Psychiatry and Clinical Neurosciences 2009; 63: 478–485 doi:10.1111/j.1440-1819.2009.01996.

Regular Article

Correlations among self-esteem, aggression, adverse

childhood experiences and depression in inmates of a
female juvenile correctional facility in Japan pcn_1996 478..485

Naomi Matsuura, PhD,1,2* Toshiaki Hashimoto, MD, PhD3 and Motomi Toichi, MD, PhD4
1
School of Education, Tokyo University of Social Welfare, Tokyo, 2Research Center for Special Needs Education, Nara
University of Education, Nara, 3Department of Pediatrics, Tokushima Red Cross Hinomine Rehabilitation Center for the
Disabilities, Tokushima and 4School of Health Science, Kyoto University Faculty of Medicine, Kyoto, Japan

Aim: The purpose of this study was to examine the Results: Significant main effects (R2 = 0.50, P <
following hypothesis: (i) Self-esteem and aggressive- 0.001) on the depression score were observed in self-
ness, adverse childhood experiences (ACE), and a esteem (b = -0.41, P < 0.001) and aggression (b =
depressive tendency interact with each other. (ii) The 0.21, P < 0.05). Self-esteem, aggression, ACE, and
subjects show a strong depressive tendency, and each depression were found to be interrelated.
of the other factors exerts a main effect on, and inter-
Conclusion: Low self-esteem was also shown to exert
acts with, the depressive tendency.
marked effects on various factors. About half of the
Method: The subjects were 91 juveniles (all female) inmates of the facility were depressed, and the main
admitted to a female juvenile correctional facility effects of self-esteem, aggression, and the ACE score
between November 2005 and December 2006. They were shown to influence the depression score.
were aged 15–19 years, with a mean age of 17.0
(SD = 1.18) years. Self-esteem scale, aggression scale, Key words: adverse childhood experiences,
ACE questionnaire, and depression scale were aggression, depression, female correctional facility,
conducted. self-esteem.

HERE HAVE BEEN a number of studies on the behavior or overly high self-esteem promotes prob-
T correlations between self-esteem and antisocial
behavior.1–3 Donnellan et al.2 reported that low self-
lematic behavior is presently unclear.
Much research has been conducted on self-esteem
esteem in adolescence exerts negative effects on of juvenile delinquents.6–8 Research concerning the
behavior. In contrast, Baumeister et al.4 argued that low self-esteem hypothesis, in which low self-esteem
unrealistically high levels of self-esteem reflect narcis- is related to external problematic behavior, can be
sism and lead to social maladaptation. Such exagger- classified into three major categories.2 Rosenberg sug-
ated self-esteem reportedly leads to aggressiveness gested that low self-esteem weakens ties with society,
and antisocial behavior in some adolescents.5 Thus, and, consequently, promotes social maladaptation
whether low self-esteem is related to antisocial and delinquency. Hirschi9 also presented a similar
opinion by emphasizing social ties. Rogers explained
the mechanism of the appearance of problematic
behavior as follows: ‘Failure to develop positive self-
*Correspondence: Naomi Matsuura, PhD, 2-13-32, Marunouchi,
esteem leads to psychological problems and induces
Naka-ku, Nagoya-shi, 460-0002, Japan.
Email: matuuranaomi@yahoo.co.jp aggressive behavior’.10 Horney, who is considered to
Received 26 March 2008, revised 13 April 2009; accepted 23 April be a Neo-Freudian psychologist, asserted that an infe-
2009. riority complex and a sense of humiliation (including

478 © 2009 The Authors

Journal compilation © 2009 Japanese Society of Psychiatry and Neurology
Psychiatry and Clinical Neurosciences 2009; 63: 478–485 Depressive tendency of female offenders 479

low self-esteem) enhance aggressiveness and lead exhibit problems such as depression and suicide
people toward antisocial behavior.11 While the low attempts during and after adolescence. In girls, par-
self-esteem hypothesis is widely accepted, its inter- ticularly, it has been reported that more than 50% of
pretation varies, and the direction of the cause–effect those who experienced four or more ACE developed
relationship lacks sound evidence.1 depression, and that about 20% attempted suicide.28
There are also reports that the self-esteem of juvenile In Japan, Matsuura performed a questionnaire
delinquents is not low, and that the low self-esteem survey in inmates of juvenile correctional facilities
hypothesis, the controversy over which still remains, and a control group using the Japanese translation of
is still unable to explain delinquency or aggressive a simplified version of the ACE questionnaire, and
behavior. It has been supported, but at the same time compared their responses. As expected, sharp differ-
has been suggested to lack scientific evidence.2 ences were observed, suggesting the seriousness of
Matsuura et al., who investigated juvenile correc- ACE that inmates had. Furthermore, a survey at mul-
tional facilities in Japan, compared the inmates with tiple facilities in Japan revealed common features of
age- and gender-matched controls, and demonstrated ACE in inmates.14
the very low self-esteem of inmates.12 They simul- It has also been suggested by many studies that
taneously disclosed some common features and juvenile delinquents often suffer from symptoms of
characteristics in inmates through the investigation anxiety and depression. Depression and behavioral
of multiple facilities. They included developmen- disorders are more concurrent than expected, and
tal problems such as hyperactivity and impul- their relationship is so strong that it must not be
siveness, adverse childhood experiences (abuse overlooked.29–31 Angold et al. performed meta-
and inappropriate child-raising), and moderate analysis concerning the concurrence of depression
aggressiveness.13–15 Their studies also revealed and behavioral disorders in community samples of
marked internal problems such as anxiety, depres- average children, and showed that the risk of the
sion, and social withdrawal, as well as external behav- occurrence of psychiatric problems in adolescence
ioral problems such as aggressive behavior and was about 6.6 times higher (95%CI: 4.4–11.0) in
delinquency.16 those who had both conditions than in those who
There is also extensive literature concerning the did not.32 As behavioral and depressive disorders
relationship between delinquency and inappropriate innately tend to concur, some researchers have sug-
child-raising (including abuse).17–19 Abuse is consid- gested that it may be more reasonable to regard the
ered to have particularly strong effects on the occur- concurrence of both symptoms as a single diagnostic
rence of aggressive behavior, and the concurrence of category.32
serious abuse and inappropriate child-raising has In a survey of male inmates of juvenile correctional
been shown to increase the seriousness of behavioral facilities in Japan, marked depression was also noted
and emotional problems.20,21 The Adverse Childhood in about 50% of the inmates.16 While characteristics
Experiences study (ACE study) by Felliti et al.22 is par- among the inmates such as low self-esteem, increased
ticularly informative. Anda et al.23 reported that ACE aggressive behavior and serious ACE have thus been
such as child abuse and the inadequacy of the child- disclosed gradually, evaluation of their correlation
raising function of the family lead to various health- remains insufficient. Although many factors are
related problems in adulthood. These ACE studies considered to interact in a complex way to induce
were planned primarily by the Centers for Disease psychiatric disorders, investigations themselves have
Control and Prevention (CDC) and were conducted been limited in Japan, and a standardized question-
with regard to the relationship between past ACE and naire has not been developed, hampering a clear
the present state of health in more than 17 000 understanding or analysis of the situation.
adults. New findings that ACE enhance delinquency However, scientific studies to date have suggested
and dangerous behavior such as crime and drug that inmates of juvenile correctional facilities exhibit
abuse have also been reported.24–26 A clear quantita- serious internal as well as external problems. In par-
tive relationship between ACE and external problems ticular, close correlations are considered to be present
as well as their relationships with self-esteem and between self-esteem and aggressive behavior and
depression have also been clarified.22,27 Thus, it has between abusive experience and a depressive ten-
been shown with scientific evidence that children dency. Girls have been reported to respond more
exposed to an adverse environment are more likely to sensitively to stressful life events and to be more

© 2009 The Authors

Journal compilation © 2009 Japanese Society of Psychiatry and Neurology
480 N. Matsuura et al. Psychiatry and Clinical Neurosciences 2009; 63: 478–485

likely to exhibit a depressive tendency.33 In this study,

we conducted a questionnaire survey in female Japanese standardized version of the Buss-Perry
inmates of juvenile correctional facilities, and exam- Aggression Questionnaire
ined the following hypothesis: A Japanese standardized version36 of the Buss-Perry
(i) self-esteem and aggressiveness, ACE, and a Aggression Questionnaire (BAQ)37was used. This
depressive tendency interact with each other; and (ii) questionnaire consists of four scales of verbal aggres-
the subjects show a strong depressive tendency, and sion, physical aggression, hostility, and anger. There
each of the other factors exerts a main effect on, and is no consensus as to a precise definition of the
interacts with, the depressive tendency. concept of aggression, but this questionnaire is con-
sidered to allow comparison of internal aggressive-
ness between subjects and controls.
METHODS Buss et al.37confirmed the high internal consistency
and reliability of their Aggression Questionnaire.
Subjects This questionnaire was converted by Ando et al. into
The subjects were 91 juveniles (all female) admitted scales with internal consistency and stability, and its
to a female juvenile correctional facility between validity has also been established.36 The mean score
November 2005 and December 2006. They were aged of the subjects using this scale was 79.2 (SD = 15.56),
15–19 years, with a mean age of 17.0 (SD = 1.18) and Cronbach’s a = 0.87.
years. At this facility, the following questionnaire
Adverse Childhood Experiences questionnaire
was completed by all inmates. Eighty-one inmates
(ACE questionnaire)
(89.0%) provided analyzable answers to all the ques-
tions and were included in the analysis. The ACE Study was conducted primarily by the
The percent distribution of delinquency type com- Health Insurance Union of the United States and the
mitted by juveniles admitted to this facility was as CDC on childhood abuse and state of health later in
follows: drug-related crime (41.7%), larceny life based on answers to questionnaires by 17 737
(15.4%), injury/violence (10.0%), others (31.9). The affiliates to the above union.22
juveniles admitted to Japanese correctional facilities The ACE Study emphasizes the number rather than
are considered to correspond to ‘severe juvenile categories of ACE that one has been subjected to. The
delinquents’ or ‘serious juvenile offenders’ rather number of the nine items of ACE (Table 1) that one
than ‘delinquents’ in Western research papers. has undergone (0–9) is regarded as the ACE score.
The mean ACE score in our subjects was 1.94
(SD = 1.96), and a = 0.69. The range was 0–9.
Questionnaire
Rosenberg Self-Esteem Scale Table 1. Adverse Childhood Experiences questionnaire items
The 10-item self-esteem scale prepared by Rosen-
Growing up (prior to age 18) in a household with:
berg34 and translated into Japanese by Yamamoto 1. Recurrent physical abuse, e.g. pushing, grabbing,
et al.35 was used. Rosenberg considered self-respect slapping, kicking.
and positive evaluation of the self rather than a 2. Recurrent emotional abuse, e.g. you were called things
feeling of superiority or inferiority in comparison like ‘lazy’ or ‘ugly’.
with others to be self-esteem. He also considered that 3. Sexual abuse.
one has high self-esteem when one feels oneself 4. An alcohol or drug abuser.
‘good enough’ rather than ‘very good’. Low self- 5. Mother being treated violently.
esteem means self-rejection, lack of self-approval, 6. Someone who is chronically depressed, suicidal,
self-contempt, and the absence of self-respect. In this institutionalized or mentally ill.
7. One or no biological parents.
study, we used the term ‘self-esteem’ in such a sense.
8. An incarcerated household member.
Answers to questions were obtained using a 5-point
9. Neglected by parents, e.g. didn’t have enough to eat, had
scale of: ‘true’ (5 points), ‘more true than false’ (4), to wear dirty clothes.
‘neither true nor false’ (3), ‘more false than true’ (2),
yes = 1, no = 0. The sum is the Adverse Childhood
and ‘false’ (1). The mean score in our subjects was
Experiences (ACE) score.
27.9 (SD = 6.82), and a = 0.79.

© 2009 The Authors

Journal compilation © 2009 Japanese Society of Psychiatry and Neurology
Psychiatry and Clinical Neurosciences 2009; 63: 478–485 Depressive tendency of female offenders 481

explained to the inmates, and their consent is

Japanese standardized version of the Birleson obtained.
Depression Self-Rating Scale for Children In this study, tests or questionnaires other than
(DSRS-C) the DSRS-C were performed within a few days
In this study, the Birleson Depression Self-Rating after admission. The DSRS-C was conducted about
Scale for Children (DSRS-C)38was used. The DSRS-C 1 month after admission in consideration of the
consists of 18 questions for the evaluation of depres- psychological effects of the decision to send the
sion in children designed to be answered by children adolescents to juvenile correctional facilities. All
themselves about their state during the past week questionnaires were handed to the inmates individu-
using a 3-point scale (2, 1, and 0), with a full score of ally after explanation by an instructor at the facility
36. Its Japanese version prepared by Murata has been and recovered after completion.
confirmed to be reliable and valid.39 Recently, Denda
et al.40 performed a large-scale survey in elementary
and junior high school children in Hokkaido using Statistical analyses
this questionnaire. For statistical analyses, Pearson’s correlation analysis,
Birleson et al.41 determined the cut-off of the 2-factor ANOVA, and multiple regression analysis were
DSRS-C as 15, but Murata et al.39 reported that 16 performed. The level of significance was <0.05. The
was appropriate by applying the Japanese version of statistical package, SPSS 13.0J for Windows, was
the DSRS-C to Japanese children and adolescents. used.
They also set the cut-off as 16 in their recent study.40
Therefore, we also set it as 16, and classified the
subjects with a DSRS-C score of ⱖ16 as a ‘being
RESULTS
depressed’ group. Thus, DSRS-C was used as an Results of the ACE questionnaire revealed that physi-
assessment instrument for screening depression, cal, psychological, and sexual abuse had been expe-
because we did not evaluate the symptoms using a rienced by 27.2%, 23.8%, and 8.6% of the subjects,
structured interview. respectively. The inmates of the correctional facility
According to Birleson’s report, the DSRS-C is were asked questions regarding their household envi-
designed to be applied to children aged 7–13 years,38 ronment: 17.3% of the inmates indicated that ‘there
but the scale was subsequently reported to also be was alcohol abuse’; 22.2 % indicated that there was
applicable to adolescents.42,43 We selected the DSRS-C ‘someone who was chronically depressed, mentally
because it is simple and easy to answer, and some of ill, institutionalized, or suicidal’; and more than 45%
our subjects were junior high school students. The indicated that ‘either or both of the parents were
mean score of our subjects was 15.53 (SD = 6.57), missing’. A total of 12.3% of the inmates indicated
and a = 0.77. that ‘a family member was imprisoned’, and 9.9%
stated ‘I was neglected by my parents (was not
allowed to go to school, not given regular meals
Informed consent etc.)’. These results suggest that family functioning in
The juvenile correctional facility performs multiple their families had often disintegrated.
questionnaires and screening tests in all inmates at The ACE score was 4 or higher, indicating a serious
the time of their admission in order to understand condition, in 28.4% of the subjects.
and evaluate their psychological, cognitive, and The mean score of the DSRS-C was 15.5
behavioral characteristics. The above questionnaires (SD = 6.57), and the score was equal to or above the
are some of them. The results are used for the plan- cut-off point (= 16) in 39 inmates (48.1%). There-
ning of individual programs for correctional educa- fore, about half of the inmates were classified into the
tion (similar to ‘Individual Educational Plan’) and ‘being depressed’ group.
understanding of inmates. The implementation of Table 2 shows the results of correlation analysis
effective correctional education on the basis of the among various factors. Self-esteem assessed with the
understanding of inmates’ characteristics and statisti- Rosenberg Self-Esteem Scale had significant negative
cal analyses and use of the test results for research are correlations with aggression assessed with the BAQ,
sufficiently explained at parents’ meetings. Also, at the ACE score and depression assessed with the
the administration of the tests, their objectives are DSRS-C. However, aggression had significant positive

© 2009 The Authors

Journal compilation © 2009 Japanese Society of Psychiatry and Neurology
482 N. Matsuura et al. Psychiatry and Clinical Neurosciences 2009; 63: 478–485

Table 2. Correlations among self-esteem, aggression, and interaction as independent variables using a 1-way
depression interaction model. In this model, a significant nega-
tive main effect was observed in self-esteem. The mul-
1 2 3 4
tiple regression coefficient of this model significantly
1 Self-esteem – indicated a goodness of fit.
2 Aggression -0.291*** –
3 Adverse Childhood -0.240* 0.353** –
Experiences score DISCUSSION
4 Depression -0.469*** 0.329** 0.271* –

*P < 0.05; **P < 0.01; ***P < 0.001. Correlations among factors
As expected, self-esteem was negatively correlated
with the other three factors (aggression, ACE score,
correlations with the ACE score and depression. The and depression score). As self-esteem showed particu-
ACE score also had a significant positive correlation larly strong correlations with aggression and the
with depression. depression score, they are considered to be exerting
As correlation analysis identified significant corre- direct effects with each other.
lations among factors, the main effect and interac- The ACE score was also suggested to significantly
tions of each factor were evaluated to examine the affect aggression rather than self-esteem or the
hypothesis. depression score. This is considered to be in agree-
Table 3 shows the results of multiple regression ment with the findings of a previous study that abuse
analysis performed by selecting the depression score and inadequate child-raising promote aggressive
as the dependent variable and the other factors as behavior in children.44
independent variables. The main effects of self-
esteem and aggression on the dependent variable
were evaluated in the Main effect model (A). As a
Multiple regression analysis and evaluation
result, significant main effects on the depression
of models
score were observed in self-esteem and aggression. Multiple regression analysis revealed that each of
Next, the main effects of self-esteem, aggression, the three models showed a strong goodness of fit.
and the ACE score were evaluated using the Main In particular, the coefficient of determination was
effect model (B). As a result, self-esteem had a signifi- highest for the Main effect model (B), showing that
cant negative main effect on the depression score, but the main effects of self-esteem, aggression, and ACE
the other two factors showed no significant effect. score affect the depression score (interaction between
The multiple regression coefficient of this model self-esteem and aggression was eliminated). The coef-
indicated its high-level validity. ficient of determination was also 0.50 when only
Next, multiple regression analysis was performed self-esteem and aggression were used as independent
by selecting self-esteem and aggression and their variables. As it was also 0.50 in the 1-way interaction

Table 3. Self-esteem, aggression, and ACE score as predictors of depression

Main effect model (A) Main effect model (B) 1-way interaction model

Self-esteem -0.41*** -0.39*** -0.37**

Aggression 0.21* 0.18 0.20
Adverse Childhood Experiences (ACE) score 0.12
Self esteem ¥ aggression 0.20
R2 0.51*** 0.59*** 0.50***

*P < 0.05; **P < 0.01; ***P < 0.001.

Standardized beta weights (b) are shown.
R2 = adjusted multiple correlation coefficient.
ACE, Adverse Childhood Experiences.

© 2009 The Authors

Journal compilation © 2009 Japanese Society of Psychiatry and Neurology
Psychiatry and Clinical Neurosciences 2009; 63: 478–485 Depressive tendency of female offenders 483

model in which the depression score was explained that 27.2%, 23.8%, and 8.6% had suffered physical,
by self-esteem, aggression, and their interaction, self- psychological, and sexual abuses, respectively. Mats-
esteem was suggested to be an important factor in the uura et al. surveyed about 350 average high school
depressive tendency. Moreover, self-esteem was the students using a similar questionnaire, and reported
factor that most affected the depressive tendency in that the percentages of those who had experienced
all three models. Therefore, the low self-esteem in these abuse types were 0–1.2%.14
subjects of this study is considered to have exerted
very serious effects on the depressive tendency.
Trzesniewski et al.1 showed that low self-esteem
Depression group
develops into not only delinquency but also diverse Denda et al. performed the DSRS-C in about 3300
problematic behavior and health problems during elementary school children and junior high school
and after adolescence by a cohort study in Dunedin, students, and reported that 22.8% were depressed,
New Zealand. They also considered that the low self- with a cut-off of 16.40 They also reported significant
esteem hypothesis was partially, but not totally, sup- increases in the score with age. The result of this
ported by their results. We studied the effects of low study, that 48.1% of the subjects with a mean age of
self-esteem on depression, and the results supported about 17 years were depressed, is considered to indi-
the low self-esteem hypothesis. However, the corre- cate their serious depressive tendency. Depression
lation between self-esteem and deviance behaviors was shown to be very prevalent in female as well as
was complicated and intricate interactions among male juvenile correctional facilities.48,49
multiple factors were not fully evaluated. Some epidemiological surveys have demonstrated
On the basis of the data obtained in this study, we that behavioral and depressive disorders are likely
must continue to advance scientific research. to concur,31 but their cause–effect relationship is
unclear. The diagnosis of their concurrence with, and
their differential diagnosis from, other diseases or
Characteristics of subjects and effects of ACE disorders are considered to be difficult due to the
The subjects were female inmates aged 14–20 years diversity of the problems.30
who were charged with delinquency and admitted to
the juvenile correctional facility. Treatments of juve-
nile delinquents are determined through a few com-
Limitations
plicated processes. Juveniles who have eventually There were a few limitations and problems in this
been sentenced by the family court to be admitted to study. One was that all questionnaires performed
a correctional facility are sent to various facilities were self-administered. Self-administered question-
depending on their age, level of criminal tendency, naires have the disadvantage of a high possibility of
and physical and mental conditions. One out of bias but are reliable in that information known only
about every 8000 girls of the same age is sent to a to the respondents can be obtained, and informa-
juvenile correctional facility;45 female minors who tion can be collected at a low cost.50–52 For the
have committed relatively serious acts of delinquency future, self-administered questionnaires should be
and are in a grave situation are considered for admis- combined with objective evaluations by a third
sion to the facility. party.
According to a large-scale survey of those admitted The second limitation was the sample size. In this
to juvenile classification offices by Kondo et al., 56% study, the sample number was considerably deficient
exhibited behavioral disorders.46 Of the juveniles for statistical analyses. A sample number at least a
admitted to classification offices, those who have few times larger would have been reasonable for the
committed serious acts of delinquency or serious evaluation of multiple cause–effect models. While
crime are sent to correctional facilities. Therefore, the about 100 juveniles are admitted to the juvenile cor-
percentage of those with behavioral disorders among rectional facility every year, it is not easy to obtain
those sent to correctional facilities is considerably several hundred samples. There are considerable bar-
high.47 riers against investigations and research because of
There are also serious problems regarding the the nature of juvenile correctional facilities. These
child-raising environment. In the subjects of this problems should be overcome one by one, and data
study, the result of the ACE questionnaire showed should be collected and analyzed steadily.

© 2009 The Authors

Journal compilation © 2009 Japanese Society of Psychiatry and Neurology
484 N. Matsuura et al. Psychiatry and Clinical Neurosciences 2009; 63: 478–485

The third limitation involved problems with the 11 Horney K. Neorsis and Human Growth. Norton, New York,
control group. However, characteristics of the sub- 1950.
jects are considered to be revealed more clearly by 12 Matsuura N, Hashimoto T, Uno S et al. Psychological traits
performing the same questionnaires in a control of youth in training school (reformatories): From the
viewpoint of mild developmental disorders such as LD
group matched with regard to age, gender, and local
and AD/HD. Jpn. J. Learn. Disabil. 2005; 14: 83–92 (in
characteristics, and comparing the results. Compari- Japanese).
son with the results in male juvenile correctional 13 Matsuura N, Hashimoto T, Toichi M. Investigation using a
facilities is also considered to make evaluation of the LD, AD/HD screening test and Adverse Childhood Experi-
effects of gender possible. ences (ACE) questionnaire in training school (reforma-
Our hypotheses were examined and we summa- tory): Risk factors of juvenile delinquency from the
rized our results as follows: (i) self-esteem and aggres- standpoint of developmental psychopathology. Jpn. J.
siveness, ACE, and a depressive tendency negatively Child Adolesc. Psychiatry 2007; 48: 583–598 (in Japanese).
interacted with each other strongly. (ii) The subjects 14 Matsuura N. Examination of relationships between juve-
showed a strong depressive tendency, and the main nile delinquency and family adversity – Empirical research
effect of low self-esteem on depressive tendency was about Adverse Childhood Experiences (child abuse
included) in juvenile training schools. Jpn. J. Soc. Probl.
confirmed significantly in all models in the multiple
2006; 21: 73–86 (in Japanese).
regression analysis. 15 Matsuura N, Hashimoto T, Toichi M. Cognitive traits of
Multiple regression analysis also indicated that students in juvenile delinquent training school. Jpn. J.
some models show a strong goodness of fit. The find- Learn. Disabil. 2007; 16: 95–105 (in Japanese).
ings of this study are expected to contribute to the 16 Matsuura N, Hashimoto T, Toichi M. Examination of behav-
future development of research in related fields. ioral and emotional disturbances of inmates of correctional
facility – comparisons with the normative high school stu-
dents. Jpn. J. Spec. Educ. 2008; 46: 215–222 (in Japanese).
REFERENCES 17 Jonson-Reid M. Child welfare services and delinquency:
1 Trzesniewski KH, Donnellan MB, Moffitt TE et al. Low self- The need to know more. Child Welfare 2004; 83: 157–173.
esteem during adolescence predicts poor health, criminal 18 Grogan-Kaylor A, Otis MD. The effect of childhood mal-
behavior, and limited economic prospects during adult- treatment on adult criminality: A tobit regression analysis.
hood. Dev. Psychol. 2006; 42: 381–390. Child Maltreat. 2003; 8: 129–137.
2 Donnellan MB, Trzesniewski KH, Robins RW et al. Low 19 Dykman RA, McPherson B, Ackerman PT et al. Internaliz-
self-esteem is related to aggression, antisocial behavior, ing and externalizing characteristics of sexually and/or
and delinquency. Psychol. Sci. 2005; 16: 328–335. physically abused children. Integr. Physiol. Behav. Sci. 1997;
3 Trzesniewski KH, Donnellan MB, Robins RW. Stability of 32: 62–74.
self-esteem across the life span. J. Pers. Soc. Psychol. 2003; 20 Bergman AJ, Walker E. The relationship between cognitive
84: 205–220. functions and behavioral deviance in children at risk for
4 Baumeister RF, Smart L, Boden JM. Relation of threatened psychopathology. J. Child Psychol. Psychiatry 1995; 36:
egotism to violence and aggression: The dark side of high 265–278.
self-esteem. Psychol. Rev. 1996; 103: 5–33. 21 Spaccarelli S, Coatsworth JD, Bowden BS. Exposure to
5 Bushman BJ, Baumeister RF. Threatened egotism, narcis- serious family violence among incarcerated boys: Its asso-
sism, self-esteem, and direct and displaced aggression: ciation with violent offending and potential mediating
Does self-love or self-hate lead to violence? J. Pers. Soc. variables. Violence Vict. 1995; 10: 163–182.
Psychol. 1998; 75: 219–229. 22 Felitti VJ, Anda RF, Nordenberg D et al. Relationship of
6 Jang SJ, Thornberry TP. Self-esteem, delinquent peers, and childhood abuse and household dysfunction to many of
delinquency: A test of the self-enhancement thesis. Am. J. the leading causes of death in adults. The Adverse Child-
Sociol. 1998; 63: 586–598. hood Experiences (ACE) Study. Am. J. Prev. Med. 1998; 14:
7 McCarthy JD, Hoge DR. The Dynamics of delinquency and 245–258.
self-esteem. Am. J. Sociol. 1984; 90: 396–410. 23 Anda RF, Felitti VJ, Bremner JD et al. The enduring effects
8 Owens TJ. Two dimensions of self-esteem: Reciprocal of abuse and related adverse experiences in childhood: A
effects of positive self-worth and self-deprecation on ado- convergence of evidence from neurobiology and epidemi-
lescent problems. Am. Sociol. Rev. 1994; 59: 391–407. ology. Eur. Arch. Psychiatry Clin. Neurosci. 2006; 256: 174–
9 Hirschi T. Causes of Delinquency. University of California 186.
Press, Berkeley, CA, 1968. 24 Whitfield CL, Dube SR, Felitti VJ et al. Adverse childhood
10 Rogers CR. On Becoming a Person. Princeton University experiences and hallucinations. Child Abuse Negl. 2005; 29:
Press, Princeton, NJ, 1961. 797–810.

© 2009 The Authors

Journal compilation © 2009 Japanese Society of Psychiatry and Neurology
Psychiatry and Clinical Neurosciences 2009; 63: 478–485 Depressive tendency of female offenders 485

25 Felitti VJ. The relationship of adverse childhood experi- the Birleson Depression Self-Rating Scale for Children
ences to adult health: Turning gold into lead. Z. Psychosom. (DSRS-C). Jpn. J. Child Adolesc. Psychiatry 2004; 45: 424–
Med. Psychother. 2002; 48: 359–369. 436 (in Japanese).
26 Anda RF, Croft JB, Felitti VJ et al. Adverse childhood expe- 41 Birleson P, Hudson I, Buchanan DG et al. Clinical evalua-
riences and smoking during adolescence and adulthood. tion of a self-rating scale for depressive disorder in child-
JAMA 1999; 282: 1652–1658. hood (Depression Self-Rating Scale). J. Child Psychol.
27 Foege WH. Adverse childhood experiences. A public health Psychiatry 1987; 28: 43–60.
perspective. Am. J. Prev. Med. 1998; 14: 354–355. 42 Ivarsson T, Gillberg C. Depressive symptoms in Swedish
28 Anda RF, Brown DW, Felitti VJ et al. Adverse childhood adolescents: Normative data using the Birleson Depression
experiences and prescribed psychotropic medications in Self-Rating Scale (DSRS). J. Affect. Disord. 1997; 42: 59–
adults. Am. J. Prev. Med. 2007; 32: 389–394. 68.
29 Loney BR, Lima EN, Butler MA. Trait affectivity and 43 Firth MA, Chaplin L. Research note: The use of the Birleson
non-referred adolescent conduct problems. J. Clin. Child Depression Scale with a non-clinical sample of boys. J.
Adolesc. Psychol. 2006; 35: 329–336. Child Psychol. Psychiatry 1987; 28: 79–85.
30 Dadds MR, Sanders MR, Morrison M et al. Childhood 44 Herrera VM, McCloskey LA. Sexual abuse, family violence,
depression and conduct disorder: II. An analysis of family and female delinquency: Findings from a longitudinal
interaction patterns in the home. J. Abnorm. Psychol. 1992; study. Violence Vict. 2003; 18: 319–334.
101: 505–513. 45 Research and Training Institute of the Ministry of Justice.
31 Sanders MR, Dadds MR, Johnston BM et al. Childhood White paper on crime 2005. Juvenile Delinquency, Tokyo,
depression and conduct disorder: I. Behavioral, affective, 2005.
and cognitive aspects of family problem-solving interac- 46 Kondou H, Ouhashi H, Fuchigami Y. An actual condition
tions. J. Abnorm. Psychol. 1992; 101: 495–504. Survey on Conduct Disorder in correctional facility. Jpn. J.
32 Angold A, Costello EJ, Erkanli A. Comorbidity. J. Child Correct. Med. 2005; 53: 1–11 (in Japanese).
Psychol. Psychiatry 1999; 40: 57–87. 47 Kondou H, Ouhashi H, Fuchigami Y. A study on subtype of
33 Putnins AL. Substance use among young offenders: Thrills, Conduct Disorder. Jpn. J. Correct. Med. 2005; 53: 12–27 (in
bad feelings, or bad behavior? Subst. Use Misuse 2006; 41: Japanese).
415–422. 48 Penn JV, Esposito CL, Schaeffer LE et al. Suicide attempts
34 Rosenberg M. Society and Adolescent Self-image. Princeton and self-mutilative behavior in a juvenile correctional
University, Princeton, NJ, 1965. facility. J. Am. Acad. Child. Adolesc. Psychiatry 2003; 42:
35 Yamamoto M, Matsui Y, Yamanari Y. The structure of per- 762–769.
ceived aspects of self. Jpn. J. Educ. Psychol. 1982; 30: 64–68 49 Keaveny ME, Zauszniewski JA. Life events and psychologi-
(in Japanese). cal well-being in women sentenced to prison. Issues Ment.
36 Ando A, Soga S, Yamazaki K et al. Development of the Health Nurs. 1999; 20: 73–89.
Japanese Version of the Buss-Perry Aggression Ques- 50 Breuk RE, Clauser CA, Stams GJ, Slot NW, Doreleijers TA.
tionnaire (BAQ). Jpn. J. Psychol. 1999; 70: 384–392 (in The validity of questionnaire self-report of psychopathol-
Japanese). ogy and parent–child relationship quality in juvenile
37 Buss AH, Perry M. The aggression questionnaire. J. Pers. delinquents with psychiatric disorders. J. Adolesc. 2007; 30:
Soc. Psychol. 1992; 63: 452–459. 761–771.
38 Birleson P. The validity of depressive disorder in childhood 51 Dolan MC, Rennie CE. The relationship between psycho-
and the development of a self-rating scale: A research pathic traits measured by the Youth Psychopathic trait
report. J. Child Psychol. Psychiatry 1981; 22: 73–88. Inventory and psychopathology in a UK sample of conduct
39 Murata T. Examination of Birleson Depression Scale for disordered boys. J. Adolesc. 2007; 30: 601–611.
school children in Japan. Jpn. J. Latest Psychiatry 1996; 1: 52 Murad SD, Joung IM, Verhulst FC et al. Determinants of
131–138 (in Japanese). self-reported emotional and behavioral problems in
40 Denda K, Koga Y, Sasaki Y et al. Depressive Symptoms Turkish immigrant adolescents aged 11–18. Soc. Psychiatry
in a School Sample of Children and Adolescents; Using Psychiatr. Epidemiol. 2004; 39: 196–207.

© 2009 The Authors

Journal compilation © 2009 Japanese Society of Psychiatry and Neurology