Journal of Moral Education

Vol. 37, No. 2, June 2008, pp. 165–184

Belonging to and exclusion from the

peer group in schools: influences on
adolescents’ moral choices
Luba Falk Feigenberg*a, Melissa Steel Kinga, Dennis J. Barra,b
and Robert L. Selmana
a
Harvard University, USA; bFacing History and Ourselves, Boston, USA

This paper reports on a mixed methods study of adolescents’ responses to case material about
social exclusion. First, a qualitative coding method is presented that describes the way adolescents
choose and justify strategies to negotiate such situations. The responses were then analysed
quantitatively using chi square tests and multinomial logistic regression. Findings indicate that
adolescents’ interpretation of their social context was a significant factor in their choice of strategy.
Those adolescents who invoked normative rules and conventions as the most salient justifications
were more likely to recommend bystanding rather than joining in the exclusion. However,
adolescents who viewed the protagonist’s own choice as an opportunity for making long-lasting
positive changes in the social environment were more likely to recommend helping the victim.
Gender and school context also were associated with adolescents’ choice of strategy. Implications
for research in moral development as well as practical implications for school-based programming
are discussed.

Introduction: the problem of peer group social exclusion and bullying in the
schools
In her ninth grade class on ethics and history, Eve Shalen wrote an essay about her
involvement with an incident of social exclusion and ostracism that occurred in early
adolescence. About her experience with social relations in middle school, she says:
My eighth grade consisted of 28 students most of whom knew each other from the age
of five or six. Although we grew up together, we still had class outcasts. From second
grade on, a small élite group spent a large portion of their time harassing two or three of
the others. I was one of those two or three, though I don’t know why…The harassment
was subtle. It came in the form of muffled giggles when I talked and rolled eyes when I
turned around. If I was out in the playground and approached a group of people, they

*Corresponding author. Harvard Graduate School of Education, Larsen Hall 610, Appian Way,
Cambridge, MA 02138, USA. Email: luba.feigenberg@childrens.harvard.edu
ISSN 0305-7240 (print)/ISSN 1465-3877 (online)/08/020165-20
# 2008 Journal of Moral Education Ltd
DOI: 10.1080/03057240802009306
166 L. F. Feigenberg et al.

often fell silent. Sometimes someone would not see me coming and I would catch the
tail end of a joke at my expense.
There was another girl in our class who was perhaps even more rejected than I. One
day during lunch…one of the popular girls in the class came up to me to show me
something she said I wouldn’t want to miss. We walked to a corner of the playground
where a group of three or four sat. One of them read aloud from a small book, which I
was told was the girl’s diary. I sat down and, laughing till my sides hurt, heard my voice
finally blend with the others. Looking back, I wonder how I could have participated in
mocking this girl when I knew perfectly well what it felt like to be mocked myself. I
would like to say that if I were in that situation today I would react differently, but I
can’t honestly be sure. (Facing History and Ourselves, 1994, pp. 29–30).

Experiences of social exclusion such as this are quite common to early adolescents
who struggle to navigate social relationships every day. Much of the discourse on
social exclusion focuses on why adolescents who may ‘know better’ still join in when
they see their peers mocking or harassing another student (Bosworth et al., 1999;
Nansel et al., 2001). However, there is little research that examines what might help
adolescents ‘think beyond’ not joining in and, instead, influence them to get
involved to help the victim. In other words, what factors might influence adolescents
to choose to stand up for themselves or others?
In this paper, we explore the way early adolescents think about the choice to
bystand, join the perpetrators or defend the victim in a situation of social exclusion,
such as the one Eve Shalen describes in her school. First, we review the research
literature on how various factors—social cognitive, cultural and contextual—influence
adolescents’ choices about difficult social actions and decisions. We then describe the
construction and validation of a framework to classify both the range of strategies
adolescents recommend in response to the In Group Assessment, a qualitative
measure based on the Eve Shalen case material, as well as the justifications they give
for their choice (Barr, 2005). Next, we present findings about the influence of school
contextual factors on the variation in social choices. We conclude with a discussion of
the implications of our findings for research and practice in moral development and
educational programming to reduce social exclusion.

Social exclusion: definitions of the problem

Often defined as a physical, verbal or psychological action intended to cause fear,
distress or harm to the victim, social exclusion during childhood and adolescence
occurs within the context of an asymmetric power relationship, where a more
powerful child, or group of children, oppresses the less powerful one(s) (Olweus,
1993). Social exclusion in adolescence often includes ostracism, teasing, harassment
and bullying (Swain, 1998). Research in the US suggests between 5 and 27% of
adolescents admit to having excluded a peer (Nansel et al., 2001; Dake et al., 2003).
It is now recognised that social exclusion is seldom the action of only one
individual. The group dynamics required for and created by such behaviour
contributes to the overall culture and climate of schools and social groups. Not
surprisingly, schools with higher rates of exclusion are perceived as less safe (Astor
et al., 2002; Dupper & Meyer-Adams, 2002). Faced with such statistics, schools
 Moral choices 167

struggle to find effective responses to social exclusion and to prevent its negative
consequences in their hallways, cafeterias, gyms and classes.
Further, rigid rules and inflexible consequences in schools typically address the
behaviour of those identified as the ‘perpetrators’ but not that of the ‘bystanders’—
those students who either passively watch or actively incite the exclusion (Staub, 2002).
In fact, even though most adolescents believe social exclusion is wrong, they often do
not try to intervene and usually stand by passively (Tisak et al., 1997; O’Connell et al.,
1999). In order to address this problem, it is important to understand what may cause
adolescents to make choices about their own behaviour in these contexts.

Theoretical orientations and empirical evidence: a brief review

Psychological theories that focus on child and adolescent social cognitive
development primarily describe the quality of individuals’ thought processes that
relate to or influence their social behaviour. Social information processing models,
for example, portray the cognitive steps necessary for individuals to make decisions
about social action (Dodge, 1986; Dodge & Price, 1994; Crick & Dodge, 1996).
Adolescents engage in a series of thought processes and ultimately choose an action
from a range of perceived possibilities. Antisocial behaviour, or social exclusion, may
result from misperceptions of the actions and intentions of others, a deficiency that
may occur at any point in the social information process (Fontaine et al., 2002).
According to these models, adolescents’ choices are due to internal cognitive
structures and abilities where the selection of exclusionary behaviour may be an
indicator of inaccurate or distorted social perceptions (Camodeca et al., 2003).
Other cognitive developmental approaches examine adolescents’ thought processes
as related to social interactions or the understanding, negotiation and meaning of
social relationships over time (Selman, 2003). This approach focuses specifically on
the conditions under which children develop and use the ability to coordinate different
social perspectives (Collins, 2002), rather than conceptualising social competence as a
sequence of social cognitive information processing steps (Selman, 1980; Keller &
Edelstein, 1991). For example, adolescents who have more difficulty coordinating
their own and others’ points of view are at greater risk for peer conflict and may be
more likely to participate in social exclusion (Selman et al., 1992, 1997). Like social
informational processing models, however, many earlier social cognitive-develop-
mental theories (Kohlberg, 1971) located the impetus for adolescents’ choice of social
action primarily in the minds of individuals or at the individual level of analysis. By
focusing mainly on individuals’ social cognitions, these models often do not consider
how other factors may influence adolescents’ social choices, especially under
challenging, complex or ambiguous conditions (Steinberg, 2003).
Differences between adolescents’ social viewpoints and actions can also be attributed
to population level or cultural factors. Research on individuals’ membership of broad
social systems suggests that group affiliation, such as gender, race or socioeconomic
status, may shape the way they think about their choices about behaviour. For example,
gender has been shown to be an important influence on social exclusion. When faced
168 L. F. Feigenberg et al.

with such situations, early adolescent boys tend to prefer to join in with the
perpetrators, while girls tend to side with the victim (Nansel et al., 2001; Seals &
Young, 2003). In contrast, research that focuses specifically on social aggression—
defined as the manipulation of a relationship in order to damage ‘another’s self-esteem,
social status or both’ (Underwood, 2003, p. 23)—suggests girls are much more likely
than boys to be the perpetrators of socially aggressive acts (Crick et al., 1996; Galen &
Underwood, 1997). Further, girls are expected, by both boys and girls, to be the
perpetrators of socially aggressive acts (Crick & Grotpeter, 1995). Lastly, there appear
to be gender differences in preferred coping strategies and desire for adult intervention
in situations of social exclusion (Naylor et al., 2001; Gamliel et al., 2003).
While gender plays an important role in adolescents’ behaviour in social exclusion,
such behaviour does not appear to vary by race or ethnicity, at least in the US (Leff et
al., 1999; Seals & Young, 2003). Yet, when cultural characteristics are considered as
part of the larger context within which exclusion occurs, adolescents view race, as well
as gender, as important factors in thinking about when such behaviour may be
appropriate. For example, adolescents consider others’ racial background as a valid
reason for excluding individuals from particular social relationships or situations, such
as school-based clubs or peer groups (Killen & Stangor, 2001; Killen et al., 2002).
The roles individuals assume within the peer group context also affect the way
they think about social choices. Individuals tend to take on particular roles, across
both situations and time, that contribute to the structural components of intergroup
dynamics. In particular, adolescents tend to view their moral choices differently
depending on whether they are witness to an event or a perpetrator (Salmivalli et al.,
1996; Sutton & Smith, 1999). In other words, adolescents’ role in the social context
influences their choices for behaviour. In this sense, context extends beyond the
impact of external social structures and systems on social and relational behaviour.
Rather, context can be thought of as the system of ‘social activities and cultural
meanings’ in which an individual participates (Serpell, 2002). The way students
understand their choices, then, is grounded in their context and social experiences
(Burton et al., 1996).
The school context, for one, plays an especially important role in the way students
perceive risk behaviour (Kuperminc et al., 1997). For example, the level of safety
adolescents perceive in their schools and communities is related to their thinking
about the meaning of threatening behaviour (Espelage et al., 2000). When students
believe their social environment is unsafe they are more likely to interpret others’
behaviour as hostile or aggressive. The climate fostered in the school plays an
important role in students’ choices around peer group actions (Salmivalli & Voeten,
2004), such as whether they join the perpetrator(s) or defend the victim.

Research questions
This study is guided by the following research questions:
1. What strategies do students recommend for negotiating a situation of social
exclusion and what justifications do they offer for their chosen strategy?
 Moral choices 169

2. Do students’ recommended strategies and justifications for negotiating a

situation of social exclusion differ by their role in the social context, i.e.
whether they are asked to assume the role of witness or perpetrator?
3. Do students’ recommended strategies for negotiating a situation of social
exclusion vary as a function of their perceptions of the social context, i.e.
school or classroom climate?
To address these questions, we designed a mixed-methods study. First, we explored
students’ open-ended responses to case material about a situation of social exclusion
for the strategies they recommend and the justifications they offer for their choices.
Next, we translated the qualitative codes into quantitative categories in order to
examine the influences on students’ choice of strategy. We interpret the implications
of our findings for research in social development as well as school-based social
development programming.

Methods
The research context and procedure
The research described in this paper is part of an ongoing collaborative project with
Facing History and Ourselves (hereafter Facing History),1 an international
organisation that provides resources for teachers and a program for students that
focuses on human behaviour and ethics as a bridge between history and the self
(Tollefson et al., 2004).
Schools with teachers who were known to use the Facing History program and
who were willing to be included in a quasi-experimental study were recruited to
participate in this study. A second school was then recruited within the same town
that matched for similar characteristics. All students in the selected classrooms were
eligible to participate in this research. Only students who returned written parental
consent forms were included in the study; participation rates were above 75% across
the classrooms. Measures were administered during class time by the teachers in the
study and were then turned over to members of the research team. No identifying
information was collected and students’ confidentiality was guaranteed. The
students completed the measures before they had any contact with Facing History
materials and so the data are not construed as evaluation outcome data.

Setting and participants

The participants in this study were students in five public middle schools in
Massachusetts. The schools were all located in districts with a predominantly white,
middle-class student population. As Table 1 shows, there is some variation in
socioeconomic and educational characteristics across the schools, as suggested by
the percentage of students who qualify for free or reduced price lunch (information
about the students in this study and the schools they attended was provided by the
Facing History organisation).
170 L. F. Feigenberg et al.

Table 1. Sample demographics (n5168)

School n Race (% white of the total % Eligible for free/reduced

population)1 price lunch1

A 34 81.0 9.6
B 63 92.0 4.8
C 19 88.0 7.1
D 14 80.0 30.5
E 38 80.0 27.7

Notes: 1Massachusetts Department of Education (2004)

The sample for this study includes 168 students in five Grade 8 classrooms, one
from each school in the study. One hundred and four girls and 64 boys participated.
Students ranged in age from 12–13 years old. No other individual-level demographic
information is available.

Measures
After reading the case material about Eve Shalen, participants completed the In
Group Assessment. This is a survey measure, with two primary questions based on
the case study of Eve Shalen and seven follow-up open-ended items. The measure
was designed to challenge students to assume different vantage points, such as
bystander or victim, about a situation of social exclusion, as well as to capture their
thinking about the issues of interpersonal relationships and social exclusion more
generally (Barr, 2005). The content and language of the assessment questions rely
on the discourse used in Facing History, while the structure of the questions is
driven by a psychological theory about how children develop the capacity to
coordinate various social perspectives, which is central to their understanding and
negotiation of social relationships as well as their developing social awareness
(Selman et al., 1997; Selman, 2003).
For the purposes of this study, we analysed students’ responses to two questions
on the In Group Assessment. The first question asked students to respond as a
witness or bystander to the social exclusion: 1(a) List at least two different ways that
Eve could have acted when she witnessed her classmates picking on other students.
(b) Which would be the best way? (c) Why would that be the best way?
In the second question, students were asked to suggest choices when invited to
join the perpetrators: 2(a) List at least two things Eve could have done when she was
invited to join in the teasing of the other girl. (b) Which would be the best way? (c)
Why would that be the best way?
In each of the two questions, students were asked to write about the choices they
perceived for negotiating the situation and their justifications for each choice. This
allowed us to compare strategies and justifications across each of the two vantage
points. Data for this study include all students who complied with the instructions
for both questions.
 Moral choices 171

Analysis plan
Qualitative analysis: coding development
To address the first research question, we initially examined the responses to code
for the strategies students suggested for negotiating the choice to witness or join in
the social exclusion. Using a semi-grounded approach (Glaser & Strauss, 1967), we
allowed the data, rather than theory, to guide these analyses. First, we open-coded
the responses by grouping similar strategies together. We then clustered the
suggested strategies by salient themes, which resulted in three strategy categories.
The codebook for these strategy categories, including exemplars of each, is shown in
Table 2.
To examine the second portion of the responses, where students justify why their
chosen strategy would be the best one for the situation, we focused on the factors in
the social environment students perceive to be most salient when choosing a
strategy. We used an inductive approach (Boyatzis, 1998) to allow both the data and
theory to guide the coding. This analysis involved a greater amount of interpretation
on our part than the analysis for strategy codes, as we attempted to capture key
elements about how the strategy was explained and the reason given for its selection.
Table 3 provides the codebook for the four justification categories as well as
examples of each.
A three-person team coded the data. Each phase of coding began with a discussion
of the codebooks for strategy and justification (as shown in Tables 2 and 3), the
establishment of anchor responses and independent scoring until an acceptable level
of inter-rater agreement was reached. Both the strategy and justification categories
had over 92% direct agreement and Kappas greater than .9 (Bakeman & Quera

Table 2. Codebook for strategy categories

Code description Anchor response

Upstand N Requires intervention in the existing
situation of ostracism
N Articulates an action that assists the
victim, such as standing up to the group
or comforting the victim
N Aligns against the mocking of the girl
Perpetrate N Aligns with the group mocking the girl
N Action implies that the invitation to join
the mocking has been accepted
N Contributes to the existing situation of
ostracism
Bystand N Aligns with neither the victim nor the
group doing the mocking
N Avoids involvement with the existing
situation of ostracism
N Uninvolvement, active as in walking away,
or passive, by minding one’s own business
‘She could of told them to stop, and
that they were being mean.’
‘She should make friends with that
girl.’
‘She should of just told a teacher what
was going on.’
‘She could play along and make fun
of the kids too.’
‘She should go with the girls.’
‘Ignore them and just go on with her
business.’
‘Make an excuse and walk away.’
172 L. F. Feigenberg et al.

Table 3. Codebook for justification categories

Code description Anchor response

Conventional N References social norms, conventions or
rules (formal or informal) as the main
guiding principle
N Highlights the efficiency or expediency
of the recommended strategy
N Does not explicate reasoning beyond
simple explanations of cost-benefit
analyses that imply one action is simply
‘better’ than another
Safety N Indicates protection as a priority
N Perceives an immediate threat to one’s
emotional or physical well-being
N Indicates that the main goal is to stop
the current situation of ostracism
N Does not reference long term
consequences or implications of
recommended strategy
Relational N Highlights the formation or maintenance
of interpersonal relationship(s)
N Articulates desire for belonging or
connectedness with another person or
with a group of people
N Identifies a connection between people’s
experiences or emotions
Prosocial N Explains connections between the
Transformational recommended action and possible future
consequences or implications
N Speculates about the possible
development of or changes in other
people’s thinking or beliefs
N Articulates opportunities for group
dynamics to shift as a result of the
recommended action
N Implies that the recommended action
could serve as a catalyst for these changes
‘It is the right thing to do.’
‘It would be easier.’
‘Because I think it’s the best.’
‘It would keep everyone out
of trouble.’
‘So they don’t start picking
on her too.’
‘That way nothing bad
happens to the victim.’
‘To make sure she doesn’t get
hurt.’
‘And I could be considered
the ‘‘Popular’’ girl.’
‘Because she’d feel like she fit
in.’
‘Because she knows what it
feels like.’
‘They might realize they’re
doing the wrong thing and
not do it again.’
‘I believe that if enough
people are willing to do
something about a
problem, the problem
would not exist anymore.’
‘She would make a good
influence on other people.’
‘That way they would under-
stand what it’s like too.’

[1995] suggest .6 to .8 to be adequate and above .8 to be excellent). Approximately

20% of the sample was used during training and reliability procedures. Once the
training was completed, the remaining surveys were divided among the raters and
scored independently. All of the surveys were used in subsequent analyses.

Quantitative analysis
In order to examine the relationships between strategy and justification categories,
we created categorical variables and assigned each code a numerical value (0–2 for
 Moral choices 173

strategy and 0–3 for justification). These values are not meant to assign hierarchical
differences between the categories, i.e. that one is inherently ‘better’ or more
sophisticated than another. As with all categorical variables, there is no measurable
difference between the codes and the numerical value is irrelevant; the only
comparisons that can be made are of equality or inequality.

Qualitative results
Coding for strategy
The first set of clusters refers to strategies that ‘Upstand’ or that align against the
mocking of the girl. These student responses recommend actions that imply that
helping the victim is the goal. One cluster of responses suggests direct intervention in
the situation by standing up to the group doing the teasing. These types of responses
recommend that the best thing to do would be to ‘tell the group to stop’ or ‘tell them
what they were doing would hurt the girl’. A second cluster of Upstand responses takes
a different tack and recommends actions that demonstrate alignment with the victim,
such as ‘she should make friends with the girl’ or ‘she should go over and talk to that
girl’. Lastly, there are response clusters that suggest strategies that seek the help of an
adult figure to stop the teasing. Most often, these responses typically recommend that
‘she should tell a teacher what was going on’ as a way to handle the situation.
The next category group of responses suggests strategies that ‘Perpetrate’ the
teasing in reaction to the invitation to join the situation. These responses indicate
actions that escalate the situation by contributing to the violence. One cluster of
strategies aligns with the group mocking the girl and joins in the teasing. These
responses explicitly state that the best thing to do would be ‘going with the group of
popular kids’ or to ‘join the in crowd’. Another cluster of responses included in this
category suggests initiating a new conflict. For example, some responses that
recommend that ‘attacking the girls’ or ‘being mean back’ would be the best way to
handle the situation. Because these types of actions perpetuate, as opposed to
decrease, the use of violence they are counted as Perpetrate.
Lastly, the ‘Bystand’ category includes responses that indicate avoiding involve-
ment in the situation. One cluster of responses in this category group suggests that
actively detaching from the situation is the best option, such as ‘walking away’ or
‘not going with the girls’. Students also recommend that she should ‘make an
excuse, like I have to go to the bathroom’ to avoid being involved in the teasing.
Another cluster of responses describes a more passive stance that avoids both an
explicit endorsement of the exclusion and direct perpetration of teasing. These
Bystand responses advocate ‘looking on’ or ‘just standing there’ while the teasing
occurs. The explicit goal of a Bystand response is the attempt to avoid aligning with
either the victim or the group doing the teasing.

Coding for justification

Responses in the ‘Conventional’ justification category stress the importance of
formal rules or informal social norms, pointing to morals or pragmatics. Moral rules
174 L. F. Feigenberg et al.

included responses such as ‘it’s the right thing to do’ or ‘that way she won’t get in
trouble’. Additional responses that are clustered as pragmatic in the Conventional
category include a type of cost-benefit analysis, such as ‘it would work better’ or ‘it
would be easier’ that highlights the expediency or efficiency of the chosen strategy.
There is also a cluster of responses in this group that refers to what ‘most kids’ would
do or provides little explanation beyond the surface level of ‘it’s better’.
A second category of justification, ‘Safety’, prioritises the physical or emotional
welfare of one of the potentially vulnerable people involved in the situation.
Articulating a need for immediate protection in the face of a perceived threat, some
Safety responses prioritise the needs of the self or Eve by referencing the notion that
the chosen strategy would ‘keep her from getting hurt’ or ‘make them leave her
alone’. Other responses point to the need for protecting the other girl (the victim) in
the situation, such as ‘that way they’ll stop picking on the girl’. Though Safety
justifications point to the urgent need for the current situation to stop, no long-term
consequences are mentioned as a reason for the chosen action. Ultimately, the most
important factor in Safety responses is that the action is seen as a necessary response
to provide prompt shelter from harm.
Responses in the ‘Relational’ justification category emphasise a sense of belonging
or connection between people. These Relational justifications place importance on
interpersonal or intergroup relationships, such as ‘that way she’ll have a friend’ or
‘she’ll feel like she fits in’. Relational justifications also included a cluster of
responses that identify the connection between different people’s experiences and/or
emotions. Such responses frequently point to the fact that Eve ‘should know how it
feels’ because she shares the experience as a victim of ostracism. Justifications of this
type articulate an awareness of the benefits (or dangers) of affiliations between
individuals.
The fourth justification category, ‘Prosocial Transformational’, views the chosen
action as a catalyst for change. Responses in this group demonstrate the awareness
that change is possible and explain the possible future effects of the present strategy,
indicating that the chosen strategy may ‘stop this from happening again’. These
justifications articulate opportunities for something to be different as a result of the
current situation, whether the shift is manifested in ‘the bullies having a change of
heart and not doing that anymore’ or larger social structures, such as ‘the situation
may not exist anymore’.

Quantitative results
Chi square tests
To explore the second research question, Chi square tests were conducted to
examine whether students recommended different strategies when asked to respond
to the situation presented in the In Group Assessment from the vantage point of a
bystander as compared with a potential perpetrator. Table 4 shows the distribution
of the strategies and justifications across the two questions.
 Moral choices 175

Overall, students tended to recommend different strategies from the two vantage
points (x2586.8, p,.001). For example, when asked to take the position as a
bystander to social exclusion, the majority of responses recommended an Upstand
strategy, such as ‘tell them to stop’ or ‘stand up to those girls’. In contrast, when
responding to the question about being invited to join the exclusion, the majority of
responses suggested a Bystand strategy, such as ‘walk away’ or ‘make an excuse’.
There was also a significant difference in the justifications between the two
vantage points (x258.52, p,.05). When asked to justify the strategies they chose in
the role of someone invited to join in the ostracism, adolescents tended to use more
relational justifications than when in the role of witness. This shift may occur
because in considering what to do when asked to actively join in the teasing,
individuals become more concerned with the current relationships involved and less
concerned with changing social dynamics. That the justifications also shifted with
changes in vantage point suggests that participants (who were promised anonymity)
took this interview seriously and did not answer in ways that were designed simply to
present themselves or their choices as socially or ethically desirable.

Multinomial logistic regression

Like Chi square analyses, multinomial logistic regression is used to analyse
categorical data. The Chi square analyses demonstrated which strategies occurred
more often in the sample. Multinomial logistic regression is predictive in that it
calculates the likelihood that students will pick one strategy versus another.
Therefore, it allows a comparison between students’ choices of action, showing
which actions are more likely to occur. Also, it is similar to ordinary least squares
regression in that it allows us to examine the effect of other variables that may
contribute to adolescent decision-making, such as their justifications of their actions,
their gender and their social context. The model was built to examine the effect of
individual variables as well as overall model fit.

Table 4. Descriptive statistics for Strategy and Justification categories (n5168)

Q1: Witness to Q2: Invited to join the Chi square test

ostracism ostracism

Strategy (%)
Upstand 78.2 30.2 x2586.8***
Perpetrate 5.2 13.6
Bystand 16.7 56.2
Justification (%)
Conventional 23.3 21.0 x258.52*
Safety 30.1 28.2
Relational 26.7 40.3
Prosocial Transformational 19.9 10.5

Notes: *p,.05, **p,.01, ***p,.001

176 L. F. Feigenberg et al.

Multinomial logistic regression was used to predict adolescents’ choice of strategy,

specifically from the vantage point of being invited to join in the ostracism. This
question was chosen as the outcome because of the significant shift in adolescents’
choice of strategy between the vantage points as reported above. That the majority of
students chose an Upstand strategy from the vantage point of a witness implied that
their responses may be influenced by what they consider to be the most ‘socially
acceptable’ action, as opposed to what they may actually believe is best for their own
situations.
Multinomial logistic regression analyses always require a comparison group. In
this analysis, Bystand strategies were the ‘excluded’ category. In Table 5, therefore,
each strategy column (Perpetrate and Upstand) is compared not to each other but,
rather, to the excluded category, Bystand. Positive coefficients are significant
predictors of a greater likelihood of choosing a Perpetrate or Upstand strategy.
Negative coefficients are related to a greater likelihood of recommending a Bystand
strategy.
Table 5 shows the coefficient estimates in the final model, which includes variables
to capture the influence of predictor variables: gender, school context, and the
justification. Only conventional and prosocial transformational justifications were
included in the final model as neither safety nor relational justifications were
significant predictors in the model building process. There were no significant
interactions among predictor variables. The model in the left column represents the
likelihood of a choice to respond with a bystanding strategy versus a perpetrate
strategy. In the right column, the model predicts the likelihood of the choice of
bystanding over upstanding.
First, we examined which factors might influence an adolescent’s choice to
perpetrate (e.g. join the ostracism) versus to bystand (e.g. remain uninvolved in the
incident). In the left column, gender was significantly associated with a choice to

Table 5. Multinomial logistic regression model to predict the likelihood of a choice to bystand as
opposed to perpetrate or upstand

Perpetrate Upstand

b Exp(b) b Exp(b)

Gender 2.05*** 7.75 .41 1.51

Justification
Conventional 21.99* 7.34 .18 1.19
Prosocial Transformational 2.54 .58 1.81*** .17
School
School A 21.24 0.29 1.84* .16
School B 2.45 .64 2.22*** .11
School C 1.00 2.72 21.42 .24
School D .52 1.68 21.18 .31

Notes: n5168; 22LL590.25; df514; omitted category is Bystand. There were no significant
interactions between predictor variables. *p,.05, **p,.01, ***p,.001.
 Moral choices 177

remain uninvolved (b52.05, p,.001). Girls were more likely than boys to
recommend bystanding over perpetrating; the reverse is also true: there was a
greater likelihood that boys would choose to join the perpetrators than girls. This
gender difference is well supported in prior research as an important factor in
adolescents’ social choices (O’Connell et al., 1999; Cowie, 2000; Ma, 2002; Seals &
Young, 2003).
Our findings also suggest that the way individuals perceive their context is
essential to their choices. In our sample, students tended not to choose a Perpetrate
strategy when they prioritised Conventional rules or socio-moral norms. As the
model in the left column indicates, students who used a Conventional justification
were less likely to choose a Perpetrate strategy than a Bystand strategy (b521.99,
p,.05). In other words, when students focused on conventions, they also tended to
suggest a strategy that does not actively participate in the maltreatment of another,
but does not actively seek a remedy either.
A response of this type might highlight appropriate standards of behaviour, such
as ‘She could of (sic) made an excuse and left. You shouldn’t pick on people.’ In this
response, the choice to avoid the situation is justified by referring to social norms
that dictate what people ‘should’ and ‘shouldn’t’ do. Similarly, another response
suggests that ‘going on with her business’ would be the best thing to do because
‘joining the popular group might get her in trouble’. This student also recommends
remaining uninvolved and alludes to more formalised rules as the main motivating
factor. The possibility of getting into trouble is a deterrent to participating in the
exclusion of another.
In fact, this finding resonates with research on efforts to prevent social exclusion,
harassment and bullying in schools. Schools that work to develop clear rules that are
consistently implemented by all members of the community do see a decrease in the
incidences of social exclusion and bullying (Glover et al., 1998). Having a clear sense
of standards or consequences of treating each other in certain ways is important to
students’ thinking about participating in exclusion or not. These adolescents
believed it would be better to remain uninvolved for reasons of social conventions
and, thus, choose bystanding as the most appropriate strategy in this situation.
While rules and policies were used to justify students’ choice to bystand, they were
not a predictor in their choice to upstand. The third finding from this study is that
those students who demonstrated an awareness of possibilities for change were more
likely to recommend Upstanding strategies. As our model indicates, students who
used Prosocial Transformational justifications were more likely to suggest helping
the victim as the best strategy over bystanding (b51.81, p,.001). The
recommendation to upstand is accompanied by the belief that change in the larger
social context is a possibility.
One response of this type exemplifies the connection between perceiving
opportunities for change and a recommendation to help the victim: ‘I think that
she should have said something because it might have had some kind of effect on the
group’s presence (sic) and future actions.’ This student articulates an awareness that
speaking up may have an impact beyond the immediate situation and could alter
178 L. F. Feigenberg et al.

others’ behaviour. The recommendation to upstand is accompanied by the belief

that prosocial change in the larger social context is possible.
Lastly, school context was a significant predictor variable. Adolescents in School
A and School B were more likely to choose defending the victim as opposed to
remaining uninvolved (b51.84, p,.05; b52.22, p,.001). To try and understand
this effect of context to a greater degree, we included school-level variables in the
model to account for sociocultural characteristics, such as race, parental educational
attainment and family income. None of these variables was significantly related to
the choice of one strategy or another. This finding suggests that the differences
between social choices in schools A and B may be associated with contextual factors
occurring inside each school, such as the school climate or the non-academic aspects
of school (Freiberg, 1999).

Discussion
In this study we explored the way adolescents responded to case material about an
incident of social exclusion. We developed a qualitative coding framework to capture
the strategies adolescents recommend for negotiating the situation as well as the
justifications they offer for their chosen strategy. Using these coding categories, we
analysed the differences between adolescents’ strategies and justifications from the
vantage point of witness as well as perpetrator. We also examined the various
influences, both at the individual and contextual levels, on adolescents’ chosen
strategy. Overall, the findings from this study, and in particular the method of
coding, offer a way to assess the interplay between contextual factors and
adolescents’ choice of action in social situations, such as an incident of social
exclusion.
Findings suggest that the factors adolescents perceive as most salient in their social
environment—as captured by the justification categories—were significantly related
to their choice of strategy. For example, attention to rules and social norms seemed
to play an important role in adolescents’ choice to bystand as opposed to the choice
to perpetrate. Adolescents who used conventional justifications in their responses
were more likely to recommend remaining uninvolved than joining in the exclusion.
While not perpetuating a situation of exclusion may be a more socially or morally
acceptable choice, our findings also suggest that a focus on rules did not influence
adolescents to choose an upstanding strategy rather than bystanding. It was when
adolescents perceived their action as presenting possibilities for prosocial change that
they were more likely to recommend helping the victim. These adolescents interpreted
opportunities for change in their social context and perceived possibilities for things to
be different, or better, in the future. Adolescents who feel their social environment
welcomes participation, and is open to being transformed, are more likely to get
involved when they witness exclusion or other acts of injustice.
Theoretically, it is important for adolescents to think beyond merely remaining
uninvolved and instead consider standing up for the person being harmed as a way to
make change, not just for the individual but for the group or the context as well. It is
 Moral choices 179

also of practical importance when thinking about how school or classroom climates
can most effectively support students in this type of thinking in order to prepare
citizens who will actively contribute to a peaceful and just society.
This does not mean that other interpretations of the context, such as those
focused on safety or relationships, are immature, invalid or not as ‘good’ or ethical in
either the descriptive or the prescriptive sense. As our findings suggest, justification
categories are not meant to reflect an adolescent’s internal cognitive capacity or
developmental maturity alone. Rather, justification categories represent an
individual’s social analysis, as informed by their developmental capacities, past
experiences and their perception of their immediate situation at a particular moment
in time (LaRusso & Selman, 2003). We do not assume individuals who justified a
strategy on this measure with concerns about safety or conventions, for example, are
necessarily less ethical or socially mature or incapable of using a relational or
prosocial transformational justification, particularly if socialised or engaged in a
context that promotes respect, trust and care. Justification is as much a measure of
those factors in the environment that an individual perceives as most salient when
making a social choice, as it is the individual’s capacity to consider change in a
caring, respectful and trusting context.
Can students’ justifications then be used as a more nuanced way to capture their
potential moral development over time? For example, after the implementation of an
intervention—an anti-bullying curriculum or Facing History unit for example—it
may be the hope that students would be more likely to use, or at least reflect upon
the possibility of a prosocial transformational justification. Embodied within
individuals, this particular way of thinking about choices and action may be
interpreted as coming from adolescents who can hold images of themselves as
proactive moral agents. Their perceived responsibility to take action for the future
good suggests that these adolescents feel capable of working toward social change
and attempting to make a positive difference in the world.
However, those adolescents who offer these justifications are more likely to be
embedded already within social contexts that promote and support thinking about
possibilities for transformation. That these adolescents’ primary concern is not safety
or immediate protection implies they have the privilege, or the courage, to
incorporate these concerns into the consideration of how situations might be
different. Their use of a prosocial transformational justification can be interpreted as
an indicator of healthy individual development as well as a reflection of features of a
supportive context that promotes and supports such thinking.

Limitations
The analyses of these two questions have several potential limitations. First, as with
any method that presents individuals with a dilemma, the responses generated by the
In Group Assessment present a picture of adolescents’ thinking about their choices,
as opposed to what they might actually do in such a situation. Multiple methods of
data collection would contribute important information to examine the comparison
180 L. F. Feigenberg et al.

between moral thought and action. Secondly, in this study, school serves as a proxy
for multiple demographic variables, including race and socioeconomic status.
Thirdly, different dilemmas from the In Group Assessment, with variation in topic,
gender of protagonist, etc. might yield different outcomes, but it is important to
know if the categories of strategies and justifications have some generality. Future
research should consider ways to further differentiate these factors as a way to
explore cultural influences on social choice. To further explain contextual
differences between schools, future studies should include measurement of
classroom and/or school climate. Additional information about the atmosphere
within classrooms and school could point to the factors that are particularly
influential in the way adolescents think about social choices. Lastly, to examine the
impact of developmental differences on the variation of strategies suggested and
justifications provided, a broader age range of students should be included.

Conclusions and implications for moral education practice

Interpersonal and intergroup relationships in schools are gaining increasing attention
in research and practice in education and social development (Ladd et al., 2002;
Greenberg, 2003; Selman & Dray, 2006). As awareness of the risks of social
exclusion increases, more schools are implementing prevention programming in an
attempt to decrease the incidence of such types of behaviour (Greenberg, 2003;
Brown et al., 2004). After such interventions, teachers, counsellors, or program
evaluators may initially hope that students will recommend (or at least see as an
option) a strategy that helps or stands up for the victim. Getting involved to support
the victim of a situation of social exclusion, or injustice more broadly, might seem
like the more moral or sophisticated response.
However, practitioners and researchers alike must look beyond the strategy
students recommend for negotiating such social situations. It is not enough to know
whether an individual would suggest helping the victim or joining the perpetrators;
the justifications offered for students’ actions are also important to take into
consideration. Without a more thorough understanding of why students might
suggest particular actions, practitioners and researchers run the risk of making
potentially incorrect judgements about their framing of particular social and moral
perceptions, as well as their general social and moral development.
As developmental contexts, schools play a critical role in supporting adolescents’
thinking about moral choices and actions, both within the schools themselves and in
the broader societal context. Schools can provide adolescents with an environment
that encourages healthy choices and promotes positive relationships between
individuals. Ultimately, it is the responsibility of educators and researchers to
collaborate and communicate in their exploration of the factors that contribute to
these kinds of learning contexts. Adults play a part in the design and maintenance of
prosocial classroom and school climates that are safe and supportive, that encourage
discussion of challenging issues such as the one depicted in our measure and that
present possibilities for students to engage in change efforts. It is in these types of
 Moral choices 181

environments that we have a better chance of seeing students choose to be

upstanders, or at least to value upstanding, based on their capacity to envision and
weigh a diverse range of justifications.

Acknowledgments
The authors would like to thank the generous support of funders, the Facing History
organization and Erin O’Connor for her statistical consultation. The ‘In Group’
measure and initial coding frameworks for the measure were developed as part of the
Facing History and Ourselves ‘Improving Intergroup Relations among Youth’
research initiative funded by the Carnegie Corporation of New York (1996–1998).
The research was further supported by the Spencer Foundation and a Harvard
Graduate School of Education Dean’s Summer Research Fellowship.

Notes
1. The research initiative was funded by the Carnegie Corporation of New York and included an
outcome study (See Schultz et al., 2001) and a qualitative case study (See Barr, 1998).

References
Astor, R. A., Benbenishty, R., Zeira, A. & Vinokur, A. (2002) School climate, observed risky
behaviors and victimization as predictors of high school students’ fear and judgments of
school violence as a problem, Health, Education and Behavior, 29(6), 716–736.
Bakeman, R. & Quera, V. (1995) Analyzing interaction: sequential analysis with SDIS and GSEQ
(New York, Cambridge University Press).
Barr, D. J. (1998) Improving interpersonal and intergroup relations among youth: a study of the
processes and outcomes of Facing History and Ourselves. Unpublished Final Report to
the Carnegie Corporation of New York, Facing History and Ourselves and the Group for the
Study of Interpersonal Development.
Barr, D. J. (2005) Early adolescents’ reflection on social justice: Facing History and Ourselves in
practice and assessment, Intercultural Education, 16(2), 145–160.
Bosworth, K., Espelage, D. L. & Simon, T. R. (1999) Factors associated with bullying behavior in
middle school students, Journal of Early Adolescence, 19(3), 341–362.
Boyatzis, R. E. (1998) Transforming qualitative information (Thousand Oaks, CA, Sage).
Brown, J. L., Roderick, J., Lantieri, L. & Aber, J. L. (2004) The Resolving Conflict Creatively
program: a school-based social and emotional learning program, in: J. Zins, R. P. Weissberg,
M. C. Wang & H. J. Walberg (Eds) Building academic success on social and emotional learning:
what does the research say? (New York, Teachers College Press), 151–169.
Burton, L. M., Obeidallah, D. A. & Allison, K. (1996) Ethnographic insights on social context and
adolescent development among inner-city African-American teens, in: R. Jessor, A. Colby
& R. A. Shweder (Eds) Ethnography and human development (Chicago, University of Chicago
Press), 395–418.
Camodeca, M., Goossens, F. A., Schuengel, C. & Terwogt, M. M. (2003) Links between social
information processing in middle childhood and involvement in bullying, Aggressive
Behavior, 29(2), 116–127.
Collins, W. A. (2002) Historical perspectives on contemporary research in social development, in:
P. K. Smith & C. H. Hart (Eds) Blackwell handbook of childhood social development (Oxford,
Blackwell), 3–23.
182 L. F. Feigenberg et al.

Cowie, H. (2000) Bystanding or standing by: gender issues in coping with bullying in English
schools, Aggressive Behavior, 26(1), 85–97.
Crick, N. R., Bigbee, M. A. & Howes, C. (1996) Gender differences in children’s normative
beliefs about aggression: how do I hurt thee? Let me count the ways, Child Development, 67,
1003–1014.
Crick, N. R. & Dodge, K. A. (1996) Social information-processing mechanisms in reactive and
proactive aggression, Child Development, 67(3), 993–1002.
Crick, N. R. & Grotpeter, J. K. (1995) Relational aggression, gender and social-psychological
adjustment, Child Development, 66, 710–722.
Dake, J. A., Price, J. H. & Telljohann, S. K. (2003) The nature and extent of bullying at school,
Journal of School Health, 73(5), 173–180.
Dodge, K. A. (1986) A social information processing model of social competence in children, in:
M. Perlmutter (Ed.) Minnesota symposium in child psychology (vol. 18) (Hillsdale, NJ,
Lawrence Erlbaum), 77–125.
Dodge, K. A. & Price, J. M. (1994) On the relation between social information processing and
socially competent behavior in early school-aged children, Child Development, 65(5),
1385–1397.
Dupper, D. R. & Meyer-Adams, N. (2002) Low-level violence: a neglected aspect of school
culture, Urban Education, 37(3), 350–364.
Espelage, D. L., Bosworth, K. & Simon, T. R. (2000) Examining the social context of bullying
behaviors in early adolescence, Journal of Counseling and Development, 78(3), 326–333.
Facing History and Ourselves (1994) Holocaust and human behavior (Brookline, MA, Facing
History and Ourselves National Foundation).
Fontaine, R. G., Burks, V. S. & Dodge, K. A. (2002) Response decision processes and
externalizing behavior problems in adolescents, Development and Psychopathology, 14(1),
107–122.
Freiberg, H. J. (Ed.) (1999) School climate: measuring, improving and sustaining healthy learning
environments (London, Falmer Press).
Galen, B. R. & Underwood, M. K. (1997) A developmental investigation of social aggression
among children, Developmental Psychology, 33(4), 589–600.
Gamliel, T., Hoover, J. H., Daughtry, D. W. & Imbra, C. M. (2003) A qualitative investigation of
bullying, School Psychology International, 24(4), 405–422.
Glaser, B. G. & Strauss, A. L. (1967) The discovery of grounded theory: strategies for qualitative
research (Chicago, Aldine).
Glover, D., Cartwright, N., Gough, G. & Johnson, M. (1998) The introduction of anti-bullying
policies: do policies help in the management of change?, School Leadership & Management,
18(1), 89–105.
Greenberg, M. T. (2003) Enhancing school-based prevention and youth development through
coordinated social, emotional and academic learning, American Psychologist, 58(6/7),
466–477.
Keller, M. & Edelstein, W. (1991) The development of socio-moral meaning making: domains,
categories and perspective-taking, in: W. M. Kurtines & J. L. Gewirtz (Eds) Handbook of
moral behavior and development (vol. 2: Research) (Hillsdale, NJ, Lawrence Erlbaum
Associates), 89–114.
Killen, M., Lee-Kim, J., Mcglothlin, H. & Stangor, C. (2002) How children and adolescents
evaluate gender and racial exclusion, Monographs of the Society for Research in Child
Development, 67(4).
Killen, M. & Stangor, C. (2001) Children’s social reasoning about inclusion and exclusion in
gender and race peer group contexts, Child Development, 72(1), 174–186.
Kohlberg, L. (1971) From is to ought: how to commit the naturalistic fallacy and get away with it
in the study of moral development, in: T. Mischel (Ed.) Cognitive development and
epistemology (New York, Academic Press).
 Moral choices 183

Kuperminc, G. P., Leadbeater, B. J. R., Emmons, C. & Blatt, S. J. (1997) Perceived school
climate and difficulties in the social adjustment of middle school students, Applied
Developmental Psychology, 1(2), 76–88.
Ladd, G. W., Buhs, E. S. & Troop, W. (2002) Children’s interpersonal skills and relationships in
school settings: adaptive significance and implications for school-based prevention and
intervention programs, in: P. K. Smith & C. H. Hart (Eds) Blackwell handbook of childhood
social development (Oxford, UK, Blackwell), 394–415.
LaRusso, M. D. & Selman, R. L. (2003) The influence of school atmosphere and development on
adolescents’ perceptions of risks and prevention: cynicism versus skepticism, in: D. Romer
(Ed.) Reducing adolescent risk (San Francisco, Sage), 113–122.
Leff, S. S., Patterson, G. R., Kupersmidt, J. B. & Power, T. J. (1999) Factors influencing teacher
identification of peer bullies and victims, School Psychology Review, 28(3), 505–517.
Ma, X. (2002) Bullying in middle school: individual and school characteristics of victims and
offenders, School effectiveness and school improvement, 13(1), 63–89.
Massachusetts Department of Education (2004) Massachusetts public school districts Available
online at: http://profiles.doe.mass.edu/districts.asp (accessed 23 April 2005).
Nansel, T. R., Overpeck, M., Pilla, R. S., Ruan, W. J., Simons-Morton, B. G. & Scheidt, P.
(2001) Bullying behaviors among US youth: prevalence and association with psychosocial
adjustment, Journal of the American Medical Association, 285(16), 2094–2100.
Naylor, P., Cowie, H. & Del Rey, R. (2001) Coping strategies of secondary school children in
response to being bullied, Child Psychology & Psychiatry Review, 6(3), 114–120.
O’Connell, P., Pepler, D. J. & Craig, W. (1999) Peer involvement in bullying: insights and
challenges for intervention, Journal of Adolescence, 22(4), 437–452.
Olweus, D. (1993) Bullying in school: what we know and what we can do (Cambridge, MA, Blackwell).
Salmivalli, C., Lagerspetz, K., Bjorkqvist, K., Osterman, K. & Kaukiainen, A. (1996) Bullying as a
group process: participant roles and their relations to social status within the group,
Aggressive Behavior, 22(1), 1–15.
Salmivalli, C. & Voeten, M. (2004) Connections between attitudes, group norms and behavior in
bullying situations, International Journal of Behavioral Development, 28(3), 246–258.
Schultz, L. H., Barr, D. J. & Selman, R. L. (2001) The value of a developmental approach to
evaluating character development programmes: an outcome study of Facing History and
Ourselves, Journal of Moral Education, 30(1), 3–27.
Seals, D. & Young, J. (2003) Bullying and victimization: prevalence and relationship to gender,
grade level, ethnicity, self-esteem and depression, Adolescence, 38(152), 735–747.
Selman, R. L. (1980) The growth of interpersonal understanding (Orlando, FL, Academic Press).
Selman, R. L. (2003) The promotion of social awareness: powerful lessons from the partnership of
developmental theory and classroom practice (New York, Russell Sage Foundation).
Selman, R. L. & Dray, A. J. (2006) Risk and prevention, in: R. M. Lerner & W. Damon (Eds)
Handbook of child psychology (vol. 4: Child psychology in practice) (New York, Wiley).
Selman, R. L., Schultz, L. H., Nakkula, M., Barr, D., Watts, C. L. & Richmond, J. B. (1992)
Friendship and fighting: a developmental approach to the study of risk and prevention of
violence, Development and Psychopathology, 4(4), 519–558.
Selman, R. L., Watts, C. L. & Schultz, L. H. (Eds) (1997) Fostering friendship (New York, Aldine
de Gruyter).
Serpell, R. (2002) The embeddedness of human development within sociocultural context, Social
Development, 11(2), 290–295.
Staub, E. (2002) The psychology of bystanders, perpetrators and heroic helpers, in: L. Newman
& R. Erber (Eds) Understanding genocide: social psychological contributions (New York, Oxford
University).
Steinberg, L. (2003) Is decision making the right framework for research on adolescent risk
taking?, in: D. Romer (Ed.) Reducing adolescent risk: toward an integrated approach (Thousand
Oaks, CA, Sage).
184 L. F. Feigenberg et al.

Sutton, J. & Smith, P. K. (1999) Bullying as group process: an adaptation of the participant role
approach, Aggressive Behavior, 25, 97–111.
Swain, J. (1998) What does bullying really mean?, Educational Research, 40(3), 358–364.
Tisak, M. S., Lewis, T. & Jankowski, A. M. (1997) Expectations and prescriptions for responding
to peer aggression: the adolescent offenders’ perspective, Aggressive Behavior, 23(3),
149–160.
Tollefson, T., Barr, D. J. & Strom, M. S. (2004) Facing History and Ourselves, in: W. G. Stephan
& P. W. Vogt (Eds) Education programs for improving intergroup relations: theory, research and
practice (New York, Teachers College Press).
Underwood, M. K. (2003) Social aggression among girls (New York, Guilford Press).