Content uploaded by Daniel Brugman
Author content
All content in this area was uploaded by Daniel Brugman
Content may be subject to copyright.
87
5
MEASURING MORAL DEVELOPMENT
IN ADOLESCENTS
Jan Boom & Daniel Brugman
In the 1970s Lawrence Kohlberg extended his theory on the develop-
ment of moral judgment competence to include a social and educa-
tional component referring to the development of a just community.
Critics have focused on the first version of his theory without taking
into consideration the latter. This shortcoming becomes particularly
obvious with regard to the relation between moral judgment com-
petence and (moral) behavior. In this contribution we will discuss
this criticism, and Kohlberg’s and his colleagues’ acknowledgment of
the role of the contextual moral atmosphere in this relation.
The relations between moral competence, the contextual moral
atmosphere and their subsequent effect on (moral) behavior are still
in need of examination. We suggest that investigating these relations
requires a sharp conceptual and methodological distinction between
moral judgment competence and moral atmosphere. Our main
objective, therefore, is to present and discuss new instruments that
measure each construct separately. The contextual moral atmosphere
is measured by the perception of the adolescents in regular school
settings (as opposed to settings like just community schools, i.e.
schools with a high moral atmosphere).
The relation of these constructs with self reported pro- and
anti-social behavior over time is discussed using data of secondary
school students in Russia as illustration.
88
Measuring moral development
Measuring moral development requires agreement on what aspects
of moral development are central and how they are related. As
Kohlberg and Candee pointed out two decades ago “The first
question raised by the study of moral judgment is that of the
relationship between moral judgment and action” (Kohlberg &
Candee, 1984, p. 498). Although proponents of Kohlberg’s theory of
the development of moral judgment competence often acknowl-
edged the importance of the relation between moral judgment and
action, it has nevertheless often been criticized for its limited
usefulness when moral behavior is at issue (e.g., Kutnick, 1986). In
fact, Kohlberg originally claimed in opposition to the psychoanalytic
and the social learning views that moral reasoning was the only
moral variable that contributed to the prediction of moral action.
Significantly, he did not claim this was a strong relation, but only
that maturity of moral reasoning goes together with maturity in
moral action. This is of particular importance for adolescent behavior
problems in school contexts. Most intervention studies have been
carried out to stimulate adolescents’ moral judgment development
but with the ultimate aim to foster pro-social behavior and reduce
anti-social behavior. Unfortunately, intervention studies for stimu-
lating moral judgment development, while realizing an upward
movement in moral reasoning of about one-third stage or more, have
not been very successful in changing behavior (Kohlberg, 1985).
As these findings demonstrate, the literature on the general
relation between moral judgment competence and moral behavior
shows only a weak correlation between the two. Over the years,
numerous studies have investigated the relation between moral
judgment competence and delinquent behavior. Gregg, Gibbs, and
Basinger (1994) observed that “it has become an established finding”
that the development of moral judgment competence in juvenile
delinquents lags behind the moral competence of their non-
delinquent peers. This lag, however, is small (about 1/3 stage). It is
not well understood how such a small difference in stage of moral
competence could make such a strong difference in behavior (i.e.
delinquent or not) especially when compared to the weak effects of
89
intervention programs to stimulate moral development. A full stage
difference on one specific value domain (“property and law”) as
Gregg et al. report could be the answer. In addition, it is still an open
question whether the delay in moral competence in delinquent
adolescents is a precursor either to delinquency due to a poor edu-
cation (as most Kohlbergians believe) or to participation in a peer
group with a low moral atmosphere, or whether it is an effect of
delinquency due either to the effects of institutionalization in prison
or to cognitive dissonance (Brugman & Aleva, 2004; Leenders &
Brugman, 2005).
In non-delinquent populations, the relation between moral
judgment competence and norm transgressive behavior seems even
weaker (Høst, Brugman, Tavecchio & Beem, 1998). Norm trans-
gressive behavior includes transgressions of norms in the moral,
conventional, and personal domain (Tisak, 1995). Relations between
moral judgment competence and norm transgressive behavior were
usually found in the moral domain, not in the conventional domain
(Kuther, 2000; Kuther & Higgins-D’Alessandro, 2000). It also has
been demonstrated that other moral cognitive variables such as
moral self-relevance are of importance in predicting norm trans-
gressive behavior (Barriga, Morrison, Liau & Gibbs, 2001).
However, as early as 1970, Kohlberg recognized that moral
atmosphere is a powerful predictor of moral action. Moral atmos-
phere refers to the immediate contextual moral atmosphere a person
is engaged in. Subsequently Kohlberg (Kohlberg, Scharf, & Hickey,
1974) started with his “prison project” in which he tried to improve
the moral atmosphere in prison. Later, this project was transferred to
high schools and became known as the just community project.
In what follows we will argue that Kohlberg’s work on moral
atmosphere is more promising for understanding moral behavior
than his work on moral judgment competence was; but that
Kohlberg’s work on moral judgment competence is important too,
especially for understanding development and the relation between
competence, atmosphere, and behavior during development. The
relation between moral judgment and moral behavior can be studied
from a cognitive-developmental viewpoint or from an information-
processing viewpoint. Recent trends to move away from Kohlberg’s
90
theory (see Narvaez & Lapsley, in press; Haidt, 2001) are perhaps
understandable from the information-processing viewpoint that
dominates the North-American scene, but are at odds with the devel-
opmental viewpoint of the theory of Piaget (and also with the early
theory of Kohlberg, see Lind, 2002). In the cognitive developmental
viewpoint, a deliberate moral judgment is necessary in order to
define an act as moral (Blasi, 1999). In the information-processing
viewpoint such a moral judgment is not necessary. From the
information-processing viewpoint moral judgment is at best a
component amongst other components relevant to explain behavior
(Haidt, 2001). As causal explanation is the ultimate objective in this
view, one needs to pay attention to all factors that are correlated with
the actual behavior.
Kohlberg himself has certainly given food to the idea that
competence in making moral judgments is causally related to per-
formance. Clearly it is important from an applied perspective to
study adolescent moral behavior in such a way. Nevertheless, this is
not the only possible perspective. Its powerful basic metaphor is that
of an underlying organization (structure) which is (co)determining
observable behavior. However, another basic metaphor is that of
structure and element as in Piaget’s notion of the INRC group
(Chapman, 1988). In this framework element and structure define
each other mutually but not in a causal sense. Still another basic
metaphor relates structure to meaningful content: in this case the
underlying unity (structure) does not determine the outcome but
gives sense and meaning to it. The last two metaphors are relevant to
a truly developmental perspective as will be explained in the next
section.
In the remainder of this chapter we will first argue that the
notion of moral competence is important to understand moral
development even if its relation to moral behavior is problematic and
we will present an instrument designed to measure this construct in
a way that is as pure as possible. Second, we will argue that the
notion of moral atmosphere (as developed out of the just community
approach) offers a more promising way to understand moral
behavior and we will present an instrument designed to measure this
construct as independently as possible from the construct of moral
91
competence. Finally, we present some empirical results based on
these instruments that tell us more about the complex relation
between moral competence, moral atmosphere, and moral behavior.
Moral competence
In this section we will discuss our work related to moral judgment
competence. The ambition to provide a definition of moral compe-
tence that is as pure as possible has led us to a new measurement
approach (Boom, Brugman, & Van der Heijden, 2001; see below). We
use the phrase “as pure as possible” to mean that we want to isolate
moral judgment competence from everything else in the moral
domain that is related to but not strictly implied by this competence
(i.e. moral atmosphere and moral behavior). Admittedly, defined in
this way, the meaning of moral competence depends on the per-
spective taken.
Our perspective in this section is primarily developmental:
what changes over age (or development, or time, or experience)
regarding the ability to make moral judgments? From a developmen-
tal perspective moral competence characterizes the position of an
individual along a latent developmental trajectory. This trajectory is
mapped to a broadly conceived yardstick (measurement device),
which consists of either discrete markers called stages or a more fine-
grained instrument, such as the moral maturity score.
In this perspective attention goes to the dynamics of the
developmental learning process (e.g. reflective abstraction) or to
special characteristics of a developmental trajectory (e.g. progress
and hierarchical relations). Many considerations that are important
to explain moral behavior might be irrelevant from this viewpoint
because they only pertain to inter-individual differences between
persons or situations and not to intra-individual change.
Moral judgment competence refers to the level of sophistication
an individual has achieved in being able to make moral judgments.
Moral judgments can be seen in the context of a person explaining
and justifying certain courses of action. This can be his/her own
action or someone else’s action. In the ideal case this should make
not much difference since the question always is “what should be
92
done” irrespective of who is in what position. Psychologically,
however, such fluctuations in person or position can be of great
importance (cf. Krebs, Vermeulen, Carpendale, & Denton, 1991).
Although moral judgments are thus clearly related to behavior or
action (the question always is: what should one do?) they do not
cause behavior in any direct sense. In some cases subjects reflect
afterwards on their own acting in interpersonal and social problems
and try to find a certain consistency and accountability in how they
have behaved. In other cases, e.g. for serious problems that do not
require an immediate decision, subjects reflect in advance on what
would be the right course of action and this again involves con-
siderations of consistency and accountability. Thus competence, as it
is active in real life, involves much more than behavior. It involves
the ability to balance the question “what to do concretely” with
several other considerations such as: who one is and wants to be,
how other people might be affected, whether one will be held
responsible afterwards, and the implications this action has for
similar cases in the future.
In sum, the development of moral judgment competence refers
to intra-individual changes (or differences between ages) in complex
meaning-laden constructions. These constructions are structured and
this structure provides unity to expressions that vary widely in
content (including the content of one’s moral behavior). Imposing the
notion of development on such changes and differences is based on
the assumption that a meaningful interpretation of changes in vastly
different life-histories in terms of changing structures of thinking is
possible. From this perspective it makes more sense to claim, as
Kohlberg did, that stage-wise development of the moral judgment
competence satisfies the “hard” stage criteria by having qualitatively
distinct cognitive structures, an invariant developmental sequence,
structural wholeness, and hierarchical integration (Kohlberg, 1984;
Colby & Kohlberg, 1987).
Not surprisingly, this assumption has been challenged for
ignoring individual differences. Indeed, the numerous interactions
and coincidences and varying contexts provide a strong divergent
force generating considerable variance between and within indi-
viduals in ways of thinking concerning moral problems. However, it
93
can be assumed that there is an organizing tendency that provides a
convergent force generating recognizable and shared ways of
thinking. Precisely in so far as a centralizing tendency is plausible,
stages in moral development are to be expected. This hypothesized
convergent tendency can be the result of several effects. It is, at least
partly, the result of learning from mistakes and from reflecting on
one’s own practices. However, the convergence is probably also due
to self-organization in the interactions between people, and to being
conducted by using shared resources, like language, which are
largely pre-given in a normatively prestructured environment.
Our strategy therefore is to acknowledge that moral compe-
tence is an abstract notion, weakly and indirectly related to moral
behavior, and not focused on individual differences. We have
revised measurement procedures accordingly (see below), so that
these criticisms can be countered by invoking the notion of moral
atmosphere as a very important intermediary between moral com-
petence and moral behavior (see next section). However, as far as
moral judgment competence is concerned, we focus on the develop-
mental pattern aspects involved, such as the hierarchical relation
between the stages and the similarities between adolescents on a
more general and abstract level (see below). Recent work by Theo
Dawson (2002) addresses other important aspects of the develop-
mental pattern. Reanalyzing large longitudinal (interview) data sets,
she found support for the ordered acquisition of qualitatively distinct
modes of reasoning (in short: stages) including characteristics like
unidimensionality and gaps that separate most of the stages.
Measuring competence?
How can we assess moral judgment competence, given that it is a
rather abstract notion without a one-to-one correspondence to
behavior? It will come as no surprise that we cannot measure it
directly; we have to infer the stage of development from the
reasoning about relevant problems. In the Kohlberg tradition the
subjects themselves are simply asked. Researchers ask participants to
explain what a protagonist in a moral dilemma should do, implicitly
probing their vision on society, their selves, relationships, and so on.
94
No observations of behavior or measurement of functions are used
(or deemed useful). Before describing our instrument we will expand
a little on this original assessment methodology: the Moral Judgment
Interview (MJI: Colby & Kohlberg, 1987) from which our instrument
is derived.
In moral judgment research, it is common practice to use moral
dilemmas to provoke responses. A moral dilemma typically poses an
ambiguous situation with several action choices available. Such
dilemmas often involve situations in which different parties or per-
sons have competing claims. The essential feature, however, is that
two fundamental moral values (or principles) are competing. For
instance in the famous Heinz-dilemma, Heinz has to make the choice
whether or not to steal a drug that would save his sick and dying
wife (life versus property). The crucial point is that, at nearly all
stage levels, reasons for both courses of action can be given. The
same structure may generate arguments pro and con, and therefore
choices pro and con, for a certain course of action. On the other hand,
manifestly similar behavior (or at least behavior choices) may be
based on structurally dissimilar points of view.
In fact, in the MJI information about the action choice made by
a subject is almost completely disregarded and the focus is exclu-
sively on the reasoning of the subject. True, the interview begins with
the usual question about what the main character should do, but this
information plays almost no role in assigning a stage score to the
participant. It appeared that arguments for both possible choices can
be given and are in fact given by the participants, though some
choices are more likely with higher development. Nevertheless, the
position a participant takes in an interview is regarded as less
informative than the underlying structure of the arguments he or she
uses. In short, what is at stake in the MJI is the individual capacity to
generate judgments in an abstract, generalized, hypothetical sense.
The MJI score is meant to reflect the highest stage of moral judgment
competence an individual can attain. This is why participants are
confronted with hypothetical dilemmas – dilemmas in a fictitious
and remote context – because such dilemmas are supposed to elicit
the highest stage of moral reasoning participants are capable of. In
general, and despite all sorts of criticism, empirical research has
95
shown this to be a successful strategy (Walker, DeVries & Trevethan,
1987). Extending this line even further, in our Moral Judgment
Sorting Task, we want to know in a more direct way how sophis-
ticated their ability is to recognize given statements as moral
judgments.
The MJST
We suggest that an important element of moral judgment compe-
tence is the ability to evaluate the arguments of others, at least those
of lower stages than one’s own. We assume that the kind of thinking
characteristic of lower stages has been subjected to critical reflective
abstraction in the course of a normal subject’s developmental
trajectory. Reflective abstraction leads to the inclusion of previous
thinking in the next stage but in a reorganized form (Boom, 1991).
For example, once you come to see that in orienting yourself toward
the immediate social group – a typical stage 3 view – everybody else
is probably doing the same, you will realize the inherent limitation of
this view. In other words, grasping the structure of your own
thinking by generalizing your previous thinking activities is the
crucial step in proceeding to the next stage (in this case the social
system perspective of stage 4). Therefore, reasoning at stages below
one’s own current stage should be recognized by the subject him or
herself as deficient, regardless of the position (pro or con a certain
course of action) taken in these arguments, whereas arguments
above one’s own stage cannot be expected to be judged consistently.
In turn, we assume that an individual’s current stage can be
adequately characterized by his or her ability to criticize earlier ways
of thinking as somehow (still implicit or, later with development,
more conscious) inadequate by form.
The Moral Judgment Sorting Task (MJST: Boom, Brugman &
Van der Heijden, 2001) is a sorting task designed to test whether
subjects can recognize the inherent hierarchical ordering of stages as
defined in the official Standard Issue Scoring Manual (Colby &
Kohlberg, 1987). The task involves statements representing all the
measurable Kohlberg stages (stages 1 to 5) and intermediate stages
(resulting in 9 categories). We include in-between stages because
96
they are sufficiently demarcated in the manual as separate ways of
responding. Participants are asked to sort these nine statements,
indicative of moral structure, from simplistic to wise. The degree to
which subjects are able to sort in correspondence to Kohlberg is a
theoretical plausible indicator of their ability to think about moral
issues at an abstract level.
Empirical relations
Results obtained with the MJST so far allow a comparison between
1327 Dutch first year university students (mean age 21.62 years) and
977 Russian high school students from grade 8 (mean age 13.4 years)
and 753 from grade 10 (mean age 15.3 years). After presenting them
with a dilemma, the interviewer asked the participants to rank the
nine statements justifying what to do, using a scale ranging from
simplistic to wise. This task was repeated three times (for details see
Boom, Brugman, & Van der Heijden, 2001; Brugman, Podolskij,
Heymans, Boom, Karabanova, & Idobaeva, 2003).
In Figure 1 the average ranking given for each statement for the
three groups are plotted. The rankings display a clear smooth
increase for the university student sample. The shift in the pattern of
average rankings suggests clearly that for the less-educated and
younger (and presumably less-developed) sample of participants,
consistent sorting was more difficult. Additional analysis (using
optimal scaling) revealed that for the younger sample from Stage 2/3
onwards, the statements were effectively treated randomly. This
leads to nearly equal averages for the highest statements.
For each participant the correlation between the ordering
proposed by Kohlberg and the average ordering of the stages over
three sets given by the participants in the MJST can be computed.
This score was used for the analysis reported in the last section. The
average correlation for all university students was 0.56, for grade 8
students it was 0.26, and for grade 10 students it was 0.34.
97
Figure 1.
Moral Judgment Sorting Task: Average rankings for each stage-level of the statements
for two groups of Russian high-school students (grade 8 and 10) and for first year Dutch
university students (4 cohorts combined). Intermediate stages as used by Colby and
Kohlberg et al. (1987) are indicated by combining the adjacent stages numbers.
Perception of moral atmosphere
Colby and Kohlberg (1987, pp. 5-8), while acknowledging that moral
competence and performance may differ, and that subjects not
always use the best reasoning they are capable of, believed that
subjects only use lower stage reasoning in situations characterized by
a significant downward pressure, i.e. a low moral atmosphere.
However, even “decontextualized” moral judgments as measured in
98
the Moral Judgment Interview (or the Moral judgment Sorting Task)
and reflecting moral competence, have been found to be more sus-
ceptible to contextual factors than once thought (e.g., Krebs et al.,
1991; Teo, Becker, & Edelstein, 1995). It has been increasingly recog-
nized that moral judgment and moral action in real life mostly arise
in social situations and may be shared with members of the group to
which one belongs. This can explain why Kohlberg has supple-
mented the traditional individual approach focusing on competence
with a more social approach focusing on everyday moral issues and
the emergent social norms of the social context in which these issues
arise. The “just community theory” is a social approach in which
subjects are asked to take the perspective of the majority of the group
or institution in which they participate. The just community theory
proposes that when one wants to improve moral behavior, one has to
improve the contextual moral atmosphere in which the behavior
occurs. In this respect, secondary schools have been considered of
special importance because on the one hand a low quality of moral
atmosphere has been found in many secondary schools, while on the
other hand the educational goals of secondary schools should foster
commitment to democratic values and human rights.
The contextual moral atmosphere refers to the norms, values,
and meaning systems which participants in a social context share. In
most studies the moral atmosphere is investigated through the
perception of the participants by asking them to assess the majority
perspective on standardized real or semi-real life situations em-
bedded in that context. Power, Higgins and Kohlberg (1989, p. 109)
recognized the difficulty of the task to assess this shared perception
from individual reports: “Individual perceptions are likely to offer
only pieces of the whole, colored by individual differences of various
kinds.” Yet, these researchers were exclusively interested in con-
structing “the whole” moral atmosphere in school (“moral culture”).
Within-school differences between adolescents in their perception of
moral atmosphere were not object of research.
From our viewpoint it is necessary to investigate empirically
the contribution of each of the factors involved in the perception of
school moral atmosphere, like sex, class, grade, and school (Beem,
Brugman, Høst, & Tavecchio, 2004). Research indicates (Høst, Brug-
99
man, Tavecchio, & Beem, 1998, Table V) that the moral atmosphere
characterized as “shared” perception explains a relatively small part
of the adolescents’ perception; by far the biggest part is explained at
the individual level and informal peer group level. In this contribu-
tion we address the relations between moral competence, perception
of moral atmosphere and moral behavior at the individual level.
Consequently, we focus on an individual’s perception of the moral
atmosphere, not on the moral atmosphere of a school as a whole as
constructed by the researcher. An individual student’s perception of
the school’s moral atmosphere is assumed to reflect how well
informed he or she is about what is going on at school and how
capable of taking the perspective of most other students (cf. Taylor &
Walker, 1997). Moral atmosphere and perception of moral atmos-
phere will further be used interchangeably.
Power et al. (1989) derived six constructs for characterizing the
contextual moral atmosphere from their long-term educational
projects in which just community schools were realized, i.e. schools
with a high moral atmosphere. According to them the members of a
school community pass through a distinctive process when they
transform the moral atmosphere in their school. Each of these
constructs is divided into steps or soft stages. (Note: these steps or
soft stages are quite different from the hard stages posited for the
development of individual moral reasoning.) Four constructs were
used for the construction of the School Moral Atmosphere Question-
naire:
(1) “Valuing of the school as an institution”, which refers to the
way the adolescents value their school. Five categories are
distinguished: (a) “Rejection of the school”, (b) “Instrumental
extrinsic valuing of the school as an institution” that helps the pupils
to meet their own needs, (c) “Enthusiastic identification with the
school” at special moments, for example when a team wins an
important game, (d) “Feelings of a spontaneous community”, when
pupils feel a sense of closeness to others and an inner motivation to
help them and to serve the community as a whole, and (e) “The
school as a normative community”, in which the community can
obligate its members in special ways and members can expect others
to uphold group norms and responsibilities.
100
(2) “Stage of community”, which refers to the shared valuing of
the community as a terminal value. Four stages are distinguished. At
stage 1 social relations are dominated by the use and abuse of
physical power. At stage 2 the community is valued to the degree to
which it succeeds in satisfying the concrete needs of individuals. At
stage 3 the community is valued because of the friendships one has
with other members of the community. At stage 4 the community is
explicitly recognized as an identity that can be distinguished from its
members and their relationships. Being a member of a community
means accepting a social contract in which group norms and ideals
are respected.
(3) “Stage of collective reasoning about norms”, which refers to the
shared ways in which issues involving contradictory interests for the
group or its members are resolved. Four stages are distinguished that
are comparable to the first four stages of individual moral
competence (Power et al., 1989, pp. 115-127). Although roughly the
same ways of reasoning as constructed at the individual level are
constructed at the collective level, the notion of development in
collective moral reasoning is quite different from its use in individual
moral reasoning.
(4) “Content of the norm”, which refers to the norms within a
community: caring, trust, integration, participation, publicity,
collective responsibility, substantive and procedural fairness, equity,
and order. No transformation has been proposed in the under-
standing of this construct. As used here, the category “content of the
norm” refers to whether the predicted action in a school-related
dilemma is, for instance, to help or not to help.
The same steps and stages on each of the first three constructs
of moral atmosphere that characterize the development of a school
into a just community were used to study adolescents’ perception of
the moral atmosphere in regular secondary schools. As it turned out,
in the USA most normal secondary schools could be characterized by
the lower steps or stages of each construct, while just community
schools could be characterized by the highest (Power et al., 1989).
Their experience suggested that it may take years to transform a low
school moral atmosphere into a high school moral atmosphere. Con-
sequently, assessing adolescents’ perception of moral atmosphere in
101
regular schools in terms of the steps and stages on these constructs
entails the risk that no systematic differences will be discernible
between normal schools, classes, individuals or over a specific time
span, especially if it is relatively short. Since becoming a just
community is not a goal of these schools, it is not surprising that
proponents of the just community approach invariably found a low
quality moral atmosphere in public high schools in the USA
(Higgins, 1991; Power, 1985). In general, however, it is hard to
believe that there are no differences between secondary schools in
moral atmosphere corresponding to differences between their educa-
tional levels, religious denomination, and other sorts of difference in
pedagogical vision. In the Netherlands schools vary also in the ways
and degrees in which they organize and practice school democracy.
Empirical findings show differences between schools in social atmos-
phere, and the constructs of moral atmosphere include social charac-
teristics as well.
In studies using the School Moral Atmosphere Questionnaire
(see below), differences in adolescents’ perception of the moral
atmosphere in their school were found between samples of schools
from different countries (Netherlands, Russia, and Italy), different
types of school (reflecting different educational levels), different
schools of the same general type, and even between students within
the same class! Differences in the perception of the moral atmosphere
in one’s school are important because the perceptions generate a
corresponding self-fulfilling prophecy: for instance, students who are
inclined to think that most students are truant thereby have a good
reason to be truant themselves and vice versa (Marks & Miller, 1987).
Indeed, as we will make clear, perception of moral atmosphere in
school has been found to be a good predictor of norm-transgressive
behavior in and around school.
Moral atmosphere is a rich concept that can be applied to all
kinds of contexts: peers, family, school, and other institutions.
However, it has been used only in a few instances to describe the
social relations within a particular context and relating it to specific
behavioral outcomes.
102
The SMAQ
In order to investigate the school moral atmosphere in normal
secondary schools we started with ethnographic interviews
(Brugman, Tavecchio, Van Os & Høst, 1999). Based upon the findings
of these interviews, a new questionnaire was designed: the School
Moral Atmosphere Questionnaire (SMAQ, Høst et al., 1998).
The SMAQ is a multiple-choice instrument that aims to
measure adolescents’ perception of school moral atmosphere. The
instrument contains two standardized school dilemmas: one about
helping an unpopular classmate, the other about stealing from (and
preventing stealing from) a classmate. Thus in the measurement of
the perception of the moral atmosphere in school contextual moral
dilemmas are used that are “semi-real”; the school dilemmas are
standardized and are not moral dilemmas which the subjects them-
selves experience, at least not at that moment in school. These
dilemmas are used to measure “the content of the collective norms of
helping and stealing” and “the stage of collective reasoning” about
these norms. The questions concerning these standardized dilemmas
ask the opinion of the individual student on the majority perspective
of the students regarding the norm in their class and school (e.g.
“How many of your classmates would help a student like John, do
you think?”) and the reasoning about helping or not helping (e.g. “If
classmates don't help each other, it isn't much of a class.” Is this
reason close to one most of your classmates would give?). When an
adolescent responds that most of his or her classmates would help
and would think in this way, this adolescent perceives a collective
norm of helping in class and a stage 4 way of reasoning. A classmate,
however, may have a quite different perception. In addition, the
SMAQ contains 5-point Likert items of several subscales that refer to
“valuing the school” and “the school as a community”. These items
also ask for the majority perspective of the students.
Reliability, stability and validity of this instrument were exten-
sively investigated and were found to be acceptable (Brugman et al.,
1999; Høst et al., 1998). A principal component analysis (PCA) exe-
cuted on the moral atmosphere subscale scores at the individual level
resulted in a three components solution. The first component (MA1)
103
was called “connectedness with school” and contained the subscales
“sense of community”, “enthusiastic identification with school”, and
“positive social relations within school”. The second (MA2) was
called “constraint” and contained the subscales “negation of commu-
nity” and “rejection of the school”. The third component (MA3) was
called “collective (contextual) moral judgment” and contained the
subscales “content of the norm helping”, “content of the norm rejec-
tion of theft” and “stage of the collective reasoning about the norm”.
Empirical relations
The relation between adolescents’ perception of the moral atmos-
phere in school and their behavior has been much less studied than
the relation between moral judgment competence and behavior. It is
well known that group membership may contribute to adolescents’
development, socialization and school career. In their just commu-
nity studies, Power et al. (1989) refer to effects on adolescents’ career
planning, their moral competence development and their pro-social
behavior (cf. Kolber, Zalaznick & Noia, 1991). Battistich, Solomon,
Kim, Watson, and Schaps (1995) found correlations between elemen-
tary school students’ sense of community and their academic
attitudes and motives, social and personal attitudes, motives and
behavior, and academic achievement. In the study of Høst et al.
(1998, p. 59) and Brugman et al. (2003), much stronger relations were
found between perceived moral atmosphere and norm transgressive
behavior and pro-social behavior than between moral competence
and these behaviors. Høst et al. (1998) and Brugman et al. (1999)
found that adolescents seem to be guided in their behavior more by
what they expect most other students would do (i.e. their perception
of the school moral atmosphere) than by their moral competence. A
more detailed analysis revealed that the best predictors of behavior
were the social-moral atmosphere components “Constraint” and
“Connectedness with school”, and not the reasoning process of
“Collective moral judgment” (Brugman et al., 2003). These relations
in cross-sectional studies, however, do not clarify how the links
between moral competence, perception of moral atmosphere and
behavior should be specified.
104
Moral atmosphere as a link between moral judgment competence
and action?
How should we view the relation between moral competence and
perception of moral atmosphere? Obviously, they are more
intimately related than are moral competence and moral atmosphere
as such. According to Kohlberg (1984; Colby & Kohlberg, 1987)
moral competence functions as a lens for the perception of moral
atmosphere (but compare Gielen, 1991; Rest, 1986). Consequently,
moral competence would set a ceiling for the perception of the stage
of collective moral reasoning (Power et al., 1989). If moral compe-
tence reflects the highest stage in moral reasoning, one may expect
that students in general would perceive a lower stage of collective
moral reasoning in their schools and classes than their own moral
competence stage. This hypothesis was supported by Brugman et al.
(1999, p. 31) who reported that students’ moral competence scored
1/3 to 2/3 stage higher than their perception of students’ collective
moral reasoning in school.
On the other hand, it should be noted that in our research in
Russian secondary schools (Brugman et al., 2003) we did not find
indications of this large gap between the perception of adolescents’
collective reasoning and individual moral competence. The moral
competence of Russian adolescents was at the same stage as that of
the Dutch adolescents while Russian adolescents perceived a much
better moral atmosphere in their school than Dutch adolescents did.
In addition, Power et al. (1989) also reported that the perception of
the school moral atmosphere in a just community school could go
beyond adolescents’ moral competence. This observation seems to
make sense because the moral atmosphere in just community schools
is assumed to be a stimulant for moral competence. Indeed, one
could argue that if perception could not go beyond competence,
development is difficult. A cognitive conflict, often considered to be
the “motor” for development, is born when one perceives one’s own
kind of reasoning as somehow deficient; this conflict can be induced
by perceiving another kind of reasoning to be better than one’s own
(cf. Berkowitz & Gibbs, 1983; Gibbs, 2003, p. 32, but see pp. 62-63).
105
Although subjects may not understand this reasoning fully, they
dimly are aware of its sophistication and intuitively grasp that it
contains an improvement that takes it beyond their own reasoning.
This mechanism is also supposed to be responsible for the finding
that “recognition-based” moral competence measures (Rest, 1986)
produce higher scores than “production-based” measures like the
Moral Judgment Interview (Colby & Kohlberg, 1987). Finally, it has
been reported that a low moral atmosphere could induce a (tempo-
rary) regression in moral competence (Colby & Kohlberg, 1987). To
summarize, we expect a reciprocal relation to exist between moral
competence and perception of moral atmosphere.
In addition, Kohlberg (1981; Higgins, Power & Kohlberg, 1984)
has speculated about other links between moral competence, moral
atmosphere and behavior. Moral atmosphere could function “as a
bridge” between moral competence and behavior. Thus, the moral
atmosphere would mediate between moral competence and
behavior. Only when the moral atmosphere supports the way of
reasoning characterizing one’s moral competence will the compe-
tence be translated into behavior. For an example one may think of
persons who during the Nazi occupation were unable to carry out
what they regarded as their moral duty to help hide a Jew. The moral
atmosphere weakened their ego-strength. On the other hand the
moral competence can function as buffer in a situation with a low
moral atmosphere. An adolescent with a relatively high moral com-
petence who participates in a peer group with a low moral atmos-
phere may find him or herself unable to participate in the immoral or
antisocial behavior of the group. Here one may think of Cole’s
example of the 14-year-old white boy who not only refused to
participate in the group’s aggressive racist behavior but rescued an
African American boy from an attack by his white buddies (Coles,
1986; see Gibbs, 2003, for an extensive discussion of this example).
Figure 2 (next page) shows a significant negative relation in
Russian adolescents between moral competence measured by the
MJST and antisocial behavior in and around school, explaining 8% of
the variance in antisocial behavior. Two models were constructed
and evaluated to investigate the relations between moral judgment
106
- 0.29
Figure 2.
The relation between Moral competence (MJST: 3 sorting tasks) and Antisocial behavior
in Russian adolescents (χ2=1.7, df=2, p=.43). The strength of the relation is indicated by
the arrow (0.29) and the amount of explained variance (8%) is indicated by the number
in italics.
competence, perception of school moral atmosphere, and antisocial
behavior (Brugman & Boom, 2003). These models represent the per-
ception of moral atmosphere as a mediator between moral compe-
tence and antisocial behavior (see Figure 3), and moral competence
as a moderator between perception of moral atmosphere and anti-
social behavior (see Figure 4).
Figure 3
Moral atmosphere mediating the relation between Moral competence and Antisocial
behavior in Russian adolescents (χ2=10.8, df=9, p=.30). The strength of the relation is
indicated by the arrows and the amount of explained variance is indicated by the number
in italics.
Moral
competence
0.08
Antisocial
behavior
Moral
atmosphere
Moral
competence
Antisocial
behavior
0.70
-0.66
0.18
0.30
107
Figure 4.
Model of Moral competence as a buffer between Moral atmosphere and Antisocial
behavior.
Cross-sectional and longitudinal data over three measurement time
provide evidence for the mediation model. Figure 3 shows that sub-
stantial relations exist between moral competence and perception of
moral atmosphere and between this perception and antisocial
behavior when measured at the same occasion. However, the
originally negative relation between moral competence and anti-
social behavior (see Figure 2) has changed to a non-significant level,
that is, perception of moral atmosphere mediates between moral
competence and antisocial behavior. By adding perception of moral
atmosphere the amount of explained variance in antisocial behavior
increased substantially from 8 to 30%. The same phenomenon was
observed using longitudinal data with moral competence and
perception of moral atmosphere measured at t1 and antisocial
behavior measured at t2, although there the amount of explained
variance in antisocial behavior was lower. However, no evidence
was found for the moderator model: a high moral competence did
not buffer the relation between a negative moral atmosphere and
antisocial behavior.
In addition some results will be presented concerning these
relations using the longitudinal data and including all our measures
during these 3 measurements.
Moral
competence
Moral
atmosphere
Antisocial
behavior
108
Inspecting the cross-lagged paths between the variables showed that:
(a) Antisocial behavior (AB) had a stronger influence on percep-
tion of moral atmosphere (MA) than perception of moral atmosphere
had on antisocial behavior.
(b) Antisocial behavior had a stronger influence on moral compe-
tence (MC) than moral competence had on antisocial behavior.
(c) Moral competence had a more stable influence on perception of
moral atmosphere than moral atmosphere had on moral competence.
(d) Many cross-lagged correlations were weak or non-significant.
These results contradict both of the Kohlbergian models (mediation
and moderating).
Figure 5.
Empirically observed longitudinal relations between Moral competence, Moral atmos-
phere, and Antisocial behavior. Weak relations are characterized by a dashed line.
The rather striking conclusion is that engagement in antisocial
behavior more strongly affects both moral competence and the per-
ception of moral atmosphere than moral competence and perception
of moral atmosphere affect antisocial behavior (Figure 5, for details
see Brugman & Boom, 2003). Consequently, the low moral judgment
competence found in delinquent adolescents could be the effect of
their delinquent behavior instead of a precursor of that behavior.
Moral
atmosphere
Moral
competence
Antisocial
behavior
109
Conclusions
We argued for a conceptual distinction between aspects of moral
development that are important for understanding development and
aspects that are important for understanding (or even predicting)
behavior. We presented instruments designed to measure each: The
Moral Judgment Sorting Task and the School Moral Atmosphere
Questionnaire. This strategy puts us in a good position to examine
the relations between adolescents’ moral competence, their percep-
tion of the school moral atmosphere, and moral behavior.
Until now the literature has provided inconsistent results (i.e.
whether moral competence influenced moral perception or vice
versa) and different proposals (i.e. moral atmosphere as a mediator
between moral competence and behavior, or moral competence as a
buffer between a low moral atmosphere and behavior) regarding the
relations between those constructs. As far as we know we collected
the first longitudinal data relevant to these relations. Although our
data set was limited (age range from 14 to 17 years; only Russian
high school students) and we could not test these models on a second
dataset, we tentatively conclude that (1) perception of school moral
atmosphere mediates the relation between moral competence and
antisocial behavior, and (2) in relation to antisocial behavior both
moral competence and perception of moral atmosphere are more of
importance as effects than as precursors. Obviously, more longi-
tudinal research is needed to study the complex interactions between
moral competence, the contextual moral atmosphere, and moral
behavior. Nevertheless, we hope to have demonstrated that simple
unidirectional causal models in this area can no longer be accepted.
While it has been acknowledged that moral atmosphere could
stimulate development or regression in moral competence, the
possibility also exists that behavior can do the same, not as an
exception but as a rule.
110
References
Barriga, A.Q., Morrison, E.M., Liau, A.K., & Gibbs, J. C. (2001). Moral cognition:
Explaining the gender difference in antisocial behavior. Merrill-Palmer
Quarterly, 47, 532-562.
Battistich, V., Solomon, D., Kim, D, Watson, M., & Schaps, E. (1995). Schools as
communities, poverty levels of student populations, and performance: A
multilevel analysis. American Educational Research Journal, 32, 627-658.
Beem, A.L., Brugman, D., Høst, K., & Tavecchio, L.W.C. (2004). Students’ per-
ception of school moral atmosphere: From moral culture to social compe-
tence. A generalizability study. European Journal of Developmental Psychology,
1, 171-192.
Berkowitz, M.W., & Gibbs, J.C. (1983). Measuring the developmental features of
moral discussion. Merrill-Palmer Quarterly, 29, 399-410.
Blasi, A. (1999). Emotions and moral motivation. Journal for the Theory of Social
Behavior, 29, 1-19.
Boom, J. (1991). Collective development and the learning paradox. Human
Development, 34, 273 - 287.
Boom, J., Brugman, D., & Van der Heijden, P.G.M. (2001). Hierarchical structure
of moral stages assessed by a sorting task. Child Development, 72, 535-548.
Brugman, D., & Aleva, A.E. (2004). Developmental delay or regression in moral
reasoning by juvenile delinquents? Journal of Moral Education, 33, 319-338.
Brugman, D., & Boom, J. (2003). Moral atmosphere in school, adolescents’ moral
development and their misbehavior: A longitudinal study. Presentation at the
European Association for Research in Learning and Instruction. August 28,
Padua, Italy.
Brugman, D., Podolskij, A.I., Heymans, P.G., Boom, J., Karabanova, O., &
Idobaeva, O. (2003). Perception of moral atmosphere in school and norm
transgressive behavior in adolescents: An intervention study. International
Journal of Behavioral Development, 27, 289-300.
Brugman, D., Tavecchio, L.W.C., Van Os, B.J., & Høst, K. (1999). Students’
perception of moral atmosphere in secondary schools: their moral reasoning com-
petence, and their practical judgement in school. MOSAIC (Moral and Social
Action Interdisciplinary Colloquium) Monographs no 8. Bath: University of
Bath Press.
Chapman, M. (1988). Constructive evolution: Origins and development of Piaget's
thought. Cambridge: Cambridge University Press.
Colby, A., & Kohlberg, L. (1987). The measurement of moral judgment. Vol. 1:
Theoretical foundations and research validation. Cambridge: Cambridge Univer-
sity Press.
Coles, R. (1986). The moral life of children. Boston: Houghton Mifflin.
Dawson, T.L. (2002). New tools new insights: Kohlberg’s moral judgment stages
revisited. International Journal of Behavioral Development, 26, 154-166.
Gielen, U. (1991). Research on moral reasoning. In L. Kuhmerker, The Kohlberg
legacy for the helping professions (pp. 39-60). Birmingham, AL: Doxa.
111
Gibbs, J.C. (2003). Moral development and reality. Beyond the theories of Kohlberg and
Hoffman. London: Sage.
Gregg, V., Gibbs, J.C., & Basinger, K.S. (1994). Patterns of developmental delay
in moral judgment by male and female delinquents. Merrill-Palmer Quarterly,
40, 538-553.
Haidt, J. (2001) The emotional dog and its rational tail: A social intuitionist
approach to moral judgment. Psychological Review, 108, 814-834.
Higgins, A. (1991). The just community approach to moral education: Evolution
of the idea and recent findings. In W.L. Kurtines & J.L. Gewirtz (Eds.),
Handbook of moral behavior and development. Vol. 3: Application (pp. 111-141).
Hillsdale, NJ: Erlbaum.
Higgins, A., Power, C., & Kohlberg, L. (1984). The relationship of moral atmos-
phere to judgments of responsibility. In W.L. Kurtines & J.L. Gewirtz (Eds.),
Morality, moral behavior and moral development (pp. 74-106). New York: Wiley.
Høst, K., Brugman, D., Tavecchio, L.W.C., & Beem, A.L. (1998). Students’ percep-
tion of the moral atmosphere in secondary schools and the relationship
between moral competence and moral atmosphere. Journal of Moral Education,
27, 47-70.
Kohlberg, L. (1981). Exploring the moral atmosphere of institutions: a bridge
between moral judgment and moral action. In L. Kohlberg, The meaning and
measurement of moral development (pp. 35-53). Worcester, Mass.: Clark Univer-
sity Press.
Kohlberg, L. (1984). Essays in moral development: Vol. 2. The psychology of moral
development. The nature and validity of moral stages. San Francisco: Harper &
Row.
Kohlberg, L. (1985). The just community approach to moral education in theory
and practice. In M.W. Berkowitz & F. Oser (Eds.), Moral education: Theory and
application (pp. 27-87). Hillsdale, NJ: Erlbaum.
Kohlberg, L., & Candee, D. (1984). The relationship of moral judgment to moral
action. In L. Kohlberg, Essays on moral development: Vol. 2. The psychology of
moral development (pp. 498-581). San Francisco: Harper & Row.
Kohlberg, L., Scharf, P., & Hickey, J. (1974). The justice structure of the prison:
A theory and an intervention. Prison Journal, 51, 3-14.
Kolber, W., Zalaznick, E., & Noia, B. (1991). The impact of a Just Community
experience on student development: Three views from the Scarsdale
Alternative School. In L. Kuhmerker, The Kohlberg legacy for the helping
professions (pp. 123-145). Birmingham, AL: Doxa.
Krebs, D.L., Vermeulen, S.C.A., Carpendale, J.I., & Denton, K. (1991). Structural
and situational influences on moral judgment: The interaction between stage
and dilemma. In W.M. Kurtines & J.L. Gewirtz (Eds.), Handbook of moral
behavior and development, Vol. 2: Research (p. 139-169). Hillsdale, NJ: Erlbaum.
Kutnick, P. (1986). The relationship of moral judgment to moral action:
Kohlberg’s theory, criticism and revision. In S. Modgil & C. Modgil (Eds.),
Lawrence Kohlberg, Consensus and controversy (pp. 125-148). London: Falmer
Press.
112
Kuther, T.L. (2000). Moral reasoning, perceived competence, and adolescent
engagement in risk activity. Journal of Adolescence, 23, 599-604.
Kuther, T.L., & Higgins-D’Alessandro, A. (2000). Bridging the gap between
moral reasoning and adolescent engagement in risky behavior. Journal of
Adolescence, 23, 409-422.
Leenders, I., & Brugman, D. (2005). The moral/non-moral domain shift in
relation to delinquent behaviour in adolescents. British Journal of Develop-
mental Psychology, 23, 65-79.
Lind, G. (2002). Ist Moral lehrbar? Ergebnisse der modernen moralpsychologische
Forschung. Berlin: Logos.
Marks, G., & Miller, N. (1987). Ten years of research on the false-consensus
effect: An empirical and theoretical review. Psychological Bulletin, 102, 72-90.
Narvaez, D., & Lapsley, D. (in press). The psychological foundations of moral
character. In D. Lapsley & C. Power, (Eds.), Character psychology and character
education. Notre Dame, Ind.: University of Notre Dame Press.
Power, C. (1985). Democratic moral education in the large public high school. In
M.W. Berkowitz & F. Oser (Eds.), Moral education: Theory and application (pp.
219-238). Hillsdale, NJ: Erlbaum.
Power, C., Higgins, A., & Kohlberg, L. (1989). Lawrence Kohlberg’s approach to
moral education. New York: Columbia.
Rest, J.R. (1986). Moral development. Advances in research and theory. New York:
Praeger.
Solomon, D., Watson, M., Battistich, V., Schaps, E., & Delucchi, K. (1996).
Creating classrooms that students experience as communities. American
Journal of Community Psychology, 24, 719-748.
Taylor, J.H., & Walker, L.J. (1997). Moral climate and the development of moral
reasoning: The effects of dyadic discussions between young offenders. Journal
of Moral Education, 26, 21-43.
Teo, Th., Becker, G., & Edelstein, W. (1995). Variability in structured wholeness:
Context factors in L. Kohlberg’s data on the development of moral judgment.
Merrill-Palmer Quarterly, 41, 381-393.
Tisak, M.S. (1995). Domains of social reasoning and beyond. Annals of Child
Development, 11, 95-130.
Walker, L.J., De Vries, B., & Trevethan, S.D. (1987). Moral stages and moral
orientations in real-life and hypothetical dilemmas. Child Development, 58,
842-858.
