Translation performed by MSc.

Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes

Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

CSBI
INVENTORY OF
BEHAVIOR
SEXUAL
INFANTILE

PROFESSIONAL MANUAL

WILLIAM N. FRIEDRICH, PhD,

ABPP

1
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

1
INTRODUCTION

The Child Sexual Behavior Inventory (CSBI) is a

measure that uses the report from the parents (mother or primary)
female caregiver of sexual behavior in children at ages
from 2 to 12 years old. The purpose is to use it in the assessment of children
to those who have been sexually abused or in whom
suspected that they are being. The CSBI was developed in
recognition of the fact that sexual abuse is related to
early sexual behavior in children (Einberder and Friedrich,
1989; Friederich, Urquiza, and Beilke, 1986; Kendall – Tackett,
Williams. Y Finkelhor, 1993). Its 38 items evaluate a broad
range of sexual behaviors that cover a number of areas
including boundary issues (development or individuality
associated with difficulties maintaining interpersonal distance)
exhibitionism (revealing sexual parts of adults or children)
gender role behavior (interest in acting as or being a
member of the opposite sex), self-stimulation (touching oneself
same with the purpose of sexual pleasure), sexual anxiety,
(anguish when demonstrating adult sexuality), sexual interest
(interest in the opposite sex and in sexual behaviors), intrusion
sexual (violation of another person's privacy),
sexual knowledge (awareness of sexual behavior beyond
of typical knowledge by age level), and behavior
voyeur (an aspect of sexual interest) reflected by
efforts to observe the sexual parts of others). See the
Table No. 1 of the represented fields and their corresponding
items. These areas only define the scope of the content
and they are not intended to be used as psychometric scales
in the current version of the CSBI. The complete version of the CSBI
consists of 38 items and three clinical scales (Total CSBI,
Development Associated with Sexual Behavior (DRSB) and Items
Specifics of Sexual Abuse (SASI). A brief description of
The CSBI scales are presented in Table N. 2.

2
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes.
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

Table No. 1
Fields of Child Sexual Behavior in Children
corresponds to the item(s) CSBI

Field Items Description

Problems of 2, 7, 17, 24, 26, 32, 36 Difficulties
Limits related to the
maintenance of the
interpersonal distance.

Exhibitionism 21.33 Revealing parts

adult sexual or
boys / girls.

Conduct of 1.3 Interest in acting as

Role Gender or being a member
of the opposite sex.

Car 4, 5, 8, 12, 18 Touching himself

Stimulation for the purpose of pleasure
sexual.

Anxiety 25 Anguish when

Sexual demonstrate sexuality
adult.

Sexual Interest 6, 15, 28, 35 Interest in sex

opposite and in behaviors
sexual.

Intrusion 9, 10, 11, 13, 14, 16, Violation of the

Sexual 27, 29, 31, 34 sexual privacy of
another person.

3
 Translation done by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic Research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

Knowledge 20, 23, 37 Consciousness of

Sexual more sexual behaviors
beyond knowledge
typical by level of
age.
Conduct 19, 22, 30 An aspect of interest
Voyeur sexual reflected by
efforts to observe
the sexual parts of
others.

Table No. 2
Brief Description of the Clinical Scales of the CSBI

Scale Item Content

CSBI Total Evaluate the nine fields of sexual behavior:
problems of limits, exhibitionism
gender role behavior, self-stimulation,
sexual anxiety
sexual, sexual knowledge, and behavior
voyeur
Regarding reported behaviors that have occurred due to
the minor behavior once in the last six months to
Sexual (DRSB) less 20% of the caregivers in a sample
regulation. This scale reflects behavior
appropriate according to the age level and
gender.
Items Contains items that are empirically
Sexual abuse related after a history of abuse
Specific sexual first controlled by the effects of
(SASI) maternal and family education.

4
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for the purposes of Forensic Clinical research
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

Sexual Behavior in Sexually Abused Children

There are a number of strong reasons that sexual behavior will be

frequently a result observed in children sexually
abused. Finkelhor and Browne (1985) described sexualization
traumatic as one of the four factors that contribute to
negative impact of sexual abuse. traumatic sexualization
(preconscious and traumatic introduction to sexual behavior)
this brings a number of psychological consequences. One of
these consequences are the confusion of a part of childhood
about the difference between love and sexual behavior. Another
the consequence is that sexual matters become
outstanding for the sexually abused child.
Finkelhor and Browne also suggest a range of
immediate and long-term behavioral manifestations
including sexual concern, sexual coercion, and the inappropriate
sexualization of parents (exhibit poor sexual boundaries with
their own children as a result of their own child abuse.

In addition to the traumagenic theory, other theories as well

predict sexual behavior as a consequence of sexual abuse. The
A sexually abused child may act in a sexual manner if
this behavior has been learned as a way to
interact with adults or children and behave accordingly
sexually gives a child the opportunity to become or to fall
in an earlier victimization (Tharinger, 1990).

It has also been suggested that sexual behavior occurs only

in clearly predictable situations or circumstances (bathroom,
bedtime). This suggestion gives credit to the hypothesis of
associative-joint learning, which suggests that children
paired in sexual conduct with any person or a
context that shares a number of characteristics with a
abuse situation (adult male in a caregiver role,
agitation at bedtime after having been abused in

5
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

this context; Friederich and Gerber, 1994). Sexual abuse is

a silent variable and for many children, sexual excitement
it is part of an abuse. Recent data suggests that the
excitement that accompanies sexual abuse predicts behavior
more persistent sexual. (Hall, Mathews, and Pearce, 1997). The
reported cases have also suggested the opposite: that it is, if the
abuse is not exciting, or if the child associates it with pain, the
sexual conduct is less likely (Friederich and Reams, 1987).

Two important reviews of empirical literature on the

The impact of child sexual abuse has been pointed out by the two previous ones.
as inappropriate, sexualized play as one of the
the most predictable consequences of child sexual abuse (Browne
y Finkelhor, 1986; Kennedy-Tacket, 1993). For example, when
non-abused children
abused in clinical treatment sexually children
abusers exhibited sexualized behavior more frequently
in six of the eight studies cited in these reviews. All
the studies that contrasted sexually abused children
clinical and non-clinical children showed differences in behavior
sexualized. Moreover, sexual behavior is not a consequence
universal (most likely due in part to the nature
widely variable of sexual abuse) this is a consequence
recognizable and frequent.

Sexual victimization and its relation to subsequent behavior

sexual abuse in children has been identified in most of the
studies using various sources such as a review
psychiatric, the report from the parents, psychological evaluation,
child observation and the child's self-report (Friederich,
1993). In addition to a number of variables of sexual abuse that
they have been found to predict sexual behavior in children.
Although not consistently identified, such variables
it includes the number of perpetrators, frequency of abuse

6
 Translation by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes.
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

sexual, and duration of abuse (Friederich, 1993) as well as the

previously mentioned sexual arousal (Hall, 1997).

Since the reviews by Friederich (1993) and Kendall-Tackett,

research on the consequences of sexual abuse in children
has more sexual behaviors systematically examined.
For example, a recent study (Mc Clelllan, 1996) found that
the earliest age of onset of sexual abuse (greater than
seven years) was the most significant predictor of three types of
sexual behavior: hypersexual, exposure and victimization. The
authors also noted that sexual behavior was related to
a number of factors, including the child's age in the
moment of abuse, disorganized family environments or
altered, presence and frequency of abuse and neglect, and a
number of sexual perpetrators. Sexual behavior is more
possible to occur in children for whom the abuse begins to
early age, those who live in family environments
disorganized, where sexual abuse was accompanied by negligence
the physical abuse, or those who were sexually abused by a
number of people.

Gender role behavior and identity is another aspect of

psychosexual development that has been shown to be related to
sexual abuse in preadolescent girls (Cosentino, Meyer-
Bahlburg, Alpert and Gaines, 1993). The authors found
significantly more crossing gender conduct and conflict
gender identity in sexually abused girls when
they contrasted with two demographically comparable groups from
girls not abused, one from a pediatric clinic and the other
with patients from a psychiatric clinic.

Other forms of child mistreatment they are also

associated with the increase in sexual behavior. For example,
Deblinger, McLeer, Atkins, Ralphe, and Foa (1989) demonstrated a
increase in sexual behavior among children who have

7
 Translation performed by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Forensic Medicine. Section of Forensic Psychiatry and Psychology. 2006

physically abused. Polansky, Chalmers, Buttenweiser and

Williams (1974) determined that neglected children
they showed an increase in sexual behavior.

Although early research on sexual abuse and

sexual conduct primarily examined meticulous aspects of
sexual behavior such as sexual interest and masturbation, a
a wide range of behavior has now been included. For example, when
less nine of the different areas have been included
categorically within the CSBI (table 1). As mentioned
previously, these nine areas of the CSBI include issues
of limits, exhibitionism, gender role behavior, self
stimulation
sexual knowledge and voyeuristic behavior. Furthermore, these
areas do not constitute psychometric scales, the coefficients
alphas were calculated with the total standardization sample
already mentioned. These alpha coefficients
reasonably high, which range from .60 to .79, reflect the
integrity of these areas.

Sexually coercive behavior in children is another

consequence related to sexual abuse, also such behavior does not
is always the result of sexual abuse (Bonner, Walker, and
Berliner, 1996; Friederich and Luecke, 1988; Gray, 1996. By
For example, Gray used an analysis group to better understand the
subsequent diversity in a clinical treatment sample
on coercive sexuality in children aged between 6 to 12
years. Despite the fact that as a group, the sexual behaviors were
extremely high, a variability was evident
considerable. A 5-group solution appeared to present the
better explanation for this variability. Interestingly, the
group dominated by conduct - disorganized and children
oppositionists, challengers had reported the highest levels
low (less than 50%) of sexual abuse. Furthermore, it is true that
the parents of the boys/girls in this group were typically

8
 Translation done by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes.
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

monitored and generally do poor monitoring of behavior

from their children and (b) the families were chaotic, new
investigations in children with clearly coercive sexuality
holds the conclusion that not all sexually related behavior
inappropriate originates from a history of sexual abuse (Bonner,
1996; Gray 1996). In reality, a possible diagnostic criterion
for conduct disorder in the fourth edition of the Manual
Diagnostic and Statistical Manual of Mental Disorders (DSM-IV;
American Psychiatric Association, 1994) includes 'has
forced someone into a sexual activity" (p.90).

Future research should closely examine the point until

What are the particular areas of sexual behavior?
related to different aspects of sexual abuse,
characteristics of the family context, or other types of evil
treatment. For example, children with a history of neglect
The abuses have been described as exhibiting and showing affection.
indiscriminately, which should fall within the scope
called Boundary Problems (Steele and Pollack, 1971).

Normative Child Sexual Behavior

Child sexual behavior has never been the central focus of the
development research, despite the fact that humans,
including children, they are sexual beings. In this review of the
literature on sexual development in children, Rutter (1971)
it points out that children are sexual beings from birth and
that infants have the ability for a penile erection and
a vaginal lubrication. The information he analyzes suggests that,
Moreover, preschoolers are more sexual than children.
latent age, this difference may simply indicate that the
latent age children hide or conceal their behavior more
sexual of their parents.

9
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

The lack of research has made it difficult to determine the limits

typical of sexual behavior in all children, if they have
either abused or not. When childhood sexuality has been
studied, the information has typically been based on the
reports from caregivers or parents. For example, Phipps-
Yones, Turner, and Kauper (1993) asked 564 caregivers.
diaries if they had demonstrated a range of sexual behavior
in two groups of preschoolers, aged between 1 to 3 years and 4 to 6
years. Sexual behavior was reported to be very clearly
common, particularly in older children. For example,
approximately 30% of all children in this study
they occasionally reported masturbating, typically at the time of
sleep. Approximately one in three caregivers for the
older children reported that they have seen children of this
age group use dolls to act out sexual activity. Children
As they grew older, they were also seen with greater sexual curiosity.
that the younger children about the mechanisms of activity
sexual and were more prone to engaging in sexual games,
just like 'playing doctor'.

Small percentages of caregivers have observed behaviors such as

how to touch the sexual parts of other children or adults and
drawing sexual body parts. Behaviors that have a
very low probability what they have been reported by the
caregivers include French kisses, pretending exchanges
sexual activities among children, inserting objects into the vagina or into the
buttocks, and ask other children to kiss their genitals. The
authors did not notice real differences in gender frequency
of various behaviors with these children/young people. Another
research based on parental reports has shown that
by the age of 10, 30% to 45% of children have
touched his mothers on their genital breasts at least once
(Rosenfeld, Bailey, Siegel and Bailey, 1986).

10
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Section of Forensic Psychiatry and Psychology. 2006

Gender role conduct in preschoolers has also been

studied by a specific measure of parental reporting
designated to assess gender role behavior (Golombok and Rust,
1993). The authors noted that sex typically increases
during the preschool years and that older children increasingly
more adopt gender roles. In any case, there is a great
amount of variability in gender role behavior in
very young preschoolers.

A new approach was used by Lamb and Coakley (1993) in their

retrospective study of sexual entertainment and games. A
a sample of 128 female students from a girls' school were
surveyed considering sexual games in childhood.
Approximately 85% of the sample described a memory
of a sexual game in childhood (M age of the person in the
moment of the reported incident = 7.5 years, SD = 2.0 years). These
games fell into six categories: playing doctor,
exhibition, experiments in stimulation, kissing games,
act out sexual fantasies and others. The most sexual games
frequent included a crossing of gender-experience, the most
similar to a game was seen as including manipulation or
coercion.

Of particular relevance to the CSBI is the fact that 14% of this

show reported having kissed another child/, 26% reported
exposure of themselves, 15% reported genital touching
while dressing, 17% reported genital touching without
dress up, 6% reported using an object around the
genitals, 4% reported oral genital contact. It affirms behaviors
for all age groups presented in chapter 3 of
this manual and were as follows: 9.4% of the
children kissed another child, 21.4% exposed their parts
sexual to adults, 9.9% exposed their sexual parts to
other children, 11.1% touched another child in their parts
sexual, and 13.2% used a doll or object to masturbate.

11
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

In summary, the information from Lamb and Coaley (1993) is very

similar to the CSBI information for females. Information
similar is not available in masculine.

A number of studies on the sexual behavior of children have

examined the socio-cultural influences. For example, in the
socioeconomic field has been found to be directly
related to sexual knowledge in children (Gordon,
Schroeder and Abrams, 1991). Smith (1990) also found
significant differences in age and social class in
sexual behavior in families. For example, nudity of
Parents were much more common in families with more children.
young people and families of at least middle class.

Cross-cultural studies also provide useful information.

about normative aspects of children's sexual behavior
in countries that may be more relaxed than the United States
United about sexuality. A study of children.
preschools in Switzerland illustrates that this fact (Lindblad,
Gustafsson, Larsson, and Lundin, 1995). The results of this
the study was consistent with the findings of the CSBI from
very low frequencies of certain intrusive and self behaviors
stimulating among children who have been sexually
abused. For example, Swiss caregivers reported a range
of 1.2% insertion of objects into the vagina or their buttocks,
compared to 3.2% in the CSBI study. The Swiss study indicated
a range of 3.2% compared to the CSBI study with a
range of 1.5% to intend to have sexual intercourse. From
in any case, the authors did not find a relationship between
sexual behaviors and other behavioral problems in the child
a relationship highlighted in an earlier investigation of the
CSBI (Friederich, Grambsch, Broughton, Kuiper, and Beilke, 1991).

12
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes.
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

Parent report versus Self report

Research on the sexual development of children is difficult.

research area for many obvious reasons. First, due to
to ethical considerations, researchers are unable to
describe all areas of sexuality when examining
directly the knowledge of children's behavior
sexual. To my knowledge, the Trauma Symptoms Inventory
Child (TSCC; Briere, 1996) is the only psychometric measure
that directly evaluates at least some aspects of the
self-reported sexual behaviors and feelings in children.
One of the subscales of the TSCC, called Sexual Issues,
evaluate the extent to which the child thinks about aspects
sexual (Sexual Aspects - Concern) as well as the degree
until which a child is afflicted by sexuality (Aspects
Sexual - Anxiety). The TSCC was designated for use with children.
from ages 8 to 16. Due to the sensitive nature of
the items, the TSCC also has an alternative version (TSCC-
A) that does not include the subscale of Sexual Aspects.

In summary, many of the sexually abused children are very

young people and they cannot report validity either of their behavior or of their

level of sexual knowledge. This points to the need for

to observe an estimation or assessment of sexual behavior in children
Parents are the most appropriate observers because
that they are capable of observing the child in large numbers
of areas where sexual behavior can be demonstrated (to the
bath time, shower time, bedtime. The
mothers are generally seen as the most reporting ones
parents are valid sources of general behavioral problems
sexual (Achenbach, McConaughy, and Howell, 1987), also the
reports from mothers and fathers are significantly
correlated (Achenbach, 1990).

13
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Section of Forensic Psychiatry and Psychology. 2006

Not all parents are the same in their reporting skills.

in a valid manner the behavior of your son/daughter, particularly the
sexual behavior. This difference is due to both factors:
factors of parents and factors of children. The factors
of the parents includes the time they spend with their child
the father's ability to assess the behavior of
child /a, and other interfering variables. For example, the
research indicates that people may differ in their
willingness to respond to sexual aspects. Fisher, Byrne, White,
Kelley (1988) created the concept of erotophobia-erotophilia.
as a dimension of personality in which individuals
those who are erotophobic have learned to evade aspects and
experiences related to sex. Furthermore, this aspect was not
valued in research with sexually abused children
Erotophobic parents may be expected to see less
sexual behavior of parents who are more inclined to
observe sexual behavior.

Due to these differences in the factors of the parents, the

parental attitudes considering the normality of behavior
sexual abuse in children was assessed in the following research
with the CSBI. The parents were interviewed to answer 'yes'
It is normal for children to have
"feelings and sexual curiosity." This item was
positively correlated with each of the items of
CSBI. That result means that the parents who were in
agreement with the child's sexual behavior is normal that
They were more likely to report sexual behavior in their child
Multiple regression analysis was used to examine the
relationship of this item to the total reported sexual behavior.

Other caregivers were asked to evaluate the children in their place.

on care, with questions about trends and some caregivers
They can be more objective than the parents due to their relationship.
different with the child. The comparison of foster mothers to

14
 Translation performed by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

biological mothers were valued with the first version of 35

items from the CSBI (Friederich 1992). None were found.
significant differences between foster mothers and the
biological mothers in the total score of the CSBI.

Relationship between the CSBI and the Behavior Inventory

of the Child (CBCL).

Although the content of many items in the CSBI was influenced

for items of sexual behavior from the Behavior Inventory
Child (CBCL; Achenbach, 1991), these two measures have
clearly distinct approaches. During the course of development
from the CSBI, the CBCL was completed concurrently with the
CSBI. This procedure allowed for a wide-ranging investigation.
range of behaviors in children as well as the degree to which the
CSBI and CBCL identify sexually abused children.
specifically, Firiederich (1986) summed six items from the CBCL
from the sexual behavior scale in a previous study.

The capacity of the six items of the CSBI and the CBCL to
to identify sexually abused and non-abused children was
evaluated using the ROC analysis curve with the first version
of CSBI (Friederich 1992). Both specificities (true
negatives) and sensitivities (true positives) were
calculated and the CSBI redesigned the CBCL for each of the four
age and gender groups. For example, estimated specificities
from .05 to .38 for the CBCL and from .35 to .55 for the CSBI.
Estimated sensitivities range from .35 to .60 for the CBCL and from
.70 to .92 for the CSBI.

Discriminative analysis was used with the first 36 items.

from CSBI (Friederich, 1993). This analysis also allowed the
calculation of the relative skills of six items from the CBCL and the
CSBI that correctly classify abused or non-abused children
sexually abused. The CSBI redesigned the six items again

15
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for clinical forensic research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

from the CBCL, with a very precise estimate of 74% to 76% for the
six items from the CBCL.

These results suggest that the total accuracy for identifying

sexually abused children are improved when behavior
sexual is valued with a longer scale that evaluates a
a large number of fields of sexual behavior.

Characteristics of the CSBI

The 38 items of the CSBI are included in a brochure in which

the caregiver (mother or primary female caregiver) writes
directly its answers. The caregiver is presented with a list
of behaviors and is asked to mark how frequently
she has observed each of those behaviors in the last six
months. Each item is rated on a four-point scale that
start from 0 (never) and 3 (at least once a week). The CSBI
it takes 5 to 8 minutes to complete for most of the
caregivers or more than 15 minutes if the items are read to him
caregiver. This can be scored in less than five minutes. The
CSBI requires a fifth-grade reading level.

This manual provides information about the CSBI materials, the

administration, punctuation interpretation and characteristics
psychometrics. It also provides normative information based on
1,114 children from the general population and information from the CSBI (N
of the main sample of the abused child.

The current version of the CSBI with 38 items also differs from
the two priority versions in terms of age groups for
which the standards were calculated. In this version were
calculated age groups of 2 to 5 years and from 6 to 9 years and from 10 to
12 years whereas in previous versions groups were used
aged 2 to 6 years and 7 to 12 years. The scores for each
items were determined for each child by dividing the

16
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for the purposes of Forensic Clinical research.
Department of Legal Medicine. Section of Forensic Psychiatry and Psychology. 2006

Total gross score of the CSBI divided by the number of items (38).
This average was subsequently calculated and analyzed for
each age. The results by item represent the frequency of
the sexual behavior observed and reported by the sample
regulations. In the age group of 2 to 5 years, the results were obtained
highest scores from the rest of the age groups and in the
children aged 6 to 9 obtained the lowest scores.
In any case, at the age of 10, the scores in girls
they increase and the same happens in children but after the
11 years old. These results suggest three groups that
they would be more sensitive to age compared to the other two
groups. (See chapter 5 for the differences between the version
current and other later versions of the CSBI).

17
 Translation performed by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Section of Forensic Psychiatry and Psychology. 2006

2
Test materials, Administration and
Score

CSBI materials

The materials consist of the Professional Manual and the Brochure of

manual scoring (protocol). The manual scoring brochure (or
protocol) brings the instructions at the top of the page
to complete the test and examples of how it should be marked
answers as well as the items. Caregivers can mark
your answers directly in this Brochure. There is a Brochure
with charcoal that is used for the examiner to review the test,
however, the computerized version can also be found.
When it comes to manual scoring of the test, the raw score
it is converted to T score by gender and age found in
Annex A.

Appropriate populations

The CSBI is suitable for female caregivers. of children

males and females aged between 2 to 12 years. The sample
regulation for this instrument includes children within a
social, economic range including urban and rural environments.
The CSBI is applicable to children in a wide variety of
demographic and social contexts.

Applicability in ages 1 to 13 years.

Some clinicians have successfully reported the administration

with children one year older and one year younger than them
prescribed for this measure (with 1 year and with 13 years) using

18
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for clinical forensic research purposes.
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

standards for 2 to 5 years of age and 10 to 12 years of age

respectively to carry out the interpretation. Because not
reported data available in the CSBI of parents
of children aged 1 to 13 years, and given the considerations of
development that has been identified (occurrence about behavior
sexual increases in ages 2 to 5 years and typically
descends in ages from 10 to 12 years), the examiner is
advertised to avoid applying the CSBI standard data in
such children except for a strong development index
appropriate to the child's behavior.

Applicability to children with Mental Retardation

The normative sample does not include children with intellectual disabilities.
or physical limitations. In any case, the clinical experience
with the CSBI in the evaluation of non-abused children
mental retardation suggests a narrow range of sexual behavior with
the presence of more compulsive and self-sexual behaviors
stimulating (touching the genitals both at home and in
public places, masturbation, etc.). This assumption is excluded.
for an examination of the CSBI data from a sample
small (n = 28) of children with intellectual disability and in these
cases, the development of age has been used when referring to
standards.

Professional Requirements

The CSBI can be administered and scored by individuals.

those who do not have formal training in Clinical Psychology,
psychological orientation or related fields. The examiner must
carefully study the administration and grading of the
procedures presented in the manual. According to the
Educational and Psychological Assessment Standards (Association)
American Journal of Education Research, American Association
of Psychology and National Council for Educational Measurement,

19
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Section of Forensic Psychiatry and Psychology. 2006

1985), the interpretation of CSBI scores requires a

formal training in Psychology, Guidance, Social Work,
Psychiatry or another related field as well as training
relevant or practical experience in the interpretation of tests
psychological accredited by a university.

Administration

The materials required for the administration are the brochure.

from CSBI, a pen or pencil and a flat surface. The environment
evaluation should be relatively quiet, free of
distraction and adequately lit. The CSBI is typically
managed in individual assessment situations. Of any
way, when the CSBI is administered to caregiver groups, the
the evaluation environment should protect privacy and
confidentiality of the responses of each individual. Due to the
the nature of this instrument, some caregivers may be
not comfortable with the content of the item so they the same
they must be explicit to ensure the validity and reliability of
the answers to the items.

Specific Instructions

Deliver a CSBI brochure to the caregiver. The information

demographic can be completed by the examiner or by the
caretaker. Explain that this brochure describes a number of
behaviors that are sometimes seen in children. Ask
to the caregiver to read each item carefully and consider whether the
behavior described has been or has not been observed in the
last six months. Some caregivers may need a
specific point ("If we count back six months
from today, which would be August 18.

Instruct the caregiver to encircle the number.

Correct in the test brochure. Report that enclosing a 0

20
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

means that the boy/girl has never shown the behavior in the
six months earlier. He continued to inform that locking a 1
it means that the behavior has occurred less than once a month
in the last six months, while locking up a 2 wants
to say that the behavior has been observed one to three times by
months in the last six months. Finally, tell the caregiver that
closing a 3 means that the behavior has occurred at
less than once a week in the last six months, for example,
a foster mother should use the same criteria as
frequency (2 means that the child has shown the behavior of
once to three times a month during the time she has been with
the child /a). Tell the caregiver to complete each item and take
the time needed to complete the test. Ask the caregiver
he/she understands what is expected to be done and that he/she does anything
ask if he needs it. It is important to clarify, for example,
when caregivers affirm item 18, "Puts objects in his/her
vagina or rectum”, it should help clarify exactly what the
the caregiver has witnessed (between the buttocks vs. inside the rectum or
between the lips vs. vagina.

Item Presentation

The CSBI can be used as part of an interview.

structured or can be read aloud to the caregiver if it has
difficulty in reading. Moreover, the author is not familiar with
research that indicates whether the items are read or the
CSBI as a structured interview invalidates the
answers, the examiner should be aware that the data
The CSBI regulations are based on the reading of the items.
of the caregiver, more than having to read them. Therefore, the
non-standardized administration may invalidate the interpretation
from the T scores because the norms were collected
through the caregiver's report. If the examiner reads the
items, this must be taken into account in a report
subsequent.

21
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Section of Forensic Psychiatry and Psychology. 2006

Reports from parents and other caregivers

There are moments when parents' information is also

desirable. The parents who live together provide
similar answers. However, there are no available rules.
for the parents. The parents' data suggests that they
they tend to report few sexual behaviors. Furthermore, when
they identify such behavior tend to report it less
frequency that what mothers who live with it do
father and with the evaluated child. Particularly for each child
In young people, sexual behaviors are often viewed as
bath time or bedtime. If the parents are not there.
present at this time, they can observe less than
these behaviors. This is very useful when obtaining the report of the
parents and when they can be directly contrasted with the
reports from other caregivers.

Caregivers, teachers, grandparents, and counselors have also

completed the CSBI because they have had the opportunity to observe
the child being assessed. However, there are no standards for
these individuals and the range of situations they can
observe the amount of time they have known the child
it can also vary from the main female caregiver.
the advantage of additional observers is that they provide
information about the extent to which the behavior of the child has been
generalized to other spheres.

The degree to which additional caregivers validate each one of the

Other reports provide useful information for the examiner.
Furthermore, this is confirmed when two reports provide
similar information. There are reasons that observations may
to be divergent, like when one of the parts can be
defensive or it can be a more precise observer, or it can have
more environments or fields in which you can observe the child.

22
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

Rating

Absent Responses

Given the brevity of the CSBI, a few missing responses

present a problem for interpretation. The sample
regulation includes only those parents who completed
all the items. It is therefore important to be attentive to
if the caregiver has left absent responses for what
complete with the help of the evaluator and consider as scores
CSBI invalid if there are more than two incomplete items.
For grading purposes, replace the missing values
by a 0.

Calculating the scale of gross scores CSBI

Total CSBI

The total raw score of the CSBI is obtained by adding the

responses to the 38 items. Remember to record 0 for the
absent responses. The total raw score of the CSBI can
have a range of 0 to 114. Record the sum of responses from
all 38 items in the box to the right of
the page. Note that the shading on the score sheet is
designed for clarity and ease of use.

The answers to item 38 are included in the gross score of the

CSBI only if they clearly define from the 37 main ones.
questions. For example, if in item 38 the caregiver writes in
"to masturbate while getting ready or in the dressing room," I couldn't
to be included in the total score if the caregiver had already affirmed
item 5, masturbating with the hand. However, if the caregiver
he said 'masturbating and playing with feces while cleaning oneself or is in '

23
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes.
Department of Legal Medicine. Section of Forensic Psychiatry and Psychology. 2006

"the dressing table" for item 38 could be scored as another behavior.

because this is clearly different from item 5. I have
found that the responses to item 38 are frequently
linked very directly to the child's context and, in some cases
of sexual abuse, in the way the child was harassed
("pinches on the nipples" or "wrap the tongue around a
carrot and make moaning sounds). Therefore, it is necessary,
clarify the caregivers' responses to item 38 for
decide how well to rate them.

If the caregiver clearly reports only two behaviors in the

item 38, only one of those behaviors is recorded in the
calculation of the Total Gross Score of the CSBI: the only one affirmed with
the highest frequency (the behavior scored as a 3). The reason
for this, only one behavior for item 38 was
used in calculating the norms due to two behaviors
truly different were very unusual.

Table 3 contains examples of responses to item 38. The first

the group includes behaviors that were determined different from
any of the current items of the CSBI and were
registered within the calculation of the total gross score of the CSBI
as item 38. The second group of behaviors were
related to the current items of the CSBI, continuing the
discussion with the caregiver, they were sufficiently determined
different from the other 37 items and were rated as items
The third group includes behaviors initially listed in
item 38 but, following discussions with the caregivers, were
are determined to be very similar to the current items of the CSBI
and were not scored as item 38. Regarding the previous one
if the 'other sexual conduct' described in item 38 is
essentially the same as an item in the CSBI that was not
it was stated (a score of 0 was given), then item 38 should not
be scored and the frequency score reported for the item
38 must be transferred to the previously unclaimed item. Without

24
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes.
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

embargo, the clinician should only change the scores

after consulting with the caregiver who completed the CSBI.
Sometimes the 'other sexual behavior' reported in item 38
it is essentially the same as the behavior described in another
statement affirmed, but both are estimated in different
frequencies. When this happens, only count the score of the
item estimated at the highest frequency in calculating the score
Total gross of the CSBI. The items of the CSBI corresponding
for the behaviors included in the second and third group are
also listed in table 3.

To enhance the clarity of the examples of the second

grouping in table 3, consider the following item: Suck a
carrot and say that she is sucking a penis. The mother of
fostering has already affirmed item 37 of the CSBI, Learn more about the
sex than other children of their age. As a result of the
interview it was discovered that an 8-year-old girl had
demonstrated to both his adoptive mother and another child how
the man puts his penis inside the woman what does it look like
erection and the man's sperm. The foster mother had
item 37 is affirmed as a 1 due to the demonstration of
the girl had only happened once, after she left home
I take care. However, the foster mother stated item 38.
Sucking a carrot and saying that she is sucking a penis
like a 3 because the girl exhibited this behavior at least 10
times in the four weeks she had been at home; she
he had also demonstrated this behavior with dogs,
candles and wooden logs. The girl had revealed an exchange.
vaginal but did not reveal oral intercourse until a few months later
afternoon.

The next item in this grouping in table 3, Wore clothes

female intimacy was scored on item 38 although the mother did not
had signed the item related to the CSBI (item 1). She never
I had seen a 10-year-old boy dressed in his

25
 Translation carried out by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

dresses, blouses, and pajamas (the mother's). However, she it

she had seen with her bra and elastic stockings (panties) in two
on occasions, and because she was starting to lose underwear
he assumed that he was putting it on. It was discovered that the
the perpetrator who was sexually abusing this child
admitted to dressing him in women's intimate clothing while he was
sexually harassing.

Table 3
Examples of rating for 'Other sexual behaviors'
Described in item 38.

Asserted behaviors and scored in item 38 (These

behaviors are clearly different from any other item of the
CSBI and therefore, were scored on item 38.
Pinches and cramps in their own nipples.
Push your pelvis while lying on the floor.
Save the used plugs of your sister or mother.
Playing with feces while masturbating.
Having regular sex with a 19-year-old boy
(reported by an 11-year-old girl).
Phone calls to '1-900'.

Affirmed and scored behaviors in item 38, but

related to the item CSBI. Included for each one among the
parenthesis the item of the CSBI to which it is related. These
behaviors are related to an item of the CSBI, but
continuing discussions with the caretaker were considered
are substantially different from that item, therefore,
they were scored in item 38.

Suck on a carrot and say you are sucking a penis.

(37)
Wear women's underwear.

26
 Translation done by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

Walking around a public bathroom knowing that men

homosexuals frequent it.
Rubbing one's own breasts to the point of getting aroused.
(12)
Leaving the bathroom naked while covering his erection with a towel.
(21)
Lying about hearing under the bed of his
sister how she had sex with her boyfriend.

Affirmative behaviors but not scored in item 38. They

include for each one in parentheses the item from the CSBI
to which it is related. These behaviors are markedly
similar to the items of the CSBI, therefore, they were not
scored on item 38.

Lick the arms and moan passionately again and again.

15
Ask your brother to give himself an erection and
that he shows you.
Suck on fingers and moan.
Ask his mother the date if he is going to have sex with her.
(23)
Often does not wear elastic stockings (panties) when
she puts on a dress.
Leaving the zipper down when sitting down.
table with his mother.

DRSB and SASI

Locate the DRSB and SASI Rating box (to the

left side of the brochure page of the test) that corresponds to the
age and gender of the child being assessed. Then transcribe
each answer of the item to the appropriate space in that box.
For example, consider the SASI rating scale for a
seven-year-old child. First locate the Rating box

27
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Forensic Medicine. Section of Forensic Psychiatry and Psychology. 2006

from DRSB and SASI "Children from 6 to 9 years old". (Note that each box
it contains both scales the DRSB and SASI for each group of
genre). If the caregiver marked a 2 in item 23, record a 2
in the SASI column in the line corresponding to item 23. Do
the same for each of the items of these two scales.
Add the item responses to reach the gross scores of
the DRSB and SASI scales and record those values in the space
in white labeled 'sum' at the end of each column. See Figure 2
in the complete and graded example in the CSBI brochure.

Conversion of standard scores and how to complete the

summary box of the CSBI score.

The Appendix presents the gross score scale of the CSBI and the
corresponding T scores for each age group and
gender. Select the regulatory table according to the group of
age and gender. Locate the raw score obtained for each
one of these three scales (Total CSBI, DRSB, and SASI). Take a look at
the left end of the column to locate the
corresponding T scores, and record the raw and T scores
for each of these scales in the box called 'Summary'
of CSBI Rating" at the bottom of the rating page.
Also indicate the Regulatory table and the comparison group.
used (A2/Children, ages 6-9 years). For example, see figure 2.

28
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

3
Interpretation

Normative Comparisons

T scores are used to interpret the level of the

sexual behavior of the child in the CSBI. These scores are
standardized transformations of the raw score scale
derived from having a mean of 50 and a standard deviation of
10. The CSBI rules are scaled in such a way that the
Scores T indicate high levels of sexual behavior.
problematic. Similar is in the percentile scores, the scores
They give information about the child's scores related to
the scores of the children in the normative sample. By
example, a T score of 65 indicates that the child's score is
1.5 standard deviations above the sample mean
standardization and that this equates to or exceeds the scores at
94% of the children compared to the sample of
standardization. For all clinical scales, the T scores
those over 65 are considered clinically
significant. T scores in the range of 60 to 64 suggest
difficulty and can represent behavioral problems
subclinical (but still significant). The examiner of the CSBI
you should remember to consider the estimates or valuations of the
child's sexual behavior in the context of the child's environment
/a.

Interpreting the Clinical Scales of the CSBI

The examination of the scores of the three clinical scales of

CSBI allows for different perspectives on the child’s behavior.
This section presents interpretative hypotheses for each scale.
individually, with the warning that the CSBI scales
they are related, they are not independent.

29
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

Total CSBI

The Total Gross Score of the CSBI has been consistently

calculated through two initial versions of the CSBI
(Friederich, 1992; Friederich 1993). The total score of the CSBI
represents the sum of the responses to all 38 items. A
The answer in item 38 is calculated at the end of the total score.
gross only if the described behavior clearly differs from
all the behaviors described in the previous 37 items.

In addition, the Total raw score of the CSBI may have a range
from 0 to 114, in the sample of sexually abused children
total raw scores over 45 were very unusual (less
of 5%). Children with very high levels of sexual behavior
tend to be younger, have been abused more recently,
and exhibit high levels of other conduct problems
typically problems with aggression and level of activity.

Development related to Sexual Behaviors (DRSB)

At different ages, some sexual behaviors are

clearly more common than others. If one uses the tables of
frequency (Tables 4 and 5) as a basis for calculating these
sexual behaviors may be viewed as more normative
Many behaviors occur with some degree of frequency in
children not abused.

CSBI scores are often high along with

other problems of sexual conduct. However, when the
majority of the total gross score increase of the CSBI is
based on the DRSB items, I have found that this is
frequently in younger children in the midst of some
transition of life (parents' divorce) who were
primarily described in their assessment as more active and

30
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

intense than the average of a child of their age and gender. The
DRSB scores are also significantly related to
the sexuality of the family, and something of this elevation may
reflect greater exposure of the child to nudity or sexuality
adult compared to the average.

Specific Items of Sexual Abuse (SASI)

High SASI scores seem to be related to abuse.

sexual, particularly if the behaviors are observed, to
less in part, by other caregivers of the child. Another child or
family variables (physical abuse, inconsistent caregivers)
they often exacerbate the intensity of these behaviors. If
The child is not very communicative regarding sexual abuse.
It can be very difficult to justify it under these circumstances.
Many of the items of the SASI for various age groups and
gender even reflects sexual interest or behaviors
sexually intrusive. Consequently, the caregivers
they may need to monitor the behavior of those children more
carefully to reduce the possibility of victimization. It
recommend the behavioral interventions that they reduce the
intensity of those behaviors (Friederich, 1990). Furthermore, it
more research is needed to determine what other
harmful circumstances may be related to the
high scores of the SASI.

Sexually Aggressive Behavior

Despite clinicians increasing awareness of children

in a sexual way aggressive, recent investigations have
indicating that not all sexually aggressive children have
having a history of sexual abuse (Bonner, 1996; Gray 1996).
This seems even truer for children with a diagnosis of
conduct disorder or oppositional defiant disorder
(Gray, 1996). In an effort to increase understanding of

31
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Division. 2006

sexually aggressive children developed some.

additional items of behavior. The incidence of the data in
those items were obtained in the test-retest sample of
CSBI and all items were affirmed at the level of 0 by the
parents of children not abused aged 2 to 12 years.
These results suggest that, although sexually
aggressive behavior can occur in children with no history of sexual abuse,
such behavior is still relatively uncommon in children
abused.

When the completed CSBI is reviewed by the caregivers, the

clinicians ask if caregivers have also observed
any of the behaviors listed in Table 6 during the
last six months. By affirming any of these items, it
indicates that a child may be engaged in a behavior
very coercive sexual, and it is likely that a number of items from
CSBI have also been affirmed. A sexual behavior more
common such as self-stimulation, including masturbation,
it is less likely to be related to a history of abuse
sexually more unusual behavior, such as aggression and
coercive sexual conduct (Firederich, 1997). Of any
way, even aggression is relatively unusual and
it has a stronger relationship than self-stimulation to abuse
sexual, which is not evidence of sexual abuse (Bonner, 1996; Gray
1996).

32
 Translation done by MSc. Paula Vanesa Montero Gutiérrez for clinical forensic research purposes.
Department of Legal Medicine. Section of Forensic Psychiatry and Psychology. 2006

Table 6
Behavioral items to assess sexual aggression

Touches other children in their private parts after

whether said or not.
Plan how to sexually touch other children.
Forcing other children to engage in sexual acts.
Places their finger or an object in another child's vagina or rectum.
/a.
_______________________________________________
Note: These items were examined in a test-retest reliability study.
None of these items were affirmed in a sample of 104 children.
abused (M = 6.1 years, SD = 2.7 years).

Using the results of the CSBI

A research component of alleged abuse
sexual

The most important aspect of any abuse investigation

sexual is the testimony or what the child reports. Others
aspects of the research may include medical evidence, the
presence or absence of predisposition and risk factors
behavior consistent with the experience of a distressing event
(PTSD), premorbid functioning, the level of sexual behavior
exhibited by the child. Because parents typically do not
they have a pre-existing perspective on their child's sexuality
the CSBI data in combination with an interview with the
parents can provide valuable information about this area of
behavior that is very relevant to the aspects of abuse
sexual.

The following questions can help the clinician clarify the

CSBI results:

33
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Section of Forensic Psychiatry and Psychology. 2006

The CSBI data are directly consistent with the

testimony / story of the child.
2. Is the total score of the CSBI consistent with the aspects
of the normal functioning of the child.
3. There is a temporal relationship between the alleged abuse and the
reported sexual behaviors?
They are anyone of the behaviors reported
specifically subject to aspects of the alleged abuse?
5. There are favorable interpretations of the reported behavior
that also need to be considered (presence of a
recent stressor, elevated family sexuality, attitude of
A caretaker regarding the child's sexuality?
Is the father (or caregiver) a good monitor of behavior?
child
7. The environment of family members contributes to the
affirmation of any of the items of the CSBI (space
personal, family nudity)?
8. This is the empirically reported sexual behavior.
directly related (+) or inversely related (-) to the
following variables in the standardization sample:
maternal education (+), family income (-), hours that are
spending on care (+), parental violence (+), physical abuse
(+), stress of living (+), family sexuality (+), and exposure
pornography (+)
9. Given that the CSBI is a valid measure, the father (or
caregiver) has some reason, such as a dispute over the
guardian, that exaggerates or minimizes the level of behavior
sexual? (Note that the standardization samples
they excluded all custody cases out of consideration for
this reason).

The denial or minimization of sexual behavior can be a

aspect, for example, when a child is in care of
upbringing and the biological mother remains with the alleged offender.

34
 Translation done by MSc. Paula Vanesa Montero Gutiérrez for clinical forensic research purposes
Department of Legal Medicine. Section of Forensic Psychiatry and Psychology. 2006

The foster mother may have noticed a sexual behavior in

the child, while the biological mother does not assert behaviors
in the CSBI. In situations where denial is likely, it
it is recommended that the clinician meets with the father and starts
clearly establishing a six-month timeframe.
Then the clinician should start with the items of the DRSB
corresponding to the age and gender of the child and asking the
Father, if you have observed such behaviors over time
framed. If the father continues to deny, the clinician may want to
share your observation of the child (poor boundaries, grabbing the
single) with the father and ask the father if he has witnessed such
behaviors. Many parents (or caregivers) become more
revealing under these circumstances, and if the clinician wants, they can
review the rest of the items of the CSBI later. If the father
continues to deny the sexual behavior in the child to the clinician
you should consider ending the discussion.

The consistency of the CSBI results can be evaluated

for an interview about observed behaviors, as illustrated
in the previous example, or more formally to have caregivers
alternatives to complete the CSBI.

The CSBI provides useful information but does not provide certainty.
test of sexual abuse. Despite the fact that the decisions
consideration of guilt or innocence is determined through a system of
cut, there is absolutely no test or symptoms, other than the
presence of sexually transmitted diseases (STD)
what is consistent with sexual abuse. Even in these
uncommon situations, an STD has not yet identified the
perpetrator.

If in an eleven-year-old girl, the reports of sexual abuse

are valid, the results of the CSBI can be useful, if the
scores are either high or not. For example, the child can
report abuse, but the mother and other caregivers can

35
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

reporting little a sexual conduct at the CSBI. The information

reported via discriminative analysis indicates that 25.2% ( false- }
negative estimated for this sample) of sexually girls
abused girls aged 10 to 12 are not classified as
sexually abused based on the CSBI scores.
Consequently, the CSBI data will not be used to
question the validity of a child's account.

If sexual behavior is present, the information from the

same discriminative analysis summarized in chapter 5 (the
conduct is now directly supported by the account of the
boy/girl). In addition, some of the individual behaviors stated
the caregiver can be very consistent with the details of the
report on the girls of the alleged abuse (the mother stated the item
31). However, there is no need to have a direct relationship.
between the experience of abuse and the expressed behavior,
especially if the child is older and more time has passed
from abuse. In such cases, a child may exhibit a
increase in the frequency of a variety of behaviors.

Individual items

It is important to examine each stated item and compare it to the

frequencies properly stated listed in Table 4 and
5. The items confirmed at a high frequency are
extremely useful to consider, but by themselves there are
only a part of a long evaluation process (see
Basic Rules for Interpretation presented in this
chapter).

There are moments when individual items can be very

convincing. For example, many preschoolers all in a
care home, they start to masturbate with a certain type of doll
small plastic. Furthermore, when asked about "others
sexual behaviors," the mothers of these children reported

36
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

that each child had recently started to use a certain

unusual phrase that made a sexual connotation explicit.
the child exhibited other behaviors suggesting distress, but the
circumstances of his life predisposed him to more problems of
behavior). The specificity of these two behaviors continues with
a general increase in sexual behavior, similar to maintaining
these children together, and this information provided at the end is
important in the prosecution of the perpetrator.

Individual items can suggest how commonly a

behavior is observed in a child of a certain age and gender.
Epidemiologists have used for a long time proportions
unequal (OR), and clinicians can also use OR. For
for example, I have found that item 6 is affirmed three times
more frequently in sexually abused male children in
ages 2 to 5 years in those of the same age group
but not abused. The combination of the CSBI scores with the
reported presence of some very unusual behaviors since
then increases concerns and exacerbates inequalities of
whether the children have been sexually abused or exposed
to an adult sexuality.

So that users of all measurement scales,

including the CSBI, they must pay close attention to the base range of the
phenomenon being projected and contrasted with probability
of error. For example, suicide has a very low baseline rate in
the children, but it is highly correlated with depression, it
which has a higher rate in the population (Hawton, 1986). This
It would be a mistake, much more so since they are not frequently established.
precautions for each depressed child.

The same precautions are maintained for sexual abuse and the
sexual behavior. In children, sexual behavior is much more
common than sexual abuse. The two correlations are very
significants are an important consideration, but due to the

37
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

the same reason that depression is not an effective predictor of

suicide, sexual behavior alone is not evidence of sexual abuse.

Treatment planning

The CSBI is also useful as part of the planning for

treatment interventions and monitoring of effectiveness
treatment (Berliner and Saunders, 1996; Hall-Marley and Dannon,
1993), The instrument evaluates behaviors over the six months
previous. However, for short periods of
treatment, the caregiver can assess the behavior of the child
since the beginning of the treatment, and these new CSBI scores
and individual frequencies of the item can be compared to the
initial estimates by the caregiver. The reliability data
They suggest that the CSBI is stable over time. Lanktree and
Briere (1995) has noted that sexual aspects, measured by
the TSCC (Briere, 1996) are more resilient than generalized
anxiously to the effects of the treatment, but it has not
a similar study has been conducted with the CSBI.

Basic Rules for Interpretation.

The CSBI should only be a component of a long process.

of evaluation that typically includes at least interviews with
caregiver and the child, the use of the rating scale of
behavior to assess a wide range of behavioral issues
in the child (Child Behavior Checklist (CBCL);
Achenbach, 1991) and for older children, a measure of
auto report just like the TSCC. The CSBI should not be used
as a single standard measure in clinical evaluation, furthermore,
it has been useful as a selection measure in studies for the
research (Berenson, 1997).

The only trust in the CSBI or a portion of the CSBI to obtain

a clinical determination about the condition of the child

38
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

sexual abuse is not appropriate. The CSBI is a means to obtain

an estimate of the observer of sexual behaviors in a child
and place these estimates in the context of observations
of children in the same group for whom there is no reason to
believe that sexual abuse has occurred. Once the CSBI has
it has been completed, it is very useful to review the estimates of
caregiver. This process will give the examiner the opportunity to
place the observations in context; for example, the
the examiner can learn that a 10-year-old child who
he was caught grabbing his teenage sister's chest (item 7)
two days after having witnessed his sister and her boyfriend
caressing in one of the family's rooms.

This example emphasizes the fact that sexual behavior or

the upbringing of a child must be considered in the context and environment
in which the child develops (Friederich, 1995). During the
investigation with the CSBI, parents have described behaviors that,
despite being statistically very infrequent, they can be
understood when examining the context where these
occurred (a single man in mutual tickling of the genitals of his
niece of similar age while taking a bath). If a caregiver
satisfactorily affirms infrequent items, the interpretation
the examiner's assessment of these statements is exacerbated by means of
to continue with questions considering the context in which the
conduct occurred.

Interpretive examples

The following examples were evaluated comprehensively

using the CSBI only as a component of a long
evaluation process. The examples present eight instances
(only three examples will be included in this translation) where the
CSBI information could be used to shed light on situations
of alleged sexual abuse and together with other information helps to
formulate the case. The examples were extracted from practice

39
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes.
Department of Legal Medicine. Section of Forensic Psychiatry and Psychology. 2006

author's clinic. In each case, the information was also

obtained from interviews, the CBCL, and multiple informants.
Some changes were made to disguise who responded or
to give relevance to some clinical aspect.

Example Case 1

B.B is an 11-year-old boy in a care home. He was

evaluated for sexual abuse as he has been seen trying to
exert coercion over two younger people than him, children
as unrelated to the caregiving home pretending to practice
oral sex. B.B has been in four care homes since his
birth and has a history of physical abuse and neglect due to
both parents and the caregivers he has had. B.B gathered
criteria for a diagnosis of conduct disorder in a
evaluation that was conducted 15 months ago, right after
he started to be cared for in this home.

As part of the evaluation, B.B were interviewed.

foster mothers and their two younger siblings also from
foster care. Her foster mother completed the CBCL and the CSBI, and her
The teacher completed the TRF. The director of the home reported that
she had never seen sexually coercive behavior in B.B;
but she pointed out that he had tried at least once to
hears her daughter while she was in the shower. She also
he stated that he had seen him a few times with his legs open
sitting awkwardly, but she did not describe this as masturbation. More
Well, she thought it was an anxious response.

B.B's foster mother reported that his sexual behavior with

the two younger children had been a complete surprise for her,
in addition to having frequent 'girlfriends' at school. Since
What discovered this coercive behavior in him, she had
found a part of a magazine that contained photos

40
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

explicit from adult women. She assumed it was from B.B. but he
he denied it.

As indicated in Figure 3, B.B's total CSBI score was 3,

which corresponds to a T score of 57. This was based on
items 16 and 35. The T scores of 72 and 44 were
highlighted respectively by the DRSB and the SASI. The scores
The B.B. were significantly elevated in the subscales
from the CBCL.

During B.B's interview, he denied having been a victim of a

sexual abuse. He admitted to having been present when the two
younger children were undressed but he argued that it was
"their idea." He persisted with this denial even after
he had told her that contradicted what her foster mother had said
that he had seen it.

B.B's teacher reported that at that moment he had shown

a moderate decrease in aggressive behavior in the TRF of
your report 15 months ago, as well as an improvement in the
academic performance. She did not report any sexual conduct.
(using the TRF modification described in chapter 5).

The two younger foster brothers of B.B were

interviewed each one separately. One of them reported a
early history of sexual abuse by a male teenager;
this abuse had included oral sex and sodomy. Both young men
they were doubtful about who had started the most
recent attempt at oral sex.

This concluded that B.B's behavior could be

measuredly explained by her behavioral disorder and by her
main story of sexual abuse experienced by one of its
foster siblings. Moreover, B.B's behavior was interesting.
since there was no compelling evidence that he had been

41
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

sexually abused. Closer monitoring was recommended.

from B.B with the other children.

Example Case 2

G. was an 11-year-old girl referred by the psychologist of the

school. She was a girl who had been observed in classes.
during many occasions masturbating until achieving the
orgasm. She was squeezing her legs, both legs together and she
she was squirming in her chair and started to gasp and blush like
if I were reaching the climax. This behavior was communicated
by their colleagues, and G. has so far been rejected by their
friends. G. had exhibited similar behavior two years ago in
the moment the school psychologist was impressed by
how the parents were defensive and how the father controlled
G. during that meeting.

The CSBI and CBCL assessments were completed.

separately by both parents, and the teacher completed an RTF.
Both parents and the teacher described an immature girl.
anxious and compulsive individual. The only
sexual behavior reported by the father was in item 12, touches
his private parts when he is at home, while
that the mother affirmed three items for a total gross score
CSBI of 4 corresponding to a T score of 55: Items 12, 25,
gets angry when adults are kissing or hugging, and 35,
he is very interested in the opposite sex. The mother reported that G
she could occasionally be careless in passing her hand
through his chest or inside his pants when he was watching
a "romantic movie" on television. The mother said she assumed
that her daughter was growing up and she and her husband claimed that
they had never seen a masturbatory behavior in G. like
reported by the teacher.

42
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

The T scores (based on G's mother's report) for

DRSB and SASI were 58 and 45 respectively. See Figure
8.

In the interview, both parents denied any issues regarding

to sexual abuse, but both wanted to get help for G.
that could sleep in his own bed. The fact that G. slept in the
the bed of his mother had been facilitated because the father had
had been out of the house for six months. This was something that he
it happened periodically, and this reflected the chronic marital tension
present.

G. was interviewed separately. She admitted to having

masturbated once at school but claimed that it was not a
problem for a long time because she wanted her
companions would want her again. She attributed this
masturbation because she felt tense and wanted to relax. She
she admitted to occasionally masturbating at home when she wanted
to forget her worries. She denied having been sexually
abused. A self-monitoring system was implemented, and G. was
obedient in keeping watch over his masturbation (did not report
some) and their daily practice of relaxation methods. No
more episodes of masturbation or touching were reported
genitals neither by the teacher nor by the parents for the rest of the year
Collective. The probability of sexual abuse was estimated or assessed.
as very minimal. Child Protection Services
they decided not to investigate further.

Example Case 3

F. was a 4-year-old boy in the middle of a dispute between

his parents for custody. His mother had claimed supposed
sexual abuse by F.'s father, and the father maintained that his
his wife was physically abusing F. Both parents had already
completed an evaluation for custody in which it was found

43
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

in each of them severe personality problems. Each

father argued that his wife or husband had admitted when his
The marriage was going well and was harmonious because they had been sexually.
abused. Those revelations were denied at this time.

Each parent was interviewed and each completed the CSBI and
CBCL. The other significant ones (very close people) of each
father also completed the same measures, as they did
two other adults who knew F. through their work
just like F.'s caregiver.

F. had already been interviewed by local authorities and had

made a statement that his father "put a shovel in my rear end".
A pediatric examination after this statement did not provide
evidence consistent with anal penetration.

F.'s mother reported numerous sexual behaviors. She

affirmed 20 items, resulting in a total gross CSBI score of 39
which corresponds to a T score of over 110. F. obtained a
T score in DRSB of 91 and a T score in SASI of more than 110
(See figure 7). Reports from other adults do not report being
they are slightly close to the mother's report even still
together they spend more time with F. than their
mother; their total gross scores CSBI estimated from 2 to 8, with
all appropriate developmental behaviors. The same was
true for the CBL, where F's mother reported problems of
significant behavior in all areas, but the caregivers
they typically reported behavior within the limits
normal with the exception of somatic complaints. Nevertheless,
both parents were health professionals and each of
they accused the other of only addressing complaints about pains
head, stomach, and other physical ailments. The scores
CSBI of F. based on the report of its progenitor are
represented in Figure 7.

44
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

F. was not very revealing on all topics. He was very...

circumscribed in his game with the author, who also observed him in
game with their parents. Both were reasonably
attentive, and F obviously enjoyed this interaction. It wasn't noticeable.
no sexual conduct with the exception that his mother did
she spontaneously kissed on four separate occasions. F
he responded to this by withdrawing from her briefly and then playing.
alone. This seemed very different from F. being
cornered. The mother’s report seemed to reflect more closely
his interests and unique perceptions more than to the reality of the
situation.

Final Comments

It is important to remember that a child's behavior must be

located in the context of her family and her experiences of
life. Isolated behaviors, despite being unusual from the point
From a statistical point of view, they may have little or no importance.
clinic if they are never seen again or if they are abandoned
after reorienting the caregiver. The context of the behavior, the
persistence of behavior and the range of exhibited behaviors
should be considered and examined in reference to others
aspects of the functioning of the child. Moreover, it is important
examine the items included in the total raw score CSBI and
the results of both scales the DRSB and the SASI for
differentiate appropriate behaviors for further development
specific abusive behaviors.

Items not included in the SASI Scale

The careful reader will notice that there are two items that do not appear in
none of the six age gender scales SASI (Items 1, Visually
like the opposite sex and 36, Puts her mouth on the bust of her
mother or other women). There are 6 items that appear in
only one of the age scales gender SASI and another 10

45
 Translation done by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes.
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

items that appear only in two age scales gender

from SASI. A total of 8 items appear in all 6 scales.
age gender.

These items illustrate the dilemma of using behaviors that have

a low base estimate, or frequency of occurrence to predict
any phenomenon. A review of the simple frequencies
affirmed listed for both genders in all three
groups (See tables 4 and 5) typically illustrate that these behaviors
are shown more frequently or at a higher intensity,
for sexually abused children. However, these are
also behaviors that are generally not reported
frequently by older age groups. This phenomenon creates
a low base estimate and makes differences between groups more
difficult to determine using conventional statistics.

When this is considered with the relatively short number of

sexually abused children compared to the number of
children from each age group in the normative sample, it is not
surprise failing to find significant differences
statistics. In summary, the use of covariates (education
maternal and family income) also resulted in a more focused approach
conservator of identifying differences.

The results once again point to the usefulness of the Total score.
CSBI, which reflects all the behaviors included in the scale.

The results also indicate the importance of not using items

solos, out of context, to reach a determination.

46
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes.
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

4
Description and Regulatory Information

The normative data from the CSBI to represent 1,114 children.

combined from three non-clinical samples: (a) 723 children
who was in the waiting area of the Clinical Community
pediatric in Rochester, Minnesota; (b) 111 children who
they were in the waiting area of a Community of Medicine
Clinic in Rochester, Minnesota; and (c) 280 children from the area
Los Angeles, California, most of them with low income and a
minority status.

Influence of Demographic Variables

Several analyses were conducted on the normative data to evaluate

the potential influence of demographic variables on
raw scores of the CSBI (non-standardized). First, the
affirmed frequencies and item means were examined through
from each age represented to determine any variation
related to age in the scores. This procedure suggested
what more than two age groups (2-6 years and 7-12 years) were used
primarily in the CSBI research (Friederich 1992;
Friederich 1993), three more groups they reflected precisely
variations in development (2-5 years, 6-9 years and 10-12 years;
see Figure 1).

The relationship of a number of demographic variables to the score

The total CSBI was calculated based on the entire normative sample.
using multiple regression. Four variables were recorded
as a block (family income, gender, age in years of the child
Mother's education). Age (F= 9.3, p<.00001) and
maternal education (F= 5.0, p <.00001) were
significantly related and together they amounted to

47
 Translation carried out by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

approximately 19% of the explained variation in the scores.

However, neither gender nor family income were
significantly associated. These findings indicated that the
younger children have significantly higher scores
taller than the older children, and the mothers with more years of
schooling reported more sexual behavior than what it
they did those with lower education.

The influence of various child and family variables was

examined by the normative sample. The variables child
they included the quality of peer relationships, ethnic status
(Caucasian, non-Caucasian) and hours per week in care. The
family variables included marital status (single or not)
daily stress, domestic violence (the presence of physical abuse)
of the child or of the spouse), the total number of children in the
family, and the sexuality of the family (a more relaxed approach to
co-sleeping, co-bathing, family nudity, opportunities
to watch movies or adult magazines and witness a
sexual exchange.

The block of family variables previously described (

income, age and gender of the child, maternal education were
registered first. The next step was to register one of the
additional variables.

Of the eight additional variables evaluated, four were

significant at the p<.05 level or better: domestic violence (F =
2.1, p<.03), hours per week in care (F = 3.8, p<.0001)
daily stress (F = 2.8, p<.005) and family sexuality (F =
9.9, p<.00001). The only variation, reflected in the change in R 2,
it was less than 1% for domestic violence and daily stress, 1.2%
for four hours a week in care and 7.5% for sexuality
of the family.

48
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

Apparently the total number of hours of care per week

contributes significantly and statistically to increase
on reporting child sexual behavior in children for whom
There is no reason to suspect sexual abuse. This reason is more
difficult to determine without further information. In addition, the age was
a co-variable in the regression analysis, the younger children
typically spend more hours per week in care, and the children
the younger ones in the normative sample were more about
sexuals than those of the older children.

Parents who reported more family sexuality also

higher levels of sexual behavior were reported in children
from 2 to 12 years of age, even after controlling for the effects
of many more family variables. These findings make
emphasis on the fact that the sexuality of the family is one
a very important contextual variable that must be considered
when evaluating the possible reasons related to behavior
child sexual abuse.

The parents in the normative sample were also evaluated.

regarding whether they agreed with the item 'is
it's normal for children to have feelings and curiosity
sexual". Once again, controlling for the effects of age, gender,
family income and maternal education, the response of the parents
this question was significant (F = 2.2, p<.03), which means
to say that the parents who agreed with this item
They reported more sexual behavior in the CSBI than the parents who did not.
they agreed with this item. Nevertheless, the
differences in the answers to this question totaled less than
the 1% of the only variation.

Based on the demographic analyses reported here and on the

specific gender patterns in the assertion of certain
items, both age and gender were considered as
important variables under which to standardize the CSBI. The

49
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic Research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

norms based on racial groups were not derived by giving a

insignificant contribution of race after it was
considering the influence of maternal education and income
familiar.

Calculation of the Standards

The normative data were derived from the Total CSBI and the
DRSB scales, and SASI of data standardization of the
show, based on the six combinations of age - gender (
boys and girls aged 2 to 5 years and boys and girls aged 10 to 12 years of
age). The T scores for each scale, according to age and the
gender, appear in the Appendix. The Averages and Deviations
Standard of gross scores of the CSBI for each group of
age and gender in the standardization sample appear in the
Table 10. This table also presents the means and deviations.
standard of a sample of 512 abused children
sexually.

50
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

5
Development and Validity

The development of the CSBI began in 1983 when it started to become

Of course, the sexual behavior items in the CBCL (Achenbach,
1991) were very useful in discriminating sexually against children
abused of the not sexually abused (Friederich, Beilke, and)
Urquiza, 1987). New and behaviorally specific items
were written and combined with reported sexual behaviors
during interviews conducted with child caregivers
sexually abused. Forty items were initially
developed and used in a pilot study of 71 children
abused and 35 children sexually abused (Purcell, 1986).

The mean scores of the two groups differed.

significantly one from the other.

The original reservation of 40 items was then expanded to 48.

items, and 16 of the 40 original items were rewritten. These
changes were made after interviews with 32
caregivers of sexually abused children
identification of an additional number of sexual behaviors
observed in these children. The 48 items included 3 items
about the child's exposure to family nudity and 5
items belonging to somatic behavior. During the
data collection, the parents have indicated that the 5 items
additional ones were poorly drafted. In addition, 13 out of 48
items were taken from the final analysis, leaving a total of 35
items.

This 35-item version of the CSBI was later studied

(Friederich, 1992) with 880 children, ages 2 to 12 years,
selected for the absence of sexual abuse and 276 children with
double documentation of history of sexual abuse. The score

51
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

gross Total CSBI differed between the two groups after the
control for age, gender, maternal education, and family income.

Sexually abused children differed from children.

sexually not abused in 27 out of 35 items. In each case,
the sexually abused children were more likely to
sexual.

The scale of item 35 was redefined, with 6 items eliminated, 3

items re-drafted, and 7 items added to the 27 items that
they differed significantly in the study by Friederich (1992).

This version of 36 items was then reviewed with a new

normative sample (N = 141) and a sexually abused sample
(N = 133). Among the group differences, 35 were significant.
of the 36 items. (Friederich, 1993).

The current version of 38 items was based on the study of

Friederich in 1993. Of the 36 items, 22 were retained, 3
twelve items were rewritten for clarity and
simplicity.

In addition, two validity items were added as part of

a normative study: Play with a friend (changed to You wore shoes)
in the test re test) and have breakfast. These two items of
validity were designated to determine if each item in the
measure was read by the caretaker who completes the CSBI. They
they were added because some parents had told him
the clinicians that the listed sexual behaviors were somewhat
difficult to read given their own history of sexual abuse. These
parents reported that, instead of reading each item, they
they simply enclosed a response, typically 0. This
concluded that the appropriate affirmation of the validity items
closing a 3 for the item, You wore shoes indicated the possibility
that the caregiver had read each item.

52
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

A decision was made to assess the attitudes of parents.

about the normality of sexual behavior in children.
In addition, for the 38 behavioral items, parents were asked
respond with 'true' or 'false' to the statement 'It is normal that the'
children have feelings and sexual curiosity." The parents in
the normative sample who answered 'false' estimated their
child significantly in general as very low.

Validity Scales

Despite the fact that the two validity items helped identify
parents who read each item, it was decided that, given the brevity of the
in measure, the validity items will not provide sufficient interpretation.
Therefore, these items were discarded from the version
CSBI current. In addition, other reporting measures for
longer parents (the CBCL) do not contain validity items. It is
recommended that, during the following interviews with the
caregiver, the clinicians determine precisely if each item
It was read and interpreted appropriately.

DRSB Scale

During the use of the CSBI, it started to become evident that

certain items were useful in helping to determine if a
particular conduct was from the perspective of development
appropriate or atypical, consequently increasing the possibility of
a story of sexual abuse. These behaviors in the six groups
different genders and ages that were affirmed by at least
20% of the mothers of the children in the normative sample
as having occurred at least once in the last six
months were included in the scale of Development related to
Sexual Conducts (DRSB). The number of items in the scale
DRSB ranges from a low one for boys and girls

53
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

ages 10 to 12 years old, up to a maximum of five for

children aged 2 to 5 years.

The 20% level was chosen because it included more children.

/as that in a clinical cut-off of 1.5 standard deviations over the
media (above 9%), or a cut of 1.0 standard deviation
about the average (more than 1.7%). In addition, the behaviors included in
the DRSB scale was all significant, usually the
contrary, correlated with age, even when it is done the
control of maternal education, family income and gender. By
For example, the behavior "stays close to people" is in the
DRSB scale for children aged 2 to 5 years and girls aged 2 to 5
years old, but it is not included in the age scale
adults. Another behavior, "is very interested in sex
"opposite" is an item on the DRSB scale for both genders in
age groups from 10 to 12 years old but not for any gender
age groups of 2 to 5 years and from 6 to 9 years, and this the most
It is likely that it is related to normal maturation.

The DRSB scale was developed to create great objectivity

about the behavior in children. These items illustrate that
hay many behaviors sexual what no they need
automatically be seen as pathological. For example, the
knowledge that 60.4% of children aged 2 to 5 years
touching their genitalia at home may normalize the reaction
from the parents who have begun to see them for the first time
at four years old, touching their own genitals.

The DRSB scale correlates with age, other issues with

behavior, sexuality of the family as well as with a history of
sexual abuse. As mentioned in chapter 3, the scores
elevated can be considered as another manifestation
to externalize behavior, particularly where the family is
comfortable regarding sexuality and has generated a
interruption or alteration in the parents.

54
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic Research purposes
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

SASI Scale

The SASI scale was developed to capture more

precisely the relationship between sexual abuse and sexual conduct.

For example, in a multiple regression analysis, sexual abuse

adds 23% of the variation in the prediction of the gross score
Total CSBI for children aged 2 to 5 years, while
that sexual abuse accounts for over 37% of the variation in the
prediction of the gross score of the SASI for the same group of
children. In this same analysis, the relative predictive power of the
family sexuality is significantly reduced when
recorded in an equation predicting the raw SASI score,
although it remains a significant predictor of the raw score
total of the CSBI and the gross DRSB score for all groups of
age and gender.

The differences between the groups (normative samples of abuse

sexual) for the six age groups - gender were
determined using ANCOVA for all 38 items of the
CSBI by order to identify items to be used in
build the SASI scale. The covariates for each analysis
they were the family income and maternal education. Due to the
Number of analysis, a probable value of .005 was established for
include it in this scale. Despite this probable value
excluded numerous items that differed significantly
probabilities, this determined that a strict criterion for the
inclusion was necessary. The number of items was included in a
range from low 11 to 14 for boys and girls from 10 to 12 years old,
respectively, and over 22 and 23 for boys and girls aged 2 to
5 years old respectively.

In summary, although the items of the SASI scale are more

directly related to the experience of sexual abuse, the

55
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

the same interpretative precautions apply to the score

Thus, the total score of the CSBI: sexual abuse cannot
predict based on a single test.

Standardization of the CSBI

Item 38 of the CSBI, with the two validity items and the item of
the father's attitude was administered to the three situations
regulations to complete the diversity of the sample (See
Chapter 4). In addition, for validation purposes, the following was obtained
a clinical sample (N = 512) with a documented history of
sexual abuse from various places in the United States of North America
and in Canada. The staff members of centers where those
children were being seen reported that the children
they have been abused and that is why they had been separated, so
soon as determined by the social services of sexual abuse.

The characteristics of the children in the abuse sample

sexual sons presented in Tables 11 and 12.

The data in these children were collected from the

following sites: Seattle WA (n = 111), Tulsa, OK (n = 19);
Pórtland, OR (n = 61); Pórtland, ME (n = 8); Pittsburg, PA (n = 9);
San Fernando Valley, CA (n = 72); Rochester, MN (n = 68);
Burlington, VT (n = 110); Calgary, AL (n = 39); and Montreal, QU (n
= 15). The total sample included 512 abused children.
sexually those who had been offended in that year
mainly. The average time since it elapsed
abuse was 10.9 months (SD= 13.9 months). The average duration
the abuse was 9.3 months (SD = 12.3 months). A total of 55 of
the children sexually abused in the abuse sample
They had medical findings consistent with sexual abuse.

The demographic characteristics of the sexual abuse sample

clearly indicate the lowest levels of education in the

56
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic Research purposes
Department of Legal Medicine. Section of Forensic Psychiatry and Psychology. 2006

parents and family income. For these reasons, all the

mean differences between the analysis groups used
ANCOVA, with maternal education and family income as two of
the co-variables.

Table 11
Age and education characteristics of the sample of abuse
sexual of the CSBI.

Demographic Characteristics M SD
Age (Years) 7.44 2.63
12.24
Educación paterna (años completados) 12.13 2.80
Note. N = 512

Reliability and Validity

The following sections describe the reliability and validity of

CSBI in the normative group and in the sample of children
sexually abused.

Reliability

The reliability analysis of the total CSBI scale in the

The regulatory sample demonstrated a reasonable internal consistency.
(r a = .72). The alpha coefficient for the Total CSBI scale in the
the sample of sexually abused is .92. When analyzed for
each of the three groups discarded by gender, the
Alpha coefficients are similar. That is, for the sample
According to the regulations, the alpha coefficients are .72 for ages 2 to 5.
71 for ages 6 to 9 years, and 71 for ages 10 to 12.
years. For the sample of sexual abuse, the alpha coefficients for
the same age groups are .93, .91, and .91, respectively. The
Very high reliability in the clinical sample is common in measures

57
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic Research purposes
Department of Legal Medicine. Section of Forensic Psychiatry and Psychology. 2006

clinics and this is related to the increased variation for the

scale in the clinical sample.

Test – Retest of Reliability

The test-retest reliability was calculated for both.

versions: for the version of 35 items and the current version of
38 items of the CSBI. The study of the 35-item version.
(Friederich, 1992) used 94 parents in the normative sample, 70
of those who completed a second protocol again in
average of 4 weeks later (74.5%). The test correlation -
retest was .85. It was obtained from three months of test data.
retest of 24 sexually abused children, resulting in a
significant correlation of .47 (p<02). All 24 of these
children had received some therapy in the meantime, and it
I hoped that his behavior could change over time.

For the version of 38 items, 63 out of 87 parents (72.4% conclusion)

they completed a second CSBI with an average of 2 weeks
after. The test-retest correlation was .91. Taken together,
these results strongly support the test reliability
CSBI retest.

Inter-rater reliability

Inter-rater reliability is very important because

which directs matters of objectivity considering conduct
sexual in children as well as the clarity of the items
individuals of the CSBI.

The estimates of father and mother in item 38 of the CSBI

were correlated by 23 pairs in the normative sample.
These parents were married and lived together. The correlation
for the total gross score of the CSBI was high and significant (r =

58
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic Research purposes
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

.43, p<.05). Taken together, these results support the

inter-rater reliability of the CSBI.

Validity

Inter correlations Scale

The inter-correlations of the clinical scale of the CSBI in the sample

regulations were all significant at the .01 level or better in
a range of .35 (DRSB with SASI for children aged 10 to 12 years) to .91
(SASI with Total CSBI of children aged 6 to 9 years). Those
correlations on the scale inter correlations in the sample of
sexual abuse is presented in Table 13.

Convergent and Discriminant Validity

With the development of the CSBI and the publication of its first version
(Friederich 1991), research has established its usefulness. Due to
example. The total gross score CSBI discriminates between pre
sexually abused students and those not sexually abused
(Allen, 1993); Hewitt and Friederich, 1991; Hewitt, Friederich, and
Children
sexually abused in school age also differ from the
children treated for emotional or psychiatric problems
who do not have a history of sexual abuse (Friederich,
Jaworski, Huxsahl, and Bengtson, 1997; Katz – Plotkin, 1991). Another
research has found that the average of the total raw score
CSBI differs significantly between sexually active boys and girls.
aggressive (without a history of sexual abuse) and abused children
and non-aggressive sexually (Bonner, 1996). The average score
total gross CSBI also outlined the sexual aggression of the
physical and non-physical aggression of children (Burton, 1996).

In addition, the total gross score of the CSBI correlates with a

history of sexual abuse but not with other types of mistreatment,

59
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

including physical abuse, psychological abuse, neglect and

emotional abandonment (Higgins and McCabe, 1996).

A recent study found that the mean of the raw score

Total CSBI varied significantly among sexually active girls
abused and two comparable groups of girls aged 6 to 12 years old
age of those who have not been sexually abused (Cosentino,
Meyer-Bahlburg, Alpert, Weinberg, and Gaines, 1995). The
group comparison were girls from a department of
pediatric or psychiatric patients. The authors found that
the items from the CSBI pertaining to boundary problems,
sexual aggression and self-stimulation were the most used for
discriminate between sexually abused and non-abused girls.

In a study of the effects of the treatment, Cohen and Mannarino

they showed the relationship between the CSBI and the Scale for Children
Attributions and Perceptions (CAPS; Mannarino, Cohen, and Berman,
1994), an instrument that evaluates specific attributions of
sexually abused children. The CSBI found to be
related to the CAPS to the measures of the attributions of the
children 6 months after treatment and 12 months after
of treatment. These findings suggest that decreases in
the sexual behavior was parallels for more attributions
upbringing in the child.

Two studies (Friederich and Jawoski, 1995; Friederich 1997) have

examined the relationship of the CSBI to the self-reporting of the aspects
sexual in the TSCC. In both studies, a relationship was noted
significant. However, for the second sample, which was the
larger, the relationship was not statistically significant
after the application of the strict Bonferroni correction. The
relationship of the CSBI to both sexual contents in the Rorschach
(Exner, 1974; Friederich, 1997) and sexual content in the Test of
Perception for children by Robert (Robert and McArthur,

60
 Translation carried out by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes.
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

1982) has also been studied (Friederich and Share). Despite

that the relationships were not significant they were
positive (in the expected direction).

The CSBI is related to reports from other parents.

types of behavior. For example, the total raw score of CSBI was
significantly related to the T scores of the CBCL
Internalizing depression anxiety withdrawal y
Externalizing (aggression, overactivity) in two studies with
earlier versions of the CSBI (Cosentino, 1995; Friederich,
1991). In this sample of 1,114 non-abused children,
the CBCL Internalizing and Externalizing scores were
also significantly associated with the total raw score
CSBI, even after contributions of age, gender,
family income and maternal education were changed first
(F = 5.5, p<.0001, y F = 7.8, p<.0001, respectively).
the same is true for the relationship of the total gross score CSBI and
the six items in the CBCL (F = 8.9, p<.0001). Spitzmueller
(1991) found that the CSBI is significantly
related to the Sensitive Measure Sexual Abuse (SAS;
While, Halpin, Strom and Santilli, 1998) in a study with two
samples of preschoolers.

It has also been found that the CSBI is related to the

estimates of teachers regarding sexual behavior of
the children. The estimates of the children in the Form
Teacher Report (TRF) was available for 31 children.
n(M = age 8.2 years) in an earlier clinical sample
(Friederich, 1992). Three items (play with the sexual parts in
public, sexual games with partners, and sexual problems
[describe] were added to the standardized and valued TRF.
with a 3 for the teacher of the abused child on the same scale
used in the TRF. The scores in the three TRF items were
correlated .36 (p<.05) with item 35 of the raw score
Total CSBI. The same strategy was applied to 43 children.

61
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

age = 8.3 years) in the current clinical sample. They were

valued by their teachers and their mothers or caregivers
feminine in the 38 items of the CSBI. Scores in the three
TRF items correlated .40 (p<.05) with the score
Total gross score of the CSBI. The total gross score of the CSBI for the item
35 differentiated among four distinct groups of children
traumatized (Scarf and Foy, 1991). The authors found that the
parents of the children who were sexually abused
reported significantly higher sexual behavior than it.
reported by the parents of the children who had been
physically abused or the parents of the children
they were traumatized but had not been physically or sexually abused
exploited.

The sexuality of the family (from a more relaxed approach of

co-sleeping, co-bathing, family nudity, opportunities to see
adult movies or magazines and witness exchanges
sexual) there is a consistent correlation between sexuality of the
family and the DRSB and SASI scales for each age group of the
normative sample. As mentioned earlier, the
family sexuality is not a consistent predictor of the
SASI scores in multiple regression analyses, but it is a
predictor for the DRSB scores in the same type of
analysis. However, the fact that the sexuality of the family is
a significant correlation of both related to development and
to the specific behaviors of sexual abuse for many of the
age groups gender indicates the need to consider this
factor in the evaluation of the alleged sexual abuse of the child.

Validity construction

The relationship of various aspects of the experience of sexual abuse

from the child to the Total CSBI score has been studied for both
versions for both the 35 item version and the 38 item version
items. The severity of the abuse, a large number of

62
 Translation done by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes
Department of Forensic Medicine. Section of Psychiatry and Forensic Psychology. 2006

perpetrators, and the use of force were

significantly related to the total raw score of CSBI
item 35 (Friederich, 1992). The detailed information was in
these three variables available for 343 out of 512 children
current group of sexual abuse. From this sample, the score
gross Total was significantly correlated with the
severity of the abuse and the number of perpetrators, but not with the
use of force.

Table 14

Correlations between the DRSB and SASI Scales and Sexuality

Familiar for the Standardization Sample for groups of
age

CSBI Scale No. Of the n Sexuality

items familiar
DRSB
Children from 2 to 5 years old of 5 287.27
age
Girls from 2 to 5 years old of 4 287.21***
age
Children from 6 to 9 years old of 2 191 .25***
age.
Girls from 6 to 9 years old of 2 171.12
age.
Children from 10 to 12 years old of 1 82.16
age.
Girls aged 10 to 12 of 1 96 .19
age.
SASI
Children from 2 to 5 years old of 23 287.27
age.
Girls from 2 to 5 years old of 24 287.12
age.

63
 Translation done by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

Children from 6 to 9 years old of 23 191 .16*

age.
Girls aged 6 to 9 years old from 17 171.33
age.
Children aged 10 to 12 years old from 14 82.24*
age.
Girls aged 10 to 12 years old 96
age.

DRSB = Development related to sexual behavior; SASI = Items

specifics of sexual abuse.
p≤.01.*** p≤.001.

In a separate study (Cosentino, 1995), sexual behavior,

measured by the CSBI, was directly related to abuse
including penetration. Maing (1991) reported a relationship
significant between the severity of abuse and sexual behavior
measured by the CSBI. The CSBI has also proven to be sensitive
specific treatment approaches designated to resolve
the effects of sexual abuse, such as Total CSBI scores have
have been reduced after treatment (Berlinder and Saunders,
1996; Hall- Marley and Damon, 1993). These relationships indicate that
the CSBI is sensitive to the effects of sexual abuse.

The current normative sample (N = 1,114) was contrasted with

the sexual abuse group (N = 512) across all three
clinical scales of the CSBI and for each of the six groups of
age-gender. (See Table 10 for means and standard deviations
standard for both groups. All the differences between
groups were significant (with total raw scores CSBI
higher in the sexual abuse group), even after the
control for the effects of maternal education and family income
(see Table 15).

64
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for the purpose of Forensic Clinical research
Department of Legal Medicine. Section of Psychiatry and Forensic Psychology. 2006

A discriminative analysis was used with the scores of 38 items.

CSBI Total and SASI to determine the group members
(abused not abused) among the six age-gender groups
for which standards were established. The goal of this analysis was
determine if the scores CSBI Total and SASI could
identify the members of the group at a very high level not in a
random level.

The global accuracy classification was calculated for each group.

age - gender (using primarily equal probabilities
for groups). For the Total CSBI score, the accuracy greater than
68% in all cases, and as high as 82.8% for children
from 2 to 5 years old and from 10 to 12 years old. For the score
SASI, the accuracy ranged from 71.4% for girls aged 10 to
12 years old at 86.8% for children aged 2 to 5 years.
highly appreciated accuracy classification on the SASI scale
it typically arises from the decrease in the number of false
positives (labeled not abused as abused) more than of
increase in the number of true positives (labeled
abused as abused). See table 16 for the results.

The number of false positives is consistently very low with

the SASI scales. However, since it is a test of
exploration in non-clinical samples, the CSBI could
ultimately identify the non-abused children as
abused children / abused as abused, even if the
incidence rate of abuse for children aged 2 to 12
Age was marked as high as 15%.

For illustration, if the false positive rate is 10%,

as many as 100 out of 1000 - children, the non-clinical sample
would be incorrectly identified as sexually
abused. These data, which must be considered in the
interpretation of the results of the CSBI once again points to the

65
 Translation carried out by MSc. Paula Vanesa Montero Gutiérrez for forensic clinical research purposes.
Department of Legal Medicine. Forensic Psychiatry and Psychology Section. 2006

need to use only the CSBI as a component of a

more comprehensive evaluation.

Conclusion

The usefulness of the CSBI is directly associated with the field and
knowledge of the professional, as well as primarily of the
information contained in this manual. It is essential that the CSBI
is to be used in combination with other measures and procedures
clinical (interviews with parents and child, other assessment scales
behavior reports from teachers) to better understand
children and ensure their safety.

The usefulness of the CSBI depends on continuous research to

understand the mediators of the experience of abuse and the relationship
of the characteristics of abuse to different aspects of the
sexual behavior in children. Subsequent investigations
they will help determine how the CSBI in combination with others
objective measures can enhance our ability to
identify cases of abuse. For example, using the regulation
current and the abuse samples, the CBCL items
related to sleep problems and PTSD symptoms, in
combination with the Total CSBI score, improves accuracy
global score of the CSBI in identifying sexually abused girls
abused girls aged 10 to 12 years old (Friederich, 1997).

The research is already advancing in the study of behaviors.

sexually intrusive listed in Table 6, as well as others
sexual anxiety items, a minimally evaluated field in the
current version of the CSBI. The research on reports by
the parents is also an advancement on the part of the author. It is expected
that the CSBI continues to be refined and that resources are made available

future versions.

66
 Translation made by MSc. Paula Vanesa Montero Gutiérrez for Clinical Forensic research purposes
Department of Legal Medicine. Section of Forensic Psychiatry and Psychology. 2006

APPENDIX

NORMATIVE DATA
SCORE CONVERSIONS
GROSS TO SCORE T) FOR THE
CLINICAL SCALES OF THE CSBI BY
GENDER AND AGE

67