See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/284003090

Development and validation of a scale of social and moral judgments (ABB

scale) and its use in prison settings

Article · October 2015

DOI: 10.1016/j.erap.2015.07.002

CITATIONS

4 authors:

Abdellaoui Sid Marcel Lourel

University of Lorraine University of Lille Nord de France
41 PUBLICATIONS   109 CITATIONS    57 PUBLICATIONS   397 CITATIONS   

SEE PROFILE SEE PROFILE

Catherine Blatier Jean-Léon Beauvois

Université Grenoble Alpes 79 PUBLICATIONS   1,590 CITATIONS   
40 PUBLICATIONS   153 CITATIONS   
SEE PROFILE
SEE PROFILE

Some of the authors of this publication are also working on these related projects:

special issue of europeran review of applied psychology View project

Are Some Rabbits More Competent and Warm Than Others?: The Lay Epistemologist Is Interested in Object Value, Not in Descriptive
Parameters View project

All content following this page was uploaded by Abdellaoui Sid on 20 July 2018.

The user has requested enhancement of the downloaded file.

 Revue européenne de psychologie appliquée 66 (2016) 171–180

Disponible en ligne sur

ScienceDirect
www.sciencedirect.com

Original article

Development and validation of a scale of social and moral judgments

(ABB scale) and its use in prison settings
Construction et validation d’une échelle de jugements socio-moraux (échelle ABB)
et son utilisation en milieu carcéral
S. Abdellaoui a,∗ , M. Lourel b , C. Blatier c , J.-L. Beauvois d
a
Laboratoire INTERPSY, université de Lorraine, campus lettres et sciences humaines et sociales, 23, boulevard Albert-1er , 54015 Nancy cedex, France
b
Laboratoire RECIFES, ÉSPÉ, université d’Artois, 59000 Lille, France
c
Laboratoire interuniversitaire de psychologie, université de Grenoble-Alpes, 38040 Grenoble cedex, France
d
Laboratoire de psychologie expérimentale et quantitative, université de Nice-Sophia-Antipolis, 24, avenue des Diables-Bleus, 06000 Nice, France

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

Article history: Introduction. – The traditional approach to value judgments involves determining the position of an
Received 26 May 2015 individual on a scale designed to evaluate the underlying mechanisms and dimensions of judgments.
Accepted 17 July 2015 Objective. – The purpose of this study was to develop and validate a scale among a general population
and to apply it to individuals particularly affected by, or directly involved in, acts of transgression.
Keywords: Method. – The scale comprises three types of behavior involving an expression of personal values (atypism
Scale or idiosyncratic behavior) or a violation of moral or conventional standards. Subjects were asked to
Judgment
assess a range of actions and behaviors on three dimensions (Likert format): seriousness, excusability
Moral
Values
and rejection of the transgressor.
Tolerance Results and conclusion. – As predicted, factor analysis shows a clear hierarchy of values. The results demon-
Prison strate the multidimensional nature of the instrument and indicate good reliability. Tolerance and severity
indices were developed to understand the underlying dynamics of social and moral judgments. The study
found that inmates’ judgments of violations and transgressions differed in some respects from the judg-
ments made by the general population. The influence of context and the role of group membership as an
explanatory factor are examined from the point of view of the identity strategies used.
© 2016 Published by Elsevier Masson SAS.

r é s u m é

Mots clés : Classiquement, les jugements de valeur correspondent à une position des gens sur des échelles sans
Échelle s’attacher aux dimensions sous-jacentes aux jugements. Cet article présente l’élaboration, la validation
Jugement d’un outil auprès d’une population générale et son utilisation auprès d’une population particulièrement
Valeur concernée par la transgression. L’outil se compose de trois types de conduites évoquant, soit l’expression
Moral
d’une valeur personnelle (atypisme), soit une transgression de valeurs conventionnelles ou morales. Les
Tolérance
sujets devaient évaluer ces conduites sur trois dimensions distinctes (format Likert) relatives à la gravité
Prison
d’une transgression, son excusabilité et au rejet dont doit être l’objet le transgresseur. Le traitement des
scores bruts, réalisé au moyen d’une analyse factorielle révèle, comme attendu, une bonne hiérarchie des
valeurs. Les résultats valident la nature multidimensionnelle de l’instrument et attestent d’une fiabilité
acceptable. Des indices de tolérance et de sévérité ont été construits en vue d’une mise en relief de la
dynamique sous-jacente aux jugements socio-moraux. Les détenus produisent des jugements de trans-
gressions en quelques points différents des jugements produits par la population générale. L’influence du
contexte et de l’appartenance comme élément d’explication est évoquée sous des stratégies identitaires
mobilisées.
© 2016 Publié par Elsevier Masson SAS.

∗ Corresponding author.
E-mail address: siduniversite@yahoo.fr (S. Abdellaoui).

http://dx.doi.org/10.1016/j.erap.2015.07.002
1162-9088/© 2016 Published by Elsevier Masson SAS.
172 S. Abdellaoui et al. / Revue européenne de psychologie appliquée 66 (2016) 171–180
1. Introduction
Adherence to rules, norms and values (whether explicit or
implicit) is a major focus of both personal and collective concerns.
Adherence and compliance are often one of the key foundations of
actions or strategies aimed at promoting or maintaining a frame-
work, whether normative, regulatory or legislative. Research in this
area is closely related to research issues surrounding delinquency
and is of particular interest to those directly involved in justice, but
also concerns people serving a prison sentence. However, beyond
the specific characteristics of transgressors, the tendency to focus
on the notions of adherence and compliance often means that we
tend to overlook questions surrounding the context-dependent
nature of rules, norms and values and the variability of social and
moral judgments according to the social and relational context
in which they are made (Abdellaoui, Personnaz, & Aubry, 2004;
Abdellaoui & Pittolo, 2000; Mazé, Finkelstein, & Quentin, 2004).
For example, most people would agree that writing graffiti on a
wall at school, using public transport without paying, embezzling
public funds or abusing or mistreating someone are reprehensible
acts. In other words, most people would agree on the seriousness of
the transgression, the intentionality of the act, the level of personal
responsibility and the appropriate sanction or punishment (Howe,
1994; Nemeth & Sosis, 1990; Przygodzki & Mullet, 1997). In many
cases, a moral judgment will be a function of the acceptability of the
behavior or of the perception (i.e. the appraisal) of the transgres-
sor (Morchain, 2009). Research has shown that judgment type is
determined by a range of factors (Ebbesen & Konecni, 1981). In par-
ticular, our understanding of judgments will depend, among other
things, on the amount of information available to assess the dynam-
ics of social and moral judgments. The purpose of this study was to
develop a measurement scale designed to capture the dynamics of
judgment (specifically, social and moral judgments) and to identify
key dimensions of judgments in this area in order to improve our
understanding of the inherent characteristics of transgression situ-
ations. More specifically, the aim was to validate the tool among the
general population, to test it in a prison population and to identify
and understand its characteristics. This approach could contribute
to the development of evaluation protocol perceptions and also
supports helping the development of human and social values in
general (Mathys, Lanctôt, & Touchette, 2013; Pélissier & Alidières,
2014).
The study was based on the assumption that the analysis of
social and moral judgments must take into account the type
of value involved (i.e. violated) in an act of transgression, the
dimension involved in the judgment and the degree of covariation
between different dimensions of the judgment. To the best of our
knowledge, post-Kohlbergian conceptions of the sphere of social
and moral values have not produced standardized measurement
tools. This paper presents a tool based on the idea that human
beings are intuitive moralists. Drawing on Piaget’s work on the
development of moral judgment (1932), Kohlberg (1969, 1971)
proposed a theory to explain the cognitive framework underly-
ing individual decision-making in the context of a social-moral
dilemma. Kohlberg developed a conception in which morality can
be released (through cognitive development) from the shackles
of obligations and egocentrism first, and from rules and conven-
tions second, thereby resulting in the cognitive management of
broad universal principles. Kohlberg (1971) defined six “stages of
moral development”, or “developmental stages”, operating at three
levels: sequential, invariant and hierarchical (for a comprehen-
sive overview, see Tostain, 2000). The Kohlbergian conception led
to the development of various methods for the analysis of moral
judgments, such as those proposed by Rest (1979, 1986), although
many experts have emphasized on the limitations of work in this
area. In addition to the various criticisms leveled at the validity
of Kohlberg’s Moral Judgment Scale (Kurtines & Greif, 1974), the
developmentalist and universalist conception has been widely crit-
icized (Baumrind, 1986; Salini, 1976; Shweder, 1981). The main
criticisms include:
• the importance given to communication and socialization con-
texts in the development of moral judgment, with social learning
theorists and experts in the anthropology of communication (e.g.
Shweder) challenging the ontological nature of moral values and
explaining their development by reference to socialization con-
texts;
• the irreversible nature of progression through the various
stages of moral development; the assumption is that a person
who has advanced from the conventional level to the post-
conventional level will never again judge or assess situations
from a conventional-level perspective;
• the one-dimensional nature of categories of values; according
to Kohlberg (1983), moral values emerge from other categories
of values, with conventional values representing sub-moral val-
ues from which adolescents infer universal meanings. For others,
moral values merely represent a category of values governed by
certain principles. The assumption is that other categories of val-
ues may be governed by other principles, thus involving other
forms of judgment and appraisal. This is the view taken by Turiel
(1983) and various other scholars (Killen & Hart, 1999; Nucci &
Turiel, 1978; Tisak & Turiel, 1984; Turiel & Smetana, 1984), whose
work provided the basis for this study.
Our view is relatively close to the position taken by Turiel, who
identified three main categories of values:
• moral values: applied to any behavior involving an obligatory,
generalizable, objective and universal judgment, assessment or
appraisal. Moral values are based on a set of principles defined as
universal. Individuals and groups adhere to moral values because
they are founded on the moral good and are constructed as
moral imperatives. Examples include the prohibition of killing
and stealing;
• conventional values: applied to any individual behavior related
to a collective or social convention or involving adherence to
(or compliance with) a stable set of social norms or rules. The
production of a conventional value is generally the result of com-
pliance with or submission to a social norm or rule (with varying
degrees of consent). Typical examples include acts of politeness
and decorum that only apply in relatively specific situations;
• personal values: applied to any behavior not involving a frame-
work of specific social or normative rules, but involving an
individual decision. Here, the process of judging an action or
behavior is neither governed nor expressed explicitly or implic-
itly by a particular group or community, but is an indication
of how individual standards or practices operate rather than a
sign of adherence to collective rules. For example, no one will be
offended to learn that a person has developed a habit of drinking
orange juice at 8am on Tuesdays and Fridays.
2. Aims and objectives
The traditional assumption is that value judgments can be exam-
ined by determining the position of individuals on a scale or based
on their answers to specific questions. Research on social and
moral values has often involved using questions referring to spe-
cific cases of transgression and atypical or idiosyncratic behavior.
The aim is to encourage subjects to make judgments about specific
acts of transgression and examples of atypical behavior (focusing
on reprehensibility). To improve our understanding of personal
 S. Abdellaoui et al. / Revue européenne de psychologie appliquée 66 (2016) 171–180
and social issues related to transgression (particularly identity and
socialization processes), a tool is needed to identify the underly-
ing mechanisms of social and moral judgment. The purpose of this
study was:
• to develop a tool to measure perceptions (i.e. judgments) of dif-
ferent types of transgression and transgressors;
• to use the tool to improve our understanding of the social and
cognitive determinants of these judgments.
Although all areas of society can be affected by acts of violation
and transgression, the question of social and moral transgression is
particularly relevant in prison settings. Questions relating to how
transgressions and transgressors are judged (i.e. appraised) by oth-
ers are a major concern in prisons. The aim was to develop a tool
among a general population and to apply it to a specific population
(i.e. a prison population) particularly affected by the issues raised
by acts of transgression.
Building on the findings of Abdellaoui, Blatier, and Beauvois
(1996), this study has three objectives:
• first, to present subjects with acts of transgression and examples
of atypical or idiosyncratic behavior directly involving the three
categories of values outlined above (i.e. moral, conventional and
personal values);
• second, to develop an analytical method applicable to a general
population which, beyond traditional modes of response (grav-
ity/seriousness, defensibility, reprehensibility), can be used to
identify fundamental dimensions of judgments in terms of social
and moral values (tolerance, severity, etc.);
• third, to assess the effect of belonging to a population of inmates
on these responses and dimensions.
To achieve these objectives, we proceeded in three stages. The
first stage involved developing and pre-testing the tool. The tool
(scale) was tested among a sample of subjects of different ages
and backgrounds. The second stage involved conducting analyses to
assess the internal validity of the scale and the internal consistency
of the items as a whole and of the categories of items relating to each
of the three categories of values. Here, the paper also considers the
factor structure and constructs the validity of the scale. The third
and final stage involved conducting an analysis to assess the impact
of belonging to a prison population composed of 193 inmates on
social and moral judgments. To the best of our knowledge, there has
been little research on social and moral judgments among inmates.
However, there have been a number of studies in this area among
delinquents, notably from a developmental perspective (Basinger,
Gibbs, & Fuller, 1995). There has also been some research among
inmates on causal attribution processes (Abdellaoui, Kouabenan, &
Gilibert, 1998) and representations of violence (Przygodzki-Lionet
& Noël, 2004), among other issues. One possible explanation for
the limited amount of research in this area is the lack of a suitable
tool.
3. Development and pre-test
The assumption was that each item (i.e. each action or behav-
ior) should only correspond to (i.e. involve) one value (whether
moral, conventional or personal). The ambiguity of the object of
measurement has always been a major issue in scale development
(Thurstone & Chave, 1929). To avoid any confusion, we used a
judging method based on specific cases. A preliminary study was
conducted among 140 students in law, history and sociology. Based
on a list of thirty behaviors designed to reflect the three cate-
gories of values, the participants were asked to specify the type
173
of value (moral, conventional, personal) corresponding to each
of the thirty actions or behaviors. Prior to this, the participants
were given a definition of the concepts of “moral value”, “conven-
tional value” and “personal value”. The results of the preliminary
study were conclusive since 12 actions and behaviors identified
as most closely related to one of the three categories of values
were highlighted, indicating that there were no major differences
of opinion. Interrater reliability scores were above 85%, with some
even approaching 100%.
The final scale included 12 items corresponding to the action or
behavior of a person and referring to one of the three types of trans-
gression or atypical behavior (moral, conventional and personal).
The scale was presented in the form of a booklet, with the three
types of statements invariably shown in the same order (personal,
conventional, moral). Each item was rated on a 7-point Likert scale.
The following dimensions were measured: seriousness, excusabil-
ity and rejection.
Example: moral transgression.
Felicien goes door-to-door pretending to be a charity fundraiser
in order to raise enough money to fill up his car with gas.
Example: conventional transgression.
Regis cannot wait to drive home after a hard day at work. He
even drives through stop signs in the middle of the countryside.
Example: personal idiosyncrasy.
Pierre has got into the habit of not taking his socks off when he
goes to bed.
The following scale was used.
What the person did was:
Extremely serious 0 1 2 3 4 5 6 Not at all serious
To what extent are/is the person’s actions/behavior defensible/inexcusable?
Absolutely inexcusable 0 1 2 3 4 5 6 Entirely excusable
The person must be:
Completely rejected 0 1 2 3 4 5 6 Absolutely not rejected
4. Psychometric properties of the scale of social and moral
judgments (ABB scale)1
4.1. Subjects and procedure
The sample was composed of 521 subjects aged 16–74 years.
The mean age of the participants was 28.1 years (standard devia-
tion: 7.4). Of the sample, 41.07% were women, while 58.93% were
men. The participants came from a range of social and occupa-
tional or educational backgrounds (29.3% of the sample were senior
high school and college students, 8% were job seekers, 41.3% were
employed and 15.8% were retired, while 5.6% did not specify their
current status). The test lasted 10 to 15 minutes. The senior high
school and college students completed the scale in groups, while
the other participants completed the scale in individual sessions
and, in most cases, in the presence of the researcher. Participation
was voluntary and confidentiality was assured. Before answer-
ing the questionnaire, the participants were given the following
instruction: “I am a researcher in psychology and I would like to
know what you think about different behaviors you might come
across in everyday life. You will be given a list of 12 items. Please
rate each item on a scale from 0 to 6 on each of the three dimen-
sions (serious, defensible/excusable, rejection). Your answers will
be anonymous. Please answer without considering others’ opin-
ions”.
1
We named the scale the “ABB scale” in reference to the initials of the three
authors – Abdellaoui, Beauvois & Blatier.
174 S. Abdellaoui et al. / Revue européenne de psychologie appliquée 66 (2016) 171–180

Table 1 Table 2
General and specific reliability indices (Cronbach’s alpha). Rotated component matrix.

All items Moral items Conventional Personal 1st factor 2nd factor 3rd factor
items items Moral values Personal values Conventional values

Cronbach’s alpha .88 .78 .82 .80 Item G6 .635 (max.)

Item E6 .623
Item 6R .605
Item 3R .575
4.2. Data analysis Item R12 .569
Item R5 .180 (min.)
The tool provides several scores. For each subject, we calculated: Item R2 .412
Item E3 .510
Item G9 .510
• an overall score based on the sum of responses for all 12 behaviors Item G3 .493
and scales (possible range: 0–216). Item R9 .490
• a score for each of the 3 values characterizing the behaviors (pos- Item E12 .486
Item G2 .328
sible range: 0–72). Item G12 .457
• a score for each scale (seriousness, defensibility/excusability, and Item E2 .691
rejection). Item E5 .332
Item G5 .297
Item E9 .394 (min.)
4.3. Results and discussion Item R7 .669 (max.)
Item R10 .610
Internal reliability was calculated using Cronbach’s alpha (1951) Item R1 .583
and was found to be high (Table 1). The internal consistency of the Item E10 .580
Item E7 .553
12 questionnaire items was sufficient to provide an overall score
Item R4 .529
(standardized alpha: .88). All the items appear to be adequate since Item G10 .505
Cronbach’s alpha ranged from .74 to .91. The same applies to the Item E4 .498
inter-item consistency of each of the three categories. This finding Item E1 .480
Item G1 .466
also means that each of the subscores of the scale of social and
Item G4 .411
moral judgments can be taken into account. Item G7 .407 (min.)
Item G11 .754 (max.)
4.4. Effect of value type and subscale type Item E11 .744
Item E8 .698
Item G8 .692
A 3 × 3 ANOVA (type of value × type of subscale) with repeated Item R11 .591
measures was conducted using SPSS software. The overall anal- Item R8 .561
ysis highlighted the effect of the variable “type of value” Principal component analysis; rotation method: varimax with Kaiser normalization.
(F(2520) = 942.13; p < .001). The three types of values were found to The rotation converged in 8 iterations.
be clear and distinct, thus validating each group of items. Atypical
or idiosyncratic behaviors were judged more leniently than con-
ventional transgressions, while conventional transgressions were Based on the analysis of variance and the factor analysis, the
judged less severely than moral transgressions. three initial categories are validated. In other words, the three cat-
The results also show the effect of the “type of scale” vari- egories of items were clearly identified by the participants.
able (F(2520) = 326.75; p < .001), indicating that there were fewer
negative judgments on the “rejection” subscale than on the “defen- 4.6. Analysis of judgment scores
sibility/excusability” and “seriousness” subscales, regardless of the
category of value. There was no significant interaction between the As shown in Fig. 1, the data related to the judgment scores of
variables “type of value × type of subscale”. the entire sample population (all categories and subscales) have a
Gaussian distribution (normal distribution law).
4.5. Factor analysis

For each of the two databases, we performed a principal com- 100

ponent analysis with orthogonal varimax rotation. The analysis
revealed three factors corresponding to three types of predicted
80
values and explaining the whole variance. This was observed inde-
pendently of the subscales. The analysis highlights groups of scores
Frequency

solely determined by the nature of the value involved in the trans-

60
gression (Table 2).
The first factor corresponds to the items related exclusively to
“moral values” and accounts for 19.41% of the total variance, with 40
an eigenvalue of 6.99. Here, the factor loadings ranged from .39 to
.63. The second factor was perfectly compatible with the “personal
values” category and accounted for 11.62% of the variance, with an 20 Sigma = 25.93
eigenvalue of 4.18. The factor loadings were relatively high, ranging Mean = 121.7
from .41 to .67. Finally, as predicted, the third factor corresponded N = 521.00
0
to the “conventional values” category, accounting for 6.37% of the 0.0 20.0 40.0 60.0 80.0 100.0 120.0 140.0 160.0 180.0 200.0
total variance, with an eigenvalue of 2.29. Here, the factor loadings
ranged from .18 to .75. Fig. 1. Distribution of raw judgment scores. Subjects: general population.
 S. Abdellaoui et al. / Revue européenne de psychologie appliquée 66 (2016) 171–180 175

Table 3
Means and standard deviations of the scores for each category of values and subscales.

Types of values > subscale Personal Conventional Moral

V

Seriousness M = 21.3704 M = 10.91 M = 5.03

␴ = 3.3425 ␴ = 4.71 ␴ = 3.79
Defensibility M = 21.38 M = 10.61 M = 4.49
␴ = 3.49 ␴ = 4.81 ␴ = 3.70
Rejection M = 21.98 M = 16.64 M = 9.24
␴ = 3.42 ␴ = 5.92 ␴ = 6.09

120 80

100

60
80
Fréquence

Frequency
60
40

40

20
y
20 Sigma = 11.45
Mean = 18.8 Sigma = 13.21

0 N = 521.00 Mean = 38.2

0.0 10.0 20.0 30.0 40.0 50.0 60.0 0 N = 521.00
5.0 15.0 25.0 35.0 45.0 55.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0
5.0 15.0 25.0 35.0 45.0 55.0 65.0
Fig. 2. Moral transgression scores.
Fig. 4. Conventional transgression scores.
Based on an analysis of the judgment scores related to each cat-
egory, the scores for the moral category were found to tend in the
same direction in the case of moral values (Fig. 2). In all likelihood, range of judgments made by the same individual must be specified
the obligatory nature of moral values explains the levels of judg- (i.e. characterized). This can be done by calculating the difference
ment indicative of a lower relative threshold. This is illustrated by between the score related to the seriousness of a transgression
the fact that the normality of the distribution shifts toward the involving the most “binding” values (i.e. moral values) and the
“severity of judgments” pole. Fig. 3 also shows similar scores for score related to the rejection of a person behaving atypically or
atypical or idiosyncratic behaviors (tending toward leniency). This idiosyncratically (i.e. the least “unpleasant” scenario). In the gen-
finding is explained by a strong tendency toward consensus related eral population (n = 521), the difference ranged from −16.95 to 5.01.
to the personal and non-compromising nature of the actions and This indicator may be an explanatory factor, notably for examining
behaviors in question. the ability of a subject to distinguish between different types or
Finally, the scores related to conventional transgressions fol- cases of transgression.
lowed a Gaussian distribution (Fig. 4). Therefore, the scale may be
said to include transgressions that generate both extreme and more
4.7. Partial correlations between specific judgment scores
moderate judgments.
Table 3 shows that the severity of judgments varies significantly
The Bravais-Pearson correlation coefficient of the subscores was
depending on the category of values and subscale. Therefore, the
used to examine relationships. A significant relationship was found
between judgments of conventional transgressions and judgments
300
of moral transgressions (r = .56). However, despite being signif-
icant, there was a weak correlation between atypical behaviors
and judgments of moral transgressions (r = .16). This is probably
explained by the fact that the judgments in question relate to dif-
200 ferent objects since one of them relates to atypical or idiosyncratic
behaviors while the other relates to serious transgressions. How-
Frequency

ever, the relationship could be explained by a tendency to exhibit

severity in all three dimensions of judgment (i.e. seriousness, defen-
sibility/excusability and rejection) (Table 4).
100

Table 4
Sigma = 9.10 Correlations between the raw scores for each category of values.
Mean = 64.7
Scores by group of values Atypical Conventional
0 N = 521.00
behavior transgressions
0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0
Conventional transgressions .33a
5.0 15.0 25.0 35.0 45.0 55.0 65.0
Moral transgressions .16a .56a
a
Fig. 3. Idiosyncrasy scores (personal values). The correlation is significant at the 0.01 level (bilateral).
176 S. Abdellaoui et al. / Revue européenne de psychologie appliquée 66 (2016) 171–180
Table 5
Correlations between the raw scores for each subscale.
Scores by subscale Seriousness
Excusability/defensibility .79a
Rejection .52a
a between 31 and 45 years (19.17%), and 17 inmates aged over 45
The correlation is significant at the 0.01 level (bilateral).
4.8. Partial correlations between the subscale scores
The subscale scores were significantly correlated. Therefore, it
seems reasonable to group the subscale scores to obtain an over-
all raw judgment score or to construct an index. Although neither
of them explains the total variance (r = .79), the seriousness and
defensibility variables with similar scores could be merged. How-
ever, by keeping them separate, we will be in a better position to
isolate what relates to the act (seriousness and defensibility) from
what pertains to the person and their company (Table 5).
4.9. Discussion of the validation of the ABB scale
A statistical analysis of the results demonstrates the validity of
the overall scale. Given the relatively low number of items, the
validity of the tool used to measure social and moral judgments was
deemed to be acceptable. We may assume that the twelve items
measure the same dimension. In addition, the items relating to
the different categories of values were not ambiguous. The level of
internal consistency in each category was found to be satisfactory.
Based on Principal Component Analysis (PCA), the categories were
also found to be clearly distinct. The Principal Component Analysis
showed that the scale was homogeneous (see especially the first
factor). The second factor shows that the participants responded as
expected to the three categories of values. Overall, the judgments
were far more severe in the case of moral values than in the case
of conventional values and were also significantly more severe in
the case of conventional values than in the case of personal val-
ues. The results also show that, overall, the scales related to the
act of transgression (seriousness and defensibility/inexcusability)
differed significantly from the rejection scale (focused on the trans-
gressor).
As noted above, our aim was also to test the ABB scale among
a population directly concerned by transgression and the status of
the transgressor.
5. Use of the ABB scale among a prison population and
comparative analysis
5.1. Objectives
The idea of comparing the judgments of convicted offenders (or
alleged offenders, i.e. inmates awaiting trial) to the judgments of
non-offenders has attracted much interest from education profes-
sionals (specialized and non-specialized education), social workers
and therapists in recent years. A comparative analysis will increase
the generalizability of the scale, but may also provide a basis for
identifying specific characteristics in the types of social and moral
judgments made by inmates. We began by comparing the inmates’
judgments to those made by the general population for each of the
categories of values relating to the behaviors included in the scale
and subscales. The next stage involved determining the impact
of different criminal characteristics on the inmates’ judgments.
Finally, we compared different dimensions underlying the raw
judgments by calculating indices of tolerance and severity for each
group of subjects.
5.2. Sample population
Excusability/ One hundred and ninety-three male subjects were asked to
defensibility complete the questionnaire using the scale of social and moral
judgments, including 47 inmates aged under 18 years (24.35%), 92
.59a inmates aged between 18 and 30 years (47.66%), 37 inmates aged
years (8.81%). The mean age of the participants was 23.72 years.
Among them, 101 were recidivists, while 92 were first-time offend-
ers (non-recidivists). The sample included 79 defendants (inmates
awaiting trial) and 114 convicted offenders. The vast majority of
types of offenses were represented in the sample. We chose to take
into account the fact of having committed or not committed a crime
directly involving a sexual offense – an offense often perceived
as the most serious and intolerable type of crime among inmates.
Inmates who are convicted or alleged sexual offenders are often
subject to threats and even violence, both verbal and physical. The
assumption is that by including this variable we will be able to
measure the impact of belonging to a stigmatized category on the
perception of transgressions and those who commit them. There
were 32 inmates in the “sexual offender” category, including 16
convicted offenders and 18 inmates awaiting trial.
5.3. Comparative analysis of the two populations on the specific
dimensions of judgment
Regardless of the category of values involved in the behavior
or action, the analysis of variance showed that the “subscales” fac-
tor had the most significant effect, indicating that the judgments
made by the participants differed according to the type of subscale
(F(1792) = 645; p < .001). There was a significant interaction effect
between the type of subscale and the group to which the partici-
pants belonged (F(1792) = 109.251; p < .001). Comparative analysis
shows that the inmates and the other participants differed signifi-
cantly on the “seriousness” dimension (F(1792) = 3.58; p = .05), with
inmates tending to make less severe judgments in terms of serious-
ness (m = 35.87) compared to the other participants (m = 37.42). The
two populations also differed on the “defensibility/excusability”
dimension (F(1792) = 16.18; p < .001) since the inmates (m = 39.58)
were significantly more likely to be lenient than the other partic-
ipants (m = 36.57). By contrast, there was no significant difference
on the “rejection” dimension (F(1792) = 0.38 NS) (Fig. 5).
If we examine the scores from the point of view of the spe-
cific criminal characteristics of the prison population, we find
no significant differences among the inmates (based on the
ANOVA test). Except for the effect associated with belonging to
the category of inmates who had committed a sexual offense
(F(1.192) = 18.98; p < .001), the inmates imprisoned for a sexual
offense were more likely to be lenient (m = 49.40) compared to
other inmates (m = 43.93). However, there was an interaction effect
50
45
40
35
Degree 30
of 25
inmates
leniency 20
others
15
10
5
0
seriousness defensibility rejection
Type of subscale
Fig. 5. Means of raw judgment scores according to subscale and group membership.
 S. Abdellaoui et al. / Revue européenne de psychologie appliquée 66 (2016) 171–180
between the variables “sexual offense vs. non-sexual offense” and
“recidivist vs. first-time offender” (F(1192) = 4.61; p = .03). More
specifically, the results show that inmates imprisoned for crimes
other than sexual offenses were more likely to be lenient in terms of
defensibility/excusability (m = 48.60 compared to 37.40 for inmates
imprisoned for a sexual offense) if they were recidivists (m = 51.19).
5.4. Comparative analysis of the judgments of the two
populations according to the type of behavior
An ANOVA test was performed to compare the inmates’ judg-
ment scores with those of the general population, regardless of the
type of subscale involved in the judgment. We combined all the
scores in the three subscales for each of the three categories of
values. First, we may note that the raw judgments were statisti-
cally similar since the analysis showed no significant difference
(F(1713) = 1.67 NS) between the general population (m = 121.65,
EC = 25.92) and the prison population (m = 125.41, EC = 28.46).
Therefore, on the face of it, it is impossible to predict whether
one group will be more or less lenient than the other group simply
based on the fact of having committed (or not committed) a pun-
ishable offense. A significant interaction was also found between
membership group and type of value (F(1713) = 5.13; p = .02). More
specifically, the results show that the two populations made sig-
nificantly different judgments regardless of the type of value. Also,
results show that inmates tended to be more lenient in judg-
ing moral transgressions (F(1713) = 18.44 < .001) and conventional
transgressions (F(1713) = 6.71; p = .01). By contrast, compared to
non-inmates (F(1713) = 2.11; p < .001), inmates were significantly
less lenient in the case of atypical behaviors (Fig. 6).
Finally, we may note that in terms of raw judgment, the analysis
of variance showed no differences in the factors characterizing the
prison population. Regardless of whether they were recidivists or
first-time offenders, awaiting trial or convicted, or involved in a sex-
ual offense or another type of offense, the ANOVA test showed no
significant difference – hence the value of examining the relation-
ships between the indicators. Various indicators of tolerance (or
tolerance indices) were developed in order to highlight potential
differences or discrepancies between the judgment of transgres-
sions and the judgment of transgressors. These differences can be
assessed using a tolerance index for each category of values.
5.5. Dissociating the nature of the transgression from the
transgressor
Although the “scales*values” interaction is significant on
account of a largely unavoidable threshold effect, the three scales
70
60
50
40
Degree of 30
leniency
20
10
0
inmates others inmates others inmates others
Moral Conventional Atypical
Transgressions Transgressions behavior
Fig. 6. Means of the raw judgment scores according to the type of transgression and
group membership.
177
functioned almost identically in the three categories of values (i.e.
comparable severity and defensibility judgments and more mod-
erate rejection judgments). Regardless of membership group, the
hierarchy of the categories of values and subscales was found to
be similar. Therefore, we may consider processing the data by
combining them in such a way that the raw judgments highlight
dimensions that are more characteristic of the judgment processes.
The point is to examine the data in terms of tolerance or severity
based on a set of indicators. The aim is to develop characteristic
indicators of the dimensions involved in the dynamics of judgment.
These indicators should also serve to highlight the limited differ-
ences reflected by the interactions relating to the raw judgments.
The similarities between most of the raw judgments and the
similar results obtained for the two populations in the initial anal-
ysis suggest that a more refined analysis is needed. This will involve
incorporating indicators that capture the fundamental dimensions
of social and moral judgment. We will illustrate this approach by
examining the level of tolerance shown in the judgments made by
the general population and the prison population.
5.6. Tolerance index: definition and results
We define tolerance as showing leniency and not rejecting a
person who has committed a transgression or exhibited a behavior
(act of transgression) recognized as being serious or highly atyp-
ical. In this sense, tolerance does not equate to permissiveness
or a complete lack of judgment. This dimension of tolerance can
be measured using a simple indicator corresponding to the dif-
ference between the perceived seriousness of the transgression
on one hand and the perceived defensibility/non-excusability and
rejection of the transgressor on the other. Based on an assessment
of the act of transgression and two assessments of the transgres-
sor, the indexed scores were weighted by multiplying by two the
assessment of the seriousness of the transgression. We obtained
a tolerance index based on the following formula: 2 × G – (E + R).
The index varies around 0. A positive index indicates tolerance. An
index can be computed for each of the three values composing the
scale.
In terms of tolerance, the analysis of variance shows that there is
a significant difference between the two populations on the three
specific tolerance scores. The inmates were found to be less tol-
erant in their appraisals of atypical behaviors (F(1713) = 144.45;
p < .001), conventional transgressions (F(1713) = 80.69 p < .001) and
moral transgressions (F(1713) = 56.05; p < .001) (Fig. 7).
These results are particularly interesting since the two sample
populations differ significantly in terms of their raw seriousness,
0
-2
-4 inmates
Level of others
tolerance -6
-8
-10
-12
individual moral
conventional
Type of tolerance
index
Fig. 7. Means of the tolerance indices for each category of values and according to
group membership. The more negative the score, the greater the level of intolerance
shown toward acts involving one of the three categories of values.
178 S. Abdellaoui et al. / Revue européenne de psychologie appliquée 66 (2016) 171–180
0 0
-2 -1.99
-4 -3.94
Level of
-6
tolerance
-6.98
-8 -7.48
-10
-11.82 -11.15
-12
individual convention moral
category of values
Fig. 8. Means of indexed scores based on the category of values involved in the act
and according to the nature of the offense.
defensibility and rejection judgment scores. The inmates tended
to be more lenient, although it was impossible to establish the
differences between the various possible subscores. The results
show that we need to examine the general degree of leniency more
closely by distinguishing leniency toward the transgression and
leniency toward the transgressor. The tolerance index has the pecu-
liarity of providing an indication of the ability of a subject to assess
the seriousness of a transgression while informing us about how
the “judging” person perceived (i.e. judged) the transgressor.
Let us now consider the impact of the criminal characteris-
tics of the prison population. First, the variable “sexual offense”
was found to have a significant effect for all the tolerance indices.
Analysis shows that inmates who were alleged or convicted sex
offenders were more tolerant (n = 34). More specifically, compared
to other inmates, alleged or convicted sex offenders were found to
be more tolerant on the individual tolerance index (F(1192) = 16.18;
p < .001), the conventional tolerance index (F(1192) = 5.82; p = .02),
and the moral tolerance index (F(1192) = 13.25; p < .001). In addi-
tion, there was no significant interaction involving this variable,
indicating that the stigmatizing effect, from the point of view of
prison life (Abdellaoui, 1998), has a real impact on the appraisal and
differentiation of the transgression and the transgressor (Fig. 8).
In terms of the “recidivist vs. non-recidivist” variable, there was
no significant effect on the tolerance indices. Regardless of whether
the inmates were recidivists or non-recidivists, no difference was
found based on the indexed scores indicating the distinction made
by the inmates between the seriousness of the transgression and
the stigmatization of the transgressor. However, concerning the
“awaiting trial vs. sentenced” variable, the ANOVA test showed a
main effect, but only for moral tolerance (F(1192) = 3.87; p = .05).
Here, the fact of being in the “awaiting trial” category appeared to
have an impact on the perceived importance of separating what
relates to the transgression from what relates to the transgressor.
The inmates awaiting trial tended to be more intolerant of the most
serious types of transgression (i.e. those involving a transgression
of moral values) and of those who committed them (m = −9.49)
compared to convicted offenders (m = −5.60). More specifically,
there was a significant interaction between the “recidivist” and
“awaiting trial” factors, but again only on the moral tolerance index
(F(1192) = 7.04; p < .01). Comparative analysis indicates that recidi-
vists awaiting trial showed the greatest intolerance toward moral
transgressions. Therefore, repeat offenders (i.e. recidivists) await-
ing trial appear to have a tendency to amalgamate the seriousness of
the transgression and the undesirability of the transgressor (Fig. 9).
These findings show that simply judging acts of transgression
negatively with or without an appraisal of the transgressor is not
necessarily sufficient to determine the type of strategy involved in
the judgment-making process. It is easy to see that, broadly speak-
ing, inmates have a greater tendency to stigmatize and sanction any
act of transgression or transgressor. Self-presentation strategies,
or even avoidance of or differentiation from other transgressors,
-2
Level of -4 -3.77
tolerance -5.07
-6 on trial
of
Non-sexual -8 -7.43
moral convicted
Sexual
transgressions -10
-12
-12.92
-14
First-time Recidivists
offenders
Fig. 9. Means of indexed moral tolerance scores based on criminal past and the fact
of awaiting trial.
appear to play a central role in determining social and moral
judgments. The desire to be singled out (to which we might add
self-enhancement) is particularly strong among inmates subject to
stigmatization and rejection in prison. This is particularly true of
subjects imprisoned for sexual offenses, who face almost constant
threats. In some prisons, sex offenders are confined to a specific
area or floor, the assumption being that inmates involved in a case
of rape or attempted rape, pedophilia or child murder should not
come into contact with other inmates. Where permitted, exercise
will also tend to be located in a different area of the prison. As
we have shown, showing greater tolerance toward transgressions
could be explained by the need to distinguish what pertains to the
nature of the act from what pertains to the person who committed
the act.
Finally, the judgments made by non-recidivists awaiting trial
can also be viewed as the result of an identity strategy. Awaiting
trial is often an anxious and uncertain time. The fact of not know-
ing what might happen in the weeks, months or years ahead causes
inmates to adopt certain behaviors or beliefs designed to attenuate
the sense of powerlessness accentuated by imprisonment. Some
studies have shown that people awaiting a potentially unpleasant
event resort to different strategies to shape and change their fate
(Abdellaoui, 1998; Delhomme, 1987). These may be seen as strate-
gies designed to ward off bad luck, which tend to be more common
among people awaiting an unpleasant event than among people
who know their fate or who have been informed of a decision or
judgment that has been a long time coming. Here, the analysis of
the tolerance scores shows that defendants (i.e. inmates awaiting
trial) tend to be more intolerant of transgressors than sentenced
inmates, but are also more intolerant if they have been previously
convicted and imprisoned before receiving a second prison sen-
tence. Self-presentation strategies and attempts to distinguish and
single out their transgression (and therefore to distinguish them-
selves from other inmates) may be indicative of a desire not to be
associated with the general population of transgressors.
6. General conclusion
We have successfully developed a convenient, easy-to-use scale
of social and moral judgments applicable to both a general (ran-
dom) population and a population particularly concerned by moral
issues and adherence to social and moral rules. The greatest degree
of leniency was associated with the least serious transgressions and
those responsible for them, and vice versa. However, it appears
that while judgments of seriousness, defensibility and rejection
in the moral, conventional and personal sphere provide interest-
ing indications, it is the structure of these judgments (highlighted
by indicators such as specific tolerance indices) that provides the
most useful basis for analysis. Whether we consider the general
population or the prison population, the tool provides a number
of benefits for research and professional practice. For example, a
 S. Abdellaoui et al. / Revue européenne de psychologie appliquée 66 (2016) 171–180
better understanding of the relationship between perception of
values and transgressions (violence, fraud. . .) as they have been
particularly analyzed by Schenk and Fremouw (2012) or by Viens
(2007) about addiction problems. Integrated into a more compre-
hensive approach, it allows also the realization of psychosocial
profiles for educational or therapeutic purposes (Van der Giessen
et al., 2013). The proposed scale provides information about dif-
ferent social-cognitive behaviors and dispositions, such as the
coherence (or consistency) and discriminating nature of social and
moral judgments and the tendency to dissociate what pertains to
the person who committed the act from what pertains to the trans-
gression per se. Depending on the type of social and moral values,
the scale can also be used to identify those values that are most
resistant to strategies aimed at restoring self-image and at devalu-
ing anything that deviates from socially desirable behavior. There
are other possible indicators, such as the indicator of severity relat-
ing to each category of values. For example, the severity of a moral
judgment is measured by the formula 2 × M − (C + I). The point here
is to multiply by two the raw score of the judgments made about
moral transgression and to subtract the scores obtained for judg-
ments made about conventional transgression situations (C stands
for conventional) and those relating to atypical or idiosyncratic
behavior (I stands for individual). The tool can also be used to cal-
culate the relative weight of one of the specific tolerances indices in
relation to the overall tolerance index (i.e. including all the specific
tolerance indices). This may provide more information about the
importance of specific tolerance by dissociating it from the other
indices of moral tolerance.
Finally, the tool should not be seen as providing a basis for the
measurement of the maturity of moral judgment or as implying
a rigid determination of the principles governing the nature or
intensity of a judgment made about a transgression. Kupfersrnid
and Wonderly (1980) showed the difficulty and, in some cases, the
impossibility of demonstrating a relationship between the inter-
nalization of moral values and actual behavior. On this point, we
begin from the assumption that making social and moral judg-
ments is part of a social-cognitive strategy or project involving
a greater or lesser degree of conscious intention. This project or
strategy may be directly focused on the self and involve the fol-
lowing thoughts: “I want to be well thought of, I don’t want to
be seen in this way, I will show that what people think of me is
wrong”, etc. It may also focus on others and involve views such as
there is worse out there, those who judge me should look at them-
selves first, I’m surrounded by bad apples, etc. In other words, the
proposed tool must be seen, first and foremost, as an instrument
in which the presence of one or several factors can significantly
determine a given judgment. The results demonstrated the multi-
factorial nature of social and moral judgments and provide a basis
for conducting measurements in terms of the relationship between
what pertains to the judged object, the person judged, and the con-
text in which the object and the person are judged. However, it
is important not to underestimate the importance that the judging
subject gives to the act of judgment when asked to answer the ques-
tions included in the scale. Furthermore, using indices that take into
account the gap between what pertains to the act of transgression
and what pertains to the transgressor provides a new perspective,
notably on the results relating to the judgments made by inmates,
such as those observed in two studies conducted by Desombre,
Przygodski-Lionet, Durand-Delvigne, and Debosscher (2005) – the
first of which, unlike the second, showed that inmates were rel-
atively lenient toward transgressors. The calculation of indices
designed to reflect a specific relationship between two variables
can also be useful in cases involving socially desirable behavior,
self-enhancement and enhancement of one’s group or strategies
for warding off bad luck, which may indicate an attempt to control
one’s environment. The judgments made by subjects using this tool
179
are thus sensitive to the norm of internality. The norm of internality
tends to view individuals as being responsible for their behavior or
for what happens to them – i.e. as the “locus of control” (Beauvois,
1984, 1994). Several studies have shown a significant relation-
ship between social and moral judgments and internality scores
(Abdellaoui & Pittolo, 2000). The proposed scale may be useful for
research on modes of judgment and for understanding behavioral
strategies.
Disclosure of interest
The authors declare that they have no competing interest.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.erap.2015.07.002.
References
Abdellaoui, S. (1998). Stratégies sociocognitives du détenu : Recherche d’identité,
recherche de contrôle, avant et après jugement [Social-cognitive strategies inmate
search for identity, seeking control, before and after judgment]. Villeneuve d’Ascq:
Presses Universitaires du Septentrion.
Abdellaoui, S., Blatier, C., & Beauvois, J.-L. (1996). Dynamique socio-cognitives des
jugements de valeurs socio-morales [Socio-cognitive dynamics of judgments
socio-moral values]. In 1er Congrès international de psychologie sociale en langue
française ADRIPS, Montréal,
Abdellaoui, S., Kouabénan, R., & Gilibert, D. (1998). Acteurs du contrôle social et
explication des conduites déviantes et conformistes [Agents of social control
and explanation of deviant and conformist behavior]. Les Cahiers Internationaux
de Psychologie Sociale, 38, 46–69.
Abdellaoui, S., Personnaz, B., & Aubry, B. (2004). Message de tolérance, cohésion
d’équipe et situation de compétition [Message of tolerance, cohesion team and
competitive situation]. In XIIIe Congrès international de psychologie du travail
AIPTLF, Bologne.
Abdellaoui, S., & Pittolo, F. (2000). L’impact du statut d’expert sur la dynamique des
jugement socio-moraux en milieu hospitalier [Statutes impact (expert vs non-
expert) on the sociomoral judgments dynamic in hospital environment]. Revue
Internationale de Psychologie du Travail de Langue Française, 6(57), 511–521.
Basinger, K. S., Gibbs, J. C., & Fuller, D. (1995). Context and the measurement of moral
judgement. International Journal of Behavioral Development, 18(3), 537–556.
Baumrind, D. (1986). Sex differences in moral reasoning: Response to Walker’s
(1984) conclusion that there are none. Child Development, 57(2), 511–521.
Beauvois, J.-L. (1984). La psychologie quotidienne [The everyday psychology]. Paris:
Presses Universitaires de France.
Beauvois, J.-L. (1994). Traité de la servitude libérale. Analyse de la soumission [Treaty
liberal servitude. Analysis of the submission]. Paris: Dunod.
Cronbach, L. (1951). Coefficient alpha and the internal structure of tests. Psychome-
trika, 16, 297–334.
Delhomme, P. (1987). Attentes d’occurrences incontrôlables et stratégies
incantatoires : Conduites éthiques, choix d’un désagrément, jugements per-
ceptifs [Expectations uncontrollable occurrences and strategies incantatory: Ethical
conduct, choice of an inconvenience, perceptual judgments] Thèse de doctorat
[phd thesis]. Paris: University Paris VII.
Desombre, C., Przygodski-Lionet, N., Durand-Delvigne, A., & Debosscher, S. (2005).
Auteurs d’infractions et victimes d’agressions : Impact de la violence et du contexte
professionnel sur le jugement de punition [Offenders and victims of attacks: The
impact of violence and professional context on judgements of punishment]. Champ
Pénal. http://champpenal.revues.org/document333.html
Ebbesen, E. B., & Konecni, V. J. (1981). The process of sentencing adult felons. A
causal analysis of judicial decisions. In B. D. Sales (Ed.), Law and psychology (2)
(pp. 413–457). New York: Plenum Press.
Howe, E. S. (1994). Judged person dangerousness as weighted averaging. Journal of
Applied Social Psychology, 24(14), 1270–1290.
Killen, M., & Hart, D. (1999). Morality in everyday life: Developmental perspectives.
New York, NY: Cambridge University Press.
Kohlberg, L. (1969). Stage and sequence: The cognitive development approach to
socialization. In D. A. Goslin (Ed.), Handbook of socialization theory and research
(pp. 437–480). Chicago: Rand McNally and Compagny.
Kohlberg, L. (1971). Stages of moral development as a basis for moral education. In
C. M. Beck, B. S. Crittenden, & E. V. Sullivan (Eds.), Moral education (pp. 23–92).
Toronto: University of Toronto Press.
Kohlberg, L. (1983). Essays in moral development. Vol II: The psychology of moral
development. New York: Harper & Row.
Kupfersrnid, J. H., & Wonderly, D. M. (1980). Moral maturity and behavior: Failure
to find a link. Journal of Youth and Adolescence, 9(3), 249–261.
Kurtines, W. M., & Greif, E. (1974). The development of moral thought: Review and
evaluation of Kohlberg’s approach. Psychological Bulletin, 81, 453–470.
180 S. Abdellaoui et al. / Revue européenne de psychologie appliquée 66 (2016) 171–180
Mathys, C., Lanctôt, N., & Touchette, T. (2013). Validation d’une mesure de climat
de groupe chez des adolescentes en centre de réadaptation : Le vécu édu-
catif partagé au travers de trois dimensions essentielles [Construct validity and
confirmatory factor analysis of a group climate measure for girls in residen-
tial care service: Real life intervention through three major dimensions]. Revue
Européenne de Psychologie Appliquée [European Review of Appied Psychology],
63(2), 109–119.
Mazé, C., Finkelstein, R., & Quentin, M. (2004). Un jury sous influence : L’impact des
affects, du type d’expertise et des circonstances aggravantes sur l’activité déci-
sionnelle des jurés [Pressures over jury: Influence of feelings, type of expertise
and aggravating circumstances on jurors’ decision making processes]. Psycholo-
gie Française, 49(4), 357–372.
Morchain, P. (2009). Psychologie sociale des valeurs [Social psychology of values]. Paris:
Dunod.
Nemeth, C. J., & Sosis, R. H. (1990). A simulated jury study: Characteris-
tics of the defendant and the jurors. The Journal of Social Psychology, 90,
221–229.
Nucci, L., & Turiel, E. (1978). Social interactions and the development of social con-
cepts in presschool children. Child Development, 49, 400–407.
Pélissier, C., & Alidières, L. (2014). The role of language in prison: A “federating space
for reciprocal action” in educational support. Procedia – Social and Behavioral
Sciences, 116, 4086–4091.
Przygodzki, N., & Mullet, E. (1997). Moral judgment and aging. European Review of
Applied Psychology, 47(1), 15–21.
Przygodzki-Lionet, N., & Noël, Y. (2004). Individu dangereux et situations dan-
gereuses : Les représentations sociales de la dangerosité chez les citoyens, les
magistrats et les surveillants de prison [Dangerous individual and dangerous
situations: The social representations of dangerousness at citizens, magistrates
and supervisors of jail]. Psychologie Française, 49, 409–424.
View publication stats
Rest, J. (1979). Development in judging moral issues. Minneapolis: University of Min-
nesota Press.
Rest, J. (1986). Guide for the defining issues test. Center for the study of ethical devel-
opment, University of Minnesota Press.
Salini, R. (1976). Achievement and morality. Journal of Personality and Social
Psychology, 33, 327–337.
Schenk, A. M., & Fremouw, W. J. (2012). Individual characteristics related to prison
violence: A critical review of the literature review. Aggression and Violent Behav-
ior, 17(5), 430–442.
Shweder, R. A. (1981). Fact and artifact in trait perception: The systematic distor-
tion hypothesis. In B. A. Maher, & W. B. Maher (Eds.), Progress in experimental
personality research. New York: Academic Press.
Thurstone, L., & Chave, E. J. (1929). The measurement of attitude. Chicago: University
of Chicago Press.
Tisak, M., & Turiel, E. (1984). Children’s conceptions of moral and prudential rules.
Child Development, 55, 1030–1039.
Tostain, M. (2000). Psychologie morale et culture [Moral psychology and culture].
Grenoble: Presses universitaires, Coll. Vies Sociales.
Turiel, E. (1983). The development of social knowledge: Morality and convention.
Cambridge: Cambridge University Press.
Turiel, E., & Smetana, J. (1984). Social knowledge and action: The coordination of
domains. In W. M. Kurtine, & J. L. Gewirtz (Eds.), Morality, moral behavior and
moral development (pp. 261–282). New York: Wiley.
Van der Giessen, D., Branje, S., Overbeek, G., Frijns, T., van Lier, P. A. C., Koot, H. M.,
et al. (2013). Co-occurrence of aggressive behavior and depressive symptoms
in early adolescence: A longitudinal multi-informant study. European Review of
Applied Psychology, 63(4), 193–201.
Viens, D. M. (2007). Addiction, responsibility and moral psychology. American Journal
of Bioethics, 7(1), 17–19.