously, the interaction was decomposed from the salience of group identity
variable.

3.4.1. High salience of group identity

Results showed that low prior knowledge participants perceived a higher mastery under the anonymity than under the individuation
condition, t (52) ¼ 3.481, p < .001, while no significant difference was found among high prior knowledge participants, t (54) ¼ .021, ns.
Moreover, a significant difference showed that participants with high prior knowledge perceived greater mastery than those with low prior
knowledge under the individuation condition, t (48) ¼ 3.318, p < .01, while no difference was found under the anonymity condition, t
(58) ¼ .605, ns.

3.4.2. Low salience of group identity

Results showed no significant difference between the anonymity and individuation conditions for either low prior knowledge partici-
pants, t (50) ¼ 1.501, ns, or high prior knowledge participants, t (39) ¼ .603, ns. Moreover, high prior knowledge participants perceived
greater mastery than those with low prior knowledge under the anonymity condition, t (48) ¼ 2.845, p < .01, while no significant dif-
ference was found under the individuation condition, t (41) ¼ .586, ns.
Taken together, these findings are consistent with those obtained for performance, suggesting that participants with low prior
knowledge may perceive a similar level of mastery as those with high prior knowledge under anonymity conditions when their group
identity is rendered salient by intergroup comparison displayed in the online learning environment.
 B. Le H
enaff et al. / Computers & Education 82 (2015) 84e95 91

Fig. 2. Perceived mastery according to anonymity and salience of group identity for low and high prior knowledge students.

4. Discussion

The goal of the present study was to examining the effects of variables at the core of the SIDE model, anonymity/individuation and
salience of group identity, on performance in a team game-based learning environment designed to acquire computing and Internet
knowledge, taking into account the students' prior knowledge.
A few studies have shown that salience of group identity reinforced by intergroup comparison has an impact on motivation and per-
formance on various tasks (James & Greenberg, 1989; Worchel et al., 1998), and research based on the SIDE model has demonstrated that
anonymity influences a large variety of social behaviors by accentuating the salience of group identity and reducing interpersonal differ-
ences (e.g., Lea & Spears, 1991; Postmes, 1997; Postmes et al., 2000, 2001). However, research investigating the joint effect of anonymity/
individuation and salience of group identity on cognitive task performance remains relatively scarce to date (Lea et al., 2002; Michinov et al.,
2004; Tanis & Postmes, 2008), especially in online learning environments for which there is a serious lack of research. Based on both the
SIDE model (Postmes, 1997; Reicher et al., 1995; Spears & Lea, 1992, 1994) and studies demonstrating that intergroup comparison and
anonymity in the classroom may influence academic performance depending on students' prior knowledge (Monteil, 1988, 1993; Monteil &
Huguet, 1993, 2001), it was expected that when group identity was made salient by giving intergroup comparison feedback, students with
low prior knowledge would perform better under anonymity than individuation conditions. Results support this hypothesis. Moreover, not
only did low prior knowledge students perform better when they were anonymous than when they were individuated, they also performed
as well as students with high prior knowledge. These findings are consistent both with the SIDE model (Reicher et al., 1995; Spears & Lea,
1992, 1994) and with previous studies conducted in classroom settings (Monteil, 1988, 1993; Monteil & Huguet, 1993, 2001) where the
cognitive benefits of anonymity are mainly observed when group identity is reinforced by intergroup comparison feedback.
Based on the studies conducted in classroom settings, we expected to observe the reverse effect for students with high prior knowledge
of computing and the Internet, whereby they would perform better under individuation than anonymity conditions, but only when group
identity was rendered salient by giving intergroup comparison feedback in the online learning environment. This hypothesis was not
supported, as no difference was found between anonymity and individuation for students whose group identity was rendered salient. On
the contrary, results showed that students with high prior knowledge performed better under anonymity than individuation conditions, but
only when their group identity was not rendered salient by intergroup comparison. In other words, when group identity was not salient,
students with high prior knowledge were only able to compare themselves with other individuals belonging to their own or another group.
Because these interpersonal comparisons may be threatening for self-esteem, particularly when unfavorable to oneself, they are essentially
useful under anonymity conditions, helping participants maintain a psychological advantage relative to other students and protecting their
self-esteem. This interpretation may be plausible in that anonymity has been recognized as a positive factor protecting participants in online
learning settings (Michinov et al., 2004; Sullivan, 2002), and as a means of reducing apprehension of being evaluated by other group
members (Jessup, Connolly, & Galegher, 1990). Thus, anonymity may protect the students' self-esteem, including in an online learning
environment based on social gaming, allowing them to free themselves from concerns about their performance within their group.
Moreover, irrespective of students' prior knowledge, an unexpected result showed that performance was globally higher under the ano-
nymity than under the individuation condition. This advantage of anonymity on performance is consistent with previous findings about
brainstorming in virtual groups demonstrating that individuals in anonymous groups tend to produce more ideas than groups whose
members are personally identified (e.g., Connolly, Jessup, & Valacich, 1990; McLeod, 2000; Valacich, Dennis, & Nunamaker, 1992). Similarly,
research on peer assessment has shown that anonymity encourages students to participate (Topping, 1998; Yang & Tsai, 2010), for example,
when personal voting systems preserving anonymity are used during lectures to evaluate ones performance on quizzes (Ainsworth et al.,
2011; Draper & Brown, 2004; Kennedy & Cutts, 2005). Because these behaviors are very similar to the production of responses to online
quizzes in the present study and can be defined as additive tasks according to Steiner's (1972) classification, it is reasonable to consider that
92 B. Le H
enaff et al. / Computers & Education 82 (2015) 84e95

similar processes are at work for idea generation tasks in virtual teams and online quizzes when anonymity is preserved. However, other
findings suggest that the systematic positive effects of anonymity in groups should be treated with caution. Indeed, some studies have
demonstrated that, far from increasing performance, anonymity leads to more negative contributions in electronic systems (Postmes et al.,
2000) and encourages social loafing in groups (Karau & Williams, 1993; Shepherd, Briggs, Reinig, Yen, & Nunamaker, 1996). Nevertheless, it
has been shown that social loafing can be reduced by providing social comparison feedback about the performance of each group member
(Jung, Schneider, & Valacich, 2010; Michinov & Primois, 2005; Paulus & Dzindolet, 1993; Roy, Gauvin, & Limayem, 1996; Shepherd et al.,
1996). This interpersonal comparison is recognized as boosting performance in an idea generation task, a roughly similar task to the one
used in the present experiment consisting of completing a series of online quizzes. Indeed, information about the performance of each
participant is likely to be used in the low group identity salience condition, which helps reduce social loafing and boost the performance of
students with high prior knowledge, but only under anonymity conditions. By contrast, the effect of interpersonal comparison may be
threatening for students with low prior knowledge when only this feedback is given. They would probably prefer intergroup comparison
accentuating their group identity, this type of feedback probably stimulating the cognitive efforts made for their group, while at the same
time protecting their self-esteem.
Taken together, our results support the hypothesis that the effect on performance of group identity rendered salient by intergroup
comparison can be accentuated by anonymity in an online learning environment, and that this effect depends on the level of students' prior
knowledge. The present findings partially confirm the predictions based on the SIDE model, extending it to performance in an online
learning environment based on social gaming. Under the anonymity condition, the salience of group identity benefits students with low
prior knowledge, who perform better than those who are individuated and achieve similar performance to students with high prior
knowledge. For the latter, an absence of salience of group identity seems preferable, because they perform better under anonymous than
individuated conditions in this situation. These findings are not fully consistent with classroom studies (Monteil, 1988, 1993; Monteil &
Huguet, 1993, 2001) demonstrating that pupils with high prior knowledge perform better when they are individuated than when they
remain anonymous. However, the way participants were individualized (or anonymized) differed in Monteil's (1988, 1993) classroom
studies compared to the present study. In the former, participants were placed visibly in the classroom, expecting to be questioned by the
teacher whereas in the present study, students were identified by their teammates and/or by members of the other team only by their
username. Despite these differences in the operationalization of the anonymity/individuation variable, both studies put the participants in a
situation where they focused on their own characteristics (skills, self-image, gender, etc.), rather than on others' characteristics or contextual
cues. Consequently, students in the individuation (or visibility) condition may have been placed in a situation where they tried to protect
themselves from a potential threat to their self-esteem within their team (or class), whereas those in the anonymity condition were sys-
tematically protected from possible setbacks during the task (or lesson).
The SIDE model does not account for potential differences between students with low and high prior knowledge involved in a game-
based learning environment, and the present study is the first to offer empirical evidence about the importance of this variable to predict
cognitive performance. However, the additional measure of perceived mastery may also provide some explanation. Indeed, when group
identity was salient, students with low prior knowledge expressed greater perceived mastery under anonymity than individuation
conditions, whereas no difference was expressed by those with high prior knowledge. The latter expressed greater perceived mastery than
students with low prior knowledge, and although their quiz scores were higher than those of students with low prior knowledge, their
performance was better under anonymity than individuation conditions when group identity was not rendered salient, i.e. when they
were only engaged in interpersonal comparison. In the same condition, high prior knowledge students did not perceived a greater
mastery under anonymity than individuation condition. These differences suggest that students with high prior knowledge lack clear-
sightedness about their actual level of mastery, or are over-confident in their skills under the individuation condition. By contrast, stu-
dents with low prior knowledge seemed to perceive their level of mastery accurately in relation to their actual performance depending on
experimental conditions. Additional correlation analyses support this interpretation, because the relationship between performance and
perceived mastery was only significant among low prior knowledge students (r ¼ .26, p < .01), and not among those with high prior
knowledge (r ¼ .08, ns).
The present study is far from being definitive, and further investigations are required, notably to control for virtual team composition. In
the present study, teams were composed on a heterogeneity criterion, mixing students with different levels of prior knowledge and of
different gender. It is reasonable to consider that our results can be explained, at least in part, by this team composition. Indeed, a number of
studies have demonstrated that group composition may have a crucial impact on team functioning and performance. Although team
composition can be homogeneous or heterogeneous with multiple criteria, there is no consensus about which factor determines group and
team performance. However, meta-analyses of the relationship between team composition and team performance suggest that task-related
diversity and diversity in skills or experience have a positive impact on performance, whereas bio-demographic diversity does not seem to
be related to performance (e.g., Horwitz & Horwitz, 2007; Hülsheger, Anderson, & Salgado, 2009). Future studies should thus control for
group composition and its potential impact on performance.

5. Conclusion and practical implications

In line with the SIDE model, the present findings suggest that it is possible for students with low prior knowledge to perform as well as
those with high prior knowledge in a team game-based learning environment under anonymity conditions, when the salience of group
identity is enhanced by providing intergroup comparison feedback. As such, it is the first study to date to broaden predictions based on the
SIDE model to performance on a quizzing task in online educational settings. By demonstrating that the SIDE model can be applied to such
settings to improve cognitive performance, the present study provides new research prospects for the SIDE model and its applications to the
distance education field. It offers some practical implications for the use of social gaming in online learning, and more specifically about the
role of social psychology models in this field (Baldwin & Dandeneau, 2009). Because anonymity is a crucial component of online envi-
ronments, and because intergroup comparison can be used to stimulate team engagement, future online learning environments could be
developed to improve learning and academic performance in many domains other than computing and the Internet.
 B. Le H
enaff et al. / Computers & Education 82 (2015) 84e95 93

Acknowledgments

re
This research was supported by the Brittany Region (France) with ARED funding for Ph.D. students. Special thanks to Je
mie Nogues for
his technical expertise in programming the UL-TE@MS-Q2i team game-based learning environment, and to Elsa Chusseau for the insti-
tutional support enabling us to recruit volunteer Master Degree students to test a new online learning environment based on social gaming.
We would particularly like to thank all the trainees who have helped us to create a large quiz database over the years.

Appendix. Interface of the team game-based learning environment UL-TE@MS/Q2i (anonymity and high salience of group
identity).

References

Abrams, D., & Hogg, M. A. (1990). Social identity theory. New York: Harvester.
Ainsworth, S., Gelmini-Hornsby, G., Threapleton, K., Crook, C., O'Malley, C., & Buda, M. (2011). Anonymity in classroom voting and debating. Learning and Instruction, 21(3),
365e378. http://dx.doi.org/10.1016/j.learninstruc.2010.05.001.
Arena, D. A., & Schwartz, D. L. (2013). Experience and explanation: using videogames to prepare students for formal instruction in statistics. Journal of Science Education and
Technology. http://dx.doi.org/10.1007/s10956-013-9483-3.
Baldwin, M. W., & Dandeneau, S. D. (2009). Putting social psychology into serious games. Social and Personality Psychology Compass, 3(4), 547e565. http://dx.doi.org/10.1111/
j.1751-9004.2009.00185.x.
€m, E., & Klingenberg, B. (2013). The effect of short formative diagnostic web quizzes with minimal feedback. Computers & Education, 60, 234e242. http://
Balter, O., Enstro
dx.doi.org/10.1016/j.compedu.2012.08.014.
Bråten, I., & Strømsø, H. I. (2006). Epistemological beliefs, interest, and gender as predictors of Internet-based learning activities. Computers in Human Behavior, 22(6),
1027e1042. http://dx.doi.org/10.1016/j.chb.2004.03.026.
Cadinu, M., & Reggiori, C. (2002). Discrimination of a low-status outgroup: the role of ingroup threat. European Journal of Social Psychology, 32, 501e515. http://dx.doi.org/
10.1002/ejsp.105.
Cameron, B., & Dwyer, F. (2005). The effect of online gaming, cognition and feedback type in facilitating delayed achievement of different learning objectives. Journal of
Interactive Learning Research, 16, 243e258.
Chen, V. H. H., & Wub, Y. (2013). Group identification as a mediator of the effect of players' anonymity on cheating in online games. Behaviour & Information Technology. http://
dx.doi.org/10.1080/0144929X.2013.843721.
Coffman, T., & Klinger, M. B. (2008). Utilizing virtual worlds in education: the implications for practice. International Journal of Social Sciences, 2(1), 29e33. http://dx.doi.org/
10.1177/1326365X120220010.
Connolly, T., Jessup, L. M., & Valacich, J. S. (1990). Effects of anonymity and evaluative tone on idea generation in computer-mediated groups. Management Science, 36,
689e703. http://dx.doi.org/10.1287/mnsc.36.6.689.
Cooper, J. (2006). The digital divide: the special case of gender. Journal of Computer Assisted Learning, 22, 320e334. http://dx.doi.org/10.1111/j.1365-2729.2006.00185.x.
Daft, R. L., & Lengel, R. H. (1984). Information richness: a new approach to managerial behavior and organization design. In B. Staw, & L. L. Cummings (Eds.), Research in
organizational behavior (pp. 191e233). Greenwich, CT: JAI Press.
Dochy, F., Segers, M., & Buehl, M. M. (1999). The relation between assessment practices and outcomes of studies: the case of research on prior knowledge. Review of
Educational Research, 69(2), 145e186. http://dx.doi.org/10.2307/1170673.
Draper, S. W., & Brown, M. I. (2004). Increasing interactivity in lectures using an electronic voting system. Journal of Computer Assisted Learning, 20(2), 81e94. http://
dx.doi.org/10.1111/j.1365-2729.2004.00074.x.
Empirica. (2006). Benchmarking access and use of ICT in European schools. Bruxelles: Commission europe
enne. Available at http://www.empirica.com/publikationen/
documents/No08-2006_learnInd.pdf.
Erhel, S., & Jamet, E. (2013). Digital game-based learning: impact of instructions and feedback on motivation and learning effectiveness. Computers & Education, 67, 156e167.
http://dx.doi.org/10.1016/j.compedu.2013.02.019.
Eschenbrenner, B., Nah, F. F.-H., & Siau, K. (2008). 3-D virtual worlds in education: applications, benefits, issues, and opportunities. Journal of Database Management, 19(4),
91e110. http://dx.doi.org/10.4018/jdm.2008100106.
Farzan, R., Dabbish, L. A., Kraut, R. E., & Postmes, T. (2011). Increasing commitment to online communities by designing for social presence. In Proceedings of the ACM 2011
Conference on Computer Supported Cooperative Work, Hangzhou, China (pp. 321e330). http://dx.doi.org/10.1145/1958824.1958874.
Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7, 117e140.
94 B. Le H
enaff et al. / Computers & Education 82 (2015) 84e95

Fies, C., & Marshal, J. (2006). Classroom response systems: a review of the literature. Journal of Science Education and Technology, 15, 101e109. http://dx.doi.org/10.1007/
s10956-006-0360-1.
Fourgous, J.-M. (2012). Apprendre autrement a  l’
ere num
erique. Se former, collaborer, innover: Un nouveau modele

educatif pour une

egalit

e des chances [Different ways of learning
in the digital age. Training, collaborating, innovating: A new educational model for equal opportunities]. Paris, France: Ministe re de l'enseignement supe
rieur et de la
recherche. Available at http://www.ladocumentationfrancaise.fr/rapports-publics/124000169/index.shtml.
Goldberg, H. R., & Dintzis, R. (2007). The positive impact of team-based virtual microscopy on student learning in physiology and histology. Advances in Physiology Education,
31, 261e265. http://dx.doi.org/10.1152/advan.00125.2006.
Gue
gan, J., & Michinov, E. (2011). Communication via Internet et dynamiques identitaires: Une analyse psychosociale. Psychologie Française, 56, 223e238. http://dx.doi.org/
10.1016/j.psfr.2011.09.001.
Horwitz, S. K., & Horwitz, I. B. (2007). The effects of team diversity on team outcomes: a meta-analytic review of team demography. Journal of Management, 33, 987e1015.
http://dx.doi.org/10.1177/0149206307308587.
Huguet, P., Charbonnier, E., & Monteil, J.-M. (1999). Productivity loss in performance groups: people who see themselves as average do not engage in social loafing. Group
Dynamics: Theory, Research, and Practice, 3(2), 118e131. http://dx.doi.org/10.1037/1089-2699.3.2.118.
Hülsheger, U. R., Anderson, N., & Salgado, J. F. (2009). Team-level predictors of innovation at work: a comprehensive meta-analysis spanning three decades of research. Journal
of Applied Psychology, 94, 1128e1145. http://dx.doi.org/10.1037/a0015978.
IPSOS. (2012). Am
enagement des classes et transmission des savoirs: les attentes des enseignants [Classroom layout and transmission of knowledge: Teachers' expectations]. Enque ^te
Ipsos/Logica Business Consulting pour Steelcase Education Solutions. Available at http://www.ipsos.fr/sites/default/files/attachments/rapport_steelcase.pdf.
Jackson, G. T., & McNamara, D. S. (2013). Motivation and performance in a game-based intelligent tutoring system. Journal of Educational Psychology, 105(4), 1036e1049. http://
dx.doi.org/10.1037/a0032580.
James, K., & Greenberg, J. (1989). In-group salience, intergroup comparison, and individual performance and self-esteem. Personality and Social Psychology Bulletin, 15(4),
604e616. http://dx.doi.org/10.1177/0146167289154013.
Jessup, L. M., Connolly, T., & Galegher, J. (1990). The effects of anonymity on GDSS group process with an idea-generating task. MIS Quarterly, 14, 313e321. http://dx.doi.org/
10.2307/248893.
Jung, J. H., Schneider, C., & Valacich, J. (2010). Enhancing the motivational affordance of information systems: the effects of real-time performance feedback and goal setting in
group collaboration environments. Management Science, 56(4), 724e742. http://dx.doi.org/10.1287/mnsc.1090.1129.
Karau, S., & Williams, K. (1993). Social loafing: a meta-analytic review and theoretical integration. Journal of Personality and Social Psychology, 65(4), 681e706. http://
dx.doi.org/10.1037/0022-3514.65.4.681.
Kennedy, G. E., & Cutts, Q. I. (2005). The association between students' use of an electronic voting system and their learning outcomes. Journal of Computer Assisted Learning,
21, 260e268. http://dx.doi.org/10.1111/j.1365-2729.2005.00133.x.
Kiesler, S., Siegel, J., & McGuire, T. W. (1984). Social psychological aspects of computer-mediated communication. American Psychologist, 39, 1123e1134. http://dx.doi.org/
10.1037/0003-066X.39.10.1123.
Kim, J. (2009). “I want to be different from others in cyberspace”. The role of visual similarity in virtual group identity. Computers in Human Behavior, 25, 88e95. http://
dx.doi.org/10.1016/j.chb.2008.06.008.
Kim, J., & Park, H. S. (2011). The effect of uniform virtual appearance on conformity intention: social identity model of deindividuation effects and optimal distinctiveness
theory. Computers in Human Behavior, 27, 1223e1230. http://dx.doi.org/10.1016/j.chb.2011.01.002.
van Knippenberg, D. (2000). Work motivation and performance: a social identity perspective. Applied Psychology: An International Review, 49(3), 357e371. http://dx.doi.org/
10.1111/1464-0597.00020.
Koch, S. C., Müller, S. M., & Sieverding, M. (2008). Women and computers. Effects of stereotype threat on attribution of failure. Computers & Education, 51, 1795e1803. http://
dx.doi.org/10.1016/j.compedu.2008.05.007.
Lantz, M. E., & Stawiski, A. (2014). Effectiveness of clickers: effect of feedback and the timing of questions on learning. Computers in Human Behavior, 31, 280e286. http://
dx.doi.org/10.1016/j.chb.2013.10.009.
Lea, M., Rogers, P., & Postmes, T. (2002). SIDE-VIEW: evaluation of a system to develop team players and improve productivity in internet collaborative learning groups. British
Journal of Educational Technology, 33(1), 54e64. http://dx.doi.org/10.1111/1467-8535.00238.
Lea, M., & Spears, R. (1991). Computer-mediated communication, de-individuation and group decision-making. International Journal of Man-Machine Studies, 34, 283e301.
http://dx.doi.org/10.1016/0020-7373(91)90045-9.
Lea, M., & Spears, R. (1992). Paralanguage and social perception in computer-mediated communication. Journal of Organizational Computing, 2(3&4), 321e341. http://
dx.doi.org/10.1177/009365096023001001.
Lea, M., Spears, R., & de Groot, D. (2001). Knowing me, knowing you: anonymity effects on social identity processes within groups. Personality and Social Psychology Bulletin,
27(5), 526e537. http://dx.doi.org/10.1177/0146167201275002.
Lea, M., Spears, R., Watt, S. E., & Rogers, P. (2000). The InSIDE story: social psychological processes affecting on-line groups. In T. Postmes, R. Spears, M. Lea, & S. D. Reicher
(Eds.), SIDE-issues centre-stage: Recent developments in studies of de-individuation in groups. Amsterdam: North Holland.
Lemyre, L., & Smith, P. M. (1985). Intergroup discrimination and self-esteem in the minimal group paradigm. Journal of Personality and Social Psychology, 49(3), 660e670.
http://dx.doi.org/10.1037/0022-3514.49.3.660.
Locksley, A., Ortiz, V., & Hepburn, C. (1980). Social categorization and discriminatory behavior: extinguishing the minimal intergroup discrimination effect. Journal of Per-
sonality and Social Psychology, 39(5), 773e783. http://dx.doi.org/10.1037/0022-3514.39.5.773.
Ma, M., Oikonomou, A., & Jain, L. C. (2011). Serious games and edutainment applications. London: Springer-Verlag.
Marmarosh, C. L., & Corazzini, J. G. (1997). Putting the group in your pocket: using collective identity to enhance personal and collective self-esteem. Group Dynamics: Theory,
Research, and Practice, 1(1), 65e74. http://dx.doi.org/10.1037/1089-2699.1.1.65.
Martinovic, D., & Zhang, Z. (2012). Situating ICT in the teacher education program: overcoming challenges, fulfilling expectations. Teaching and Teacher Education, 28,
461e469. http://dx.doi.org/10.1016/j.tate.2011.12.001.
McLeod, P. L. (2000). Anonymity and consensus in computer-supported group decision making. In T. L. Griffith (Ed.), Research on managing groups and teams: Technology (pp.
175e204). Stamford, CT: JAI Press.
McQuail, D. (2005). McQuail's mass communication theory (5th ed.). London: SAGE Publications.
Michinov, N., Michinov, E., & Toczek-Capelle, M.-C. (2004). Social identity, group processes, and performance in synchronous computer-mediated communication. Group
Dynamics: Theory, Research, and Practice, 8(1), 27e39. http://dx.doi.org/10.1037/1089-2699.8.1.27.
Michinov, E., & Monteil, J.-M. (2003). Interpersonal attraction and social attraction: when group salience cancels the similar-attraction relationship. Canadian Journal of
Behavioural Science, 35(4), 305e315. http://dx.doi.org/10.1037/h0087210.
Michinov, N., & Primois, C. (2005). Improving productivity and creativity in online groups through social comparison process: new evidence for asynchronous electronic
brainstorming. Computers in Human Behavior, 21, 11e28. http://dx.doi.org/10.1016/j.chb.2004.02.004.
Monteil, J.-M. (1988). Comparaison sociale. Strate
gies individuelles et me
diations socio-cognitives. un effet de diffe

renciations comportementales dans le champ scolaire.
European Journal of Psychology of Education, 1(1), 3e18. http://dx.doi.org/10.1007/BF03326304.
Monteil, J.-M. (1993). Soi et le contexte. Paris: Armand Colin.
Monteil, J.-M., & Huguet, P. (1993). The influence of social comparison situations on individual task performance: experimental illustrations. International Journal of Psy-
chology, 28(5), 627e643. http://dx.doi.org/10.1080/00207599308246948.
Monteil, J.-M., & Huguet, P. (2001). The social regulation of classroom performances: a theoretical outline. Social Psychology of Education, 4, 359e372. http://dx.doi.org/
10.1023/A:1011345022870.
Oakes, P. J. (1987). The salience of social categories. In J. C. Turner, M. A. Hogg, P. J. Oakes, S. D. Reicher, & M. S. Wetherell (Eds.), Rediscovering the social group: A self-
categorization theory (pp. 117e141). Oxford: Basil Blackwell.
Oakes, P. J., Haslam, S. A., & Turner, J. C. (1994). Stereotyping and social reality. Malden: Blackwell Publishing.
Ong, C.-S., & Lai, J.-Y. (2006). Gender differences in perceptions and relationships among dominants of e-learning acceptance. Computers in Human Behavior, 22, 816e829.
http://dx.doi.org/10.1016/j.chb.2004.03.006.
Ouwerkerk, J. W., Ellemers, N., & de Gilder, D. (1999). Group commitment and individual effort in experimental and organizational contexts. In N. Ellemers, R. Spears, &
B. Doosje (Eds.), Social identity: Context, commitment, content (pp. 184e204). Oxford England: Blackwell Science.
Paulus, P. B., & Dzindolet, M. T. (1993). Social influence processes in group brainstorming. Journal of Personality and Social Psychology, 64, 575e586. http://dx.doi.org/10.1037/
0022-3514.64.4.575.
 B. Le H
enaff et al. / Computers & Education 82 (2015) 84e95 95

Pilegge, A. J., & Holtz, R. (1997). The effects of social identity on the self-set goals and task performance of high and low self-esteem individuals. Organizational Behavior and
Human Decision Processes, 70(1), 17e26. http://dx.doi.org/10.1006/obhd.1997.2689.
Postmes, T. (1997). Social influence in computer-mediated groups. Amsterdam: Universiteit van Amsterdam.
Postmes, T., & Branscombe, N. R. (2010). Rediscovering social identity. In T. Postmes, & N. R. Branscombe (Eds.), Key readings in social psychology. New York, NY, USA: Psy-
chology Press.
Postmes, T., Spears, R., & Lea, M. (1998). Breaching or building social boundaries? SIDE-effects of computer-mediated communication. Communication Research, 25(6),
689e715. http://dx.doi.org/10.1177/009365098025006006.
Postmes, T., Spears, R., & Lea, M. (2000). The formation of group norms in computer-mediated communication. Human Communication Research, 26(3), 341e371. http://
dx.doi.org/10.1111/j.1468-2958.2000.tb00761.x.
Postmes, T., Spears, R., & Lea, M. (2002). Intergroup differentiation in computer-mediated communication: effects of depersonalization. Group Dynamics: Theory, Research, and
Practice, 6(1), 3e16.
Postmes, T., Spears, R., Sakhel, & de Groot, D. (2001). Social influence in computer-mediated communication: the effects of anonymity on group behavior. Personality and
Social Psychology Bulletin, 27(10), 1243e1254. http://dx.doi.org/10.1177/01461672012710001.
Redersdorff, S., & Martinot, D. (2009). Being outperformed in an intergroup context: the relationship between group status and self-protective strategies. The British Psy-
chological Society, 48, 275e294. http://dx.doi.org/10.1348/014466608X334771.
Reicher, S., Spears, R., & Postmes, T. (1995). A social identity model of deindividuation phenomena. European Review of Social Psychology, 6, 161e198. http://dx.doi.org/10.1080/
14792779443000049.
Ren, Y., Harper, F. M., Drenner, S., Terveen, L., Kiesler, S., Riedl, J., et al. (2012). Building member attachment in online communities: applying theories of group identity and
interpersonal bonds. MIS Quarterly, 36(3), 841e864.
Ren, Y. Q., Kraut, R., & Kiesler, S. (2007). Applying common identity and bond theory to design of online communities. Organization Studies, 28(3), 377e408. http://dx.doi.org/
10.1177/0170840607076007.
Rogers, P., & Lea, M. (2005). Social presence in distributed group environments: the role of social identity. Behaviour & Information Technology, 24(2), 151e158. http://
dx.doi.org/10.1080/01449290410001723472.
Roy, M. C., Gauvin, S., & Limayem, M. (1996). Electronic group brainstorming: the role of feedback on productivity. Small Group Research, 27, 215e247. http://dx.doi.org/
10.1177/1046496496272002.
Shepherd, M. M., Briggs, R. O., Reinig, B. A., Yen, J., & Nunamaker, J. F., Jr. (1996). Invoking social comparison to improve electronic brainstorming: beyond anonymity. Journal of
Management Information Systems, 12, 155e170.
Spears, R., & Lea, M. (1992). Social influence and the influence of the “social” in computer-mediated communication. In M. Lea (Ed.), Contexts of computer-mediated
communication (pp. 30e65). Hemel Hempstead, England: Harvester Wheatsheaf.
Spears, R., & Lea, M. (1994). Panacea or panopticon? The hidden power in computer-mediated communication. Communication Research, 21, 427e459. http://dx.doi.org/
10.1177/009365094021004001.
Spears, R., Lea, M., & Lee, S. (1990). De-individuation and group polarization in computer-mediated communication. British Journal of Psychology, 29, 121e134. http://
dx.doi.org/10.1111/j.2044-8309.1990.tb00893.x.
Spears, R., Lea, M., Postmes, T., & Wolbert, A. (2011). A SIDE look at computer-mediated interaction: power and the gender divide. In Z. Birchmeier, B. Dietz-Uhler, & G. Stasser
(Eds.), Strategic uses of social technology: An interactive perspective of social psychology (pp. 16e39). New York, NY: Cambridge University Press.
Spears, R., & Postmes, T. (2014). Group identity, social influence, and collective action online: extensions and applications of the SIDE Model. In S. S. Sundar (Ed.), The handbook
of psychology and communication technology. West Sussex, England: Wiley-Blackwell.
Squire, K., & Barab, S. (2004). Replaying history: engaging urban underserved students in learning world history through computer simulation games. In ICLS '04 Proceedings
of the 6th International Conference on Learning Sciences (pp. 505e512).
Steiner, I. D. (1972). Group processes and productivity. Academic Press.
Sullivan, P. (2002). “It's easier to be yourself when you are invisible”: female college students discuss their online classroom experiences. Innovative Higher Education, 27(2),
129e144. http://dx.doi.org/10.1023/A:1021109410893.
Tajfel, H. (1974). Social identity and intergroup behaviour. Social Sciences Information/Information sur les sciences sociales, 13(2), 65e93. http://dx.doi.org/10.1177/
053901847401300204.
Tajfel, H. (1978). Social categorization, social identity and social comparison. In H. Tajfel (Ed.), Differentiation between groups: Studies in the social psychology of intergroup
relations (pp. 61e76). London: Academic Press.
Tajfel, H., Billig, M. G., Bundy, R. F., & Flament, C. (1971). Social categorization and intergroup behaviour. European Journal of Social Psychology, 1(2), 149e178. http://dx.doi.org/
10.1002/ejsp.2420010202.
Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. In S. Worchel, & W. Austin (Eds.), Psychology of intergroup relations (2nd ed.). (pp. 7e24).
Chicago: Nelson-Hall.
Tanis, M., & Postmes, T. (2008). Cues to identity in online dyads: effects of interpersonal versus intragroup perceptions on performance. Group Dynamics: Theory, Research, and
Practice, 12, 96e111. http://dx.doi.org/10.1037/1089-2699.12.2.96.
Tarrant, M. (2002). Adolescent peer groups and social identity. Social Development, 11(1), 110e123. http://dx.doi.org/10.1111/1467-9507.00189.
Topping, K. (1998). Peer assessment between students in colleges and universities. Review of Educational Research, 68(3), 249e276. http://dx.doi.org/10.3102/
00346543068003249.
Turner, J. C. (1982). Towards a cognitive redefinition of the social group. In H. Tajfel (Ed.), Social identity and intergroup relations (pp. 15e40). Cambridge, England: Cambridge
University Press.
Turner, J. C. (1987). The analysis of social influence. In J. C. Turner, M. A. Hogg, P. J. Oakes, S. D. Reicher, & M. S. Wetherell (Eds.), Rediscovering the social group: A self-
categorization theory (pp. 68e88). Oxford: Basil Blackwell.
Turner, J. C., & Haslam, S. A. (2001). Social identity, organizations, and leadership. In M. E. Turner (Ed.), Applied social researchGroups at work: Theory and research (pp. 25e65).
Mahwah, NJ, USA: Lawrence Erlbaum Associates Publishers.
Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Rediscovering the social group: A self-categorization theory. Oxford, England: Blackwell.
Turner, J. C., Oakes, P. J., Haslam, S. A., & McGarty, C. (1994). Self and collective: cognition and social context. Personality and Social Psychology Bulletin, 20(5), 454e463. http://
dx.doi.org/10.1177/0146167294205002.
Valacich, J. S., Dennis, A. R., & Nunamaker, J. F. (1992). Group size and anonymity effects on computer-mediated idea generation. Small Group Research, 23, 49e73. http://
dx.doi.org/10.1177/1046496492231004.
Warburton, S. (2009). Second life in higher education: assessing the potential for and the barriers to deploying virtual worlds in learning and teaching. British Journal of
Educational Technology, 40(3), 414e426. http://dx.doi.org/10.1111/j.1467-8535.2009.00952.x.
Worchel, S., Rothgerber, H., Day, E. A., Hart, D., & Butemeyer, J. (1998). Social identity and individual productivity within groups. British Journal of Social Psychology, 37(4),
389e413. http://dx.doi.org/10.1111/j.2044-8309.1998.tb01181.x.
Wouters, P., van Nimwegen, C., van Oostendorp, H., & van der Spek, E. D. (2013). A meta-analysis of the cognitive and motivational effects of serious games. Journal of
Educational Psychology, 105(2), 249e265. http://dx.doi.org/10.1037/a0031311.
Wright, S. S., & Forsyth, D. R. (1997). Group membership and collective identity: consequences for self-esteem. Journal of Social and Clinical Psychology, 16, 43e56. http://
dx.doi.org/10.1521/jscp.1997.16.1.43.
Yang, Y.-F., & Tsai, C.-C. (2010). Conceptions of and approaches to learning through online peer assessment. Learning and Instruction, 20(1), 72e83. http://dx.doi.org/10.1016/
j.learninstruc.2009.01.003.
Young, M. F., Slota, S., Cutter, A. B., Jalette, G., Mullin, G., Lai, B., et al. (2012). Our princess is in another castle: a review of trends in serious gaming for education. Review of
Educational Research, 82(1), 61e89. http://dx.doi.org/10.3102/0034654312436980.