Child Abuse & Neglect, Vol. 19, No. 8, pp.

933-942, 1995
Copyright © 1995 Elsevier Science Ltd
Pergamon Printed in the USA. All rights reserved
0145-2134/95 $9.50 + .00
0145-2134(95)00055-0

A STUDY OF THE VARIABILITY OF TRAINING AND

BELIEFS A M O N G PROFESSIONALS WHO INTERVIEW
CHILDREN TO INVESTIGATE SUSPECTED
SEXUAL ABUSE

ROBERT I. DAVEY

Consultant in Child and Adolescent Psychiatry, Family Counselling Service, Winsford, UK

JONATHAN HILL

Department of Child and Adolescent Psychiatry,

Liverpool University and the Royal Liverpool Children's Hospital, UK

Abstract--The study aimed to examine the relationship between professional background, number of qualifications,
specific training in child sexual abuse and beliefs about indicators of child sexual abuse in professionals who conduct
sexual abuse investigatory interviews. A questionnaire was sent to all professionals within a discrete geographical
area who were involved in the investigation of child sexual abuse. One hundred and six of 136 questionnaires were
returned (77.9%), 60 of which were from investigatory interviewers. The latter comprised eight different professions.
There was a considerable variation in general training and in training in child sexual abuse specifically, both of which
were more likely to have been received by medical psychological and social work respondents than residential care
workers and police officers. Beliefs about sexual abuse indicators showed a striking variation between respondents.
Most of the variation was not accounted for by the respondent variables examined. Where there was evidence for an
association, the interviewer's professional background had the strongest effect, followed by the number of qualifications
and the number of investigatory interviews conducted per year. Previous training in child sexual abuse had no
significant effect on these beliefs. Further research is needed into the optimal training necessary for members of
different professions in order to change their beliefs about child sexual abuse.

Key Words--Child sexual abuse training, Investigatory interviews, Sexual abuse indicators.

INTRODUCTION

CHILD SEXUAL ABUSE is common and may have serious long-term consequences (Baker &
Duncan, 1985; Finkelhor, 1980; Russell, 1983). The accurate detection of child sexual abuse
is important so that action can be taken to protect the victim and a high sensitivity is required
so that the maximum possible number of victims receive appropriate protection. Equally, a
high specificity is needed in order to ensure that those children who have not been victims do
not undergo lengthy investigations nor become the recipients of inappropriate child protection
measures. The extent of variability of professional training, experience and beliefs regarding
possible indicators of child sexual abuse and influences upon these are therefore of great
importance and provide the focus for this paper.

Received for publication July 9, 1993; final revision received August 29, 1994; accepted November 28, 1994.
Reprint requests should be addressed to Robert Davey, Family Counselling Service, Health Centre, High Street.
Winsford, CW7 2AS, UK.
933
934 R.I. Davey and J. Hill
Background
The investigation of possible child sexual abuse does not fall into the domain of one
professional grouping, nor can the nature of the activity be readily defined. It has been character-
ized as a diagnostic process (Quinn & Schetky, 1989; Vizard, Bentovim, & Tranter, 1987)
yet in many respects this represents an unjustified extension of the term (Glasgow & Bentall,
1989). In some respects, it may be more accurately viewed as a component of a criminal
investigation; however, the age of the alleged victim, the nature of the act and frequently the
relationship with the perpetrator means that this also is a considerable over-simplification.
Some important consequences follow from this state of affairs.
First, training in the investigation of child sexual abuse requires the acquisition of a wide
body of knowledge covering areas such as cognitive, emotional and sexual development in
the child, family dynamics, the law, and a range of interview skills that have been derived from
the integration of this body of knowledge and clinical experience. Second, the investigation of
child sexual abuse will be carried out by professionals of widely differing backgrounds, notably
the police, social services and medical services. Investigators are likely to have professional
values and training experiences which vary considerably, with the possibility that such differ-
ences will influence perceptions and beliefs concerning child sexual abuse. Against this back-
ground, child sexual abuse training is of crucial interest.
Although some studies have found a wide variation in the previous sexual abuse training
of professionals attending child sexual abuse courses (Hibbard, Sewint, & Connelly, 1987;
Hibbard & Zollinger, 1990), none have focused on a sample of all such professionals within
a defined locality. Several studies have examined the effectiveness of relatively brief training
experiences on professionals' knowledge of child sexual abuse (Hibbard, Sewint, & Connelly,
1987; Sullivan & Clancy, 1990) but there has been no research into the comparative effective-
ness of different styles and depths of training.
Professionals involved in the investigation of child sexual abuse perceive physical signs to
be stronger indicators of abuse than nonphysical signs (Boat & Everson, 1988). Of the psycho-
logical and behavioral signs, there is general uncertainty over which can be considered to
constitute reliable indicators. Although there is a large body of research into this topic, the
interpretation is difficult for a number of reasons:

1. The subjects are those who have come to the attention of the medical or social services.
These are estimated to constitute only one quarter to one third of the population of sexually
abused children (Finkelhor & Hotaling, 1984) and may be an unrepresentative subsample.
2. Such studies do not control for the family dynamics underlying the abuse, the reaction of
the family to the disclosure of the abuse or the subsequent management of the family by
professionals, which may include the removal of the child or the abuser from the family.
All of these may bear upon the psychological and behavioral status of the child.
3. The studies take no account of the prevalence of the signs in question in the general
population of nonabused children. According to Bayes' Theorem (Phillips, 1973) the proba-
bility that a child exhibiting indicator X has been sexually abused is given by the equation:

Probability Proportion Proportion of Proportion of

that a child = of abused x abused children - children with
with X has children in general X in general
been abused with X population population

Although controlled studies of sexually abused children suggest that the only sign specific
to sexual abuse, as opposed to other forms of abuse, is an increase in sexualized behavior and
sexual interest (Einbender & Friedrich, 1989; Gale, Thompson, Moran, & Sack, 1988; Golds-
 Training and beliefs of sexual abuse investigators 935
ton, Turquist, & Knutson, 1989; Livingston, 1987), the paucity of research into the nature and
extent of sexual behavior in the general population of children precludes its use as an "indica-
tor" of sexual abuse.

Aims of the Study

The study aimed to establish the professional backgrounds, general training (measured by
the respondent's number of qualifications) and specific training in child sexual abuse of all
investigatory interviewers within a given locality, to elicit the extent of variability of beliefs
regarding possible indices of child sexual abuse and to determine the extent to which that
variability could be accounted for by differences in profession and training.

METHODOLOGY

Sample
Inquiries were conducted in the County of Ciwyd, Wales (in the UK) and adjacent areas,
to establish which professionals were involved in the investigation of suspected sexual abuse.
Clwyd covers an area of 2,477 Km z, comprising both rural and urban areas. Its main town is
Wrexham, population 112,000, which contains both affluent and socially deprived communities.
These characteristics indicate that Clwyd is not an unusual area in that it contains communities
with a variety of socioeconomic characteristics. In addition, it had a policy for the investigation
of child sexual abuse (Clwyd Area Child Protection Committee, 1988) and a newly established
1-day training program for professionals dealing with sexual abuse.

Instrument
Those identified were sent an anonymous questionnaire which assessed the respondent's
profession, place of work, qualifications, number of years qualified, number of investigatory
interviews conducted in the previous year, training in child sexual abuse, age, sex and other
demographic variables. The respondent was asked to rate 22 psychological, behavioral and
physical signs and symptoms as "not indicative," "mildly indicative," "strongly indicative,"
or "convincing proof" of sexual abuse (see Appendix).

RESULTS

One hundred and six of 136 questionnaires were returned (77.9%). Of these, 60 156.6%)
were from respondents who conducted investigatory interviews. Since the investigatory inter-
viewers were of primary interest, those who did not personally conduct such interviews were
excluded. The investigatory interviewers were from eight different professional backgrounds:
one pediatrician, two child psychiatrists, three clinical psychologists, one educational psycholo-
gist, 30 local authority social workers, three NSPCC (a charitable organization) child protection
officers, nine residential care workers, and 11 police officers. For the purpose of analysis,
respondents were grouped into four professional categories: a "medical/psychological" group,
which included the child psychiatrists, the clinical psychologists, the educational psychologist
and the pediatrician (n= 7); a "social work" group, which included the social workers and
child protection officers (n = 33); residential care workers (n = 9); police officers (n = 11).
Respondents from the medical/psychological and social work groups were significantly older
and had significantly more children than those in the other two groups. The average number
936 R. I. Davey and J. Hill

Table 1. Investigatory Interviewers' Level of Training in Child Sexual Abuse

CSA Covered
in Basic Attended CSA Attended
No CSA Training or Seminar(s) or CSA
Professional Group Training Other Course Workshop(s) Course(s) Total Variance

Medical/Psychological 0 0 5 2 7 0.24
Social Work 0 2 5 26 33 0.33
Residential Care Worker 5 3 1 0 9 0.53
Police 3 2 2 4 11 1.66
Total 8 7 13 32 60 1.18

p < 0.001, Extended Fisher's Exact Test.

of qualifications per respondent was 1.35 for the medical/psychological group, 1.70 for the
social workers, .56 for the residential care workers and .00 for the police. The overall female
to male ratio was 2.4:1, with all of the police and most of the social workers being female,
compared with about half of the medical/psychological and residential care worker groups.
All these differences were significant at the p < .01 level (one-way Analysis of Variance).
There was a considerable variation in the number of children interviewed by respondents
over the previous year. Twenty had interviewed less than 3, 27 3 to 10, 7 10 to 20, and 6
more than 20. The police and social workers had interviewed significantly more than those in
the other two groups (X 2 = 6.658 [Yates' correction] p < .01).
There was a striking variation in the amount of child sexual abuse training received by
individuals (Table 1). There were significant differences between the professional groups, with
the police and residential care workers more likely to have received no training at all and the
medical/psychological respondents and social workers more likely to have attended a course
(Extended Fisher's Exact Test, p < .001). By assuming the four training categories to be
consecutive points on a continuum of child sexual abuse training, the variance for each profes-
sional group was calculated (Table 1). It can be seen that the variance in child sexual abuse
training was much greater among the police in comparison with the other three groups. There
was a significant positive association between the interviewer's number of qualifications and
his training in child sexual abuse (X 2 [Yates' correction] = 8.484, p < .01).
Table 2 shows respondents' ratings of the 22 psychological, behavioral and physical signs
as "indicators" of sexual abuse. The percentage of respondents rating a sign as "strongly
indicative" or "convincing proof" was calculated to give a measure of "endorsement" for
that sign. The signs have been placed in decreasing order of "endorsement." In order to give
a measure of consensus among raters, the variance was calculated by assuming the four rating
categories to be consecutive points on a continuous scale.
The 22 possible indicators can be divided into three broad groups: physical signs (e.g., semen in
the anus/vagina), psychological/behavioral signs (e.g., school refusal) and sexual signs (e.g., recurrent
masturbation). In general, the physical signs had the highest rate of endorsement and the psychological/
behavioral signs the lowest. Of the psychological/behavioral signs, depression, attempted suicide,
repeated running away from home, and fear of the alleged perpetrator were endorsed as indicators
by the greatest numbers of respondents. With the exception of recurrent masturbation, the sexual
signs had a high rate of endorsement. In the case of recurrent masturbation, there was no difference
in ratings according to whether the subject was a boy or a girl, but there was a marked difference
according to whether the subject was pre or postpubertal: Recurrent masturbation in postpubertal
children was endorsed by only 6.7% of respondents, compared with 30.0% for prepube~al children.
The signs for which there was the most variance in respondents' ratings were detailed knowledge
of sexual intercourse, fear of the alleged perpetrator, and recurrent masturbation in a prepubertal
boy/girl.
 66666666 6 6 6 6 6 6 6 6 6 6 6666

~ ~ ~ - -

.J

O INI~

-..~ b

~ , ~ . ~ ~ "" -~ 0 . ~11~

937
938 R.I. Davey and J. Hill

The responses of the investigatory interviewers were analyzed by the respondents' profes-
sional group, number of qualifications, training in child sexual abuse, and number of children
interviewed over the previous year. Statistical analyses were performed using the Chi-square
test and, where expected frequencies were unacceptably low, the Fisher's Exact Test and the
Extended Fisher's Exact Test. For the majority of the signs (15) there were no significant
associations between the way they were rated and the respondent variables. Table 3 shows
those signs for which there were statistically significant associations.
Four of the significant associations were with the professional group (repeated reference to
sexual matters or themes, recurrent headaches or stomachaches, recurrent masturbation in a
postpubertal girl and recurrent masturbation in a postpubertal boy), two were with the number
of qualifications (depression and recurrent headaches or stomachaches) and two were with the
number of children interviewed in the previous year (recurrent masturbation in a prepubertal
boy and recurrent masturbation in a prepubertal girl). There were no significant associations
with child sexual abuse training.
In order to establish whether the significant associations found were accounted for by the
respondents' sex, age, or number of children, the data were analyzed for associations between
these variables and the way the signs were rated. No significant associations were found.
In each of the four significant associations with professional group, those in the medical/
psychological and social work groups were most cautious in linking the signs to sexual abuse,
followed by the police officers, with the residential care workers most likely to view the signs
as indicative. For the two significant associations with the respondent's number of qualifica-
tions, the more qualified respondents tended to be more cautious in linking the signs to
sexual abuse. For the two significant associations with the number of investigatory interviews
conducted in the last year, those who had conducted more interviews were more cautious in
linking the signs to sexual abuse.

DISCUSSION

This study examined the background, training and beliefs about indicators of child sexual
abuse among professionals who interview children to investigate suspected sexual abuse. In

Table 3. Sign/Symptoms for Which There was a Significant Association Between the Way the Sign was Rated
and Respondent Characteristics
No of Children
Professional Numberof Training Interviewedin
Sign/Symptom Group Qualifications in CSA Last Year

Repeated Reference to Sexual Matters or * NS NS NS

Themes in Play or Speech
Depression (e.g., excessive misery, NS ** NS NS
sadness, loss of interest)
Recurrent Masturbation in a
Prepubertal girl NS NS NS *
Prepubertal boy NS NS NS *
Recurrent Severe Headaches/ ** ** NS NS
Stomachaches
Recurrent Masturbation in a
Postpubertal girl ** NS NS NS
Postpubertal boy ** NS NS NS
Remaining 15 Signs NS NS NS NS
*p < 0.05.
**p < 0.025.
 Training and beliefs of sexual abuse investigators 939
this sample investigatory interviewers represented eight different professional backgrounds.
The majority of such interviews were conducted by police officers and social workers, in line
with local and national recommendations. There was also a considerable variation in interview-
ers' number of qualifications and training in child sexual abuse. Professional background,
number of qualifications and child sexual abuse training were all significantly associated with
one another. Medical/psychological respondents and social workers had the most qualifications
and were the most trained in child sexual abuse. Those with more qualifications had received
more training in child sexual abuse. The most striking finding was the considerable variation
among investigatory interviewers as to which behavioral and psychological signs were per-
ceived as indicators of child sexual abuse. Most of this variation was not accounted for by
the investigatory interviewer characteristics examined. Where there was evidence for an effect,
the professional background had the strongest effect, followed by the number of qualifications
and the number of children interviewed in the last year, both of which had an equal effect.
This was a survey of all investigatory interviewers within a discrete geographical area.
Although it is possible that some were overlooked, preliminary inquiries were thorough and
this number is likely to have been low. The care taken to ensure that all investigatory interview-
ers received questionnaires accounts for the high number of respondents who did not conduct
investigatory interviews.
The response rate was high (77.9%), which lends weight to the findings. The relatively
small sample size and the small number in the professional subgroups dictate that caution
must be exercised in generalizing the findings. Although the use of a questionnaire to assess
the way a respondent investigates sexual abuse is limited, the method used gives a good
indication of respondents' beliefs about the indicators of child sexual abuse.
The variety of different professions represented among the investigatory interviewers, reflects
both the variety of situations in which child sexual abuse may become apparent and the practice
of referring some complex cases on to professions recognized as having an expertise in the
interviewing of children. This highlights the importance of interdisciplinary work in child
sexual abuse.
Medical/psychological professionals and social workers have a longer training than police
officers and residential care workers. This is reflected in a higher number of qualifications.
Naturally, some qualifications might have no relevance to working with children or to the
investigation of sexual abuse.
The variation in the child sexual abuse training of investigatory interviewers was striking,
with a substantial minority who had had no child sexual abuse training at all. This variation
was most marked in the case of the police. The very low level of child sexual abuse training
among residential care workers is noteworthy given that residential care workers are in contact
with sexually abused children on a daily basis.
These findings suggest that training in child sexual abuse is most likely to be received by
those in professions that are characterized by a high level of training generally. It is likely
that professionals who are more highly trained have a greater interest in, and better access to
further training. This paradoxical situation would perpetuate the disparity in sexual abuse
training between individuals.
The perception by most respondents that " h a r d " physical signs such as semen in the anus
or vagina or the presence of sexually transmitted disease are most strongly linked to sexual
abuse was expected and supports the findings of Boat and Everson (1988).
The finding that the majority of respondents believed that repeated reference to sexual
themes and sexualized behavior were strong indicators of sexual abuse is in line with the
literature. It is noteworthy that 73.3% believed that detailed knowledge of sexual intercourse
in a child aged under eight is indicative, given the subsequent findings of Gordon, Schroeder,
and Abrams (1990) that sexually abused children do not differ in their sexual knowledge from
940 R.I. Davey and J. Hill
nonabused children. There is a paucity of objective research data on sexual behavior in normal
children. This is highlighted by the lack of consensus as to whether masturbation in prepubertal
children is linked to child sexual abuse.
It is reassuring that common behavioral conditions, such as aggressive behavior, bedwetting,
and conduct disorder were not generally viewed as being linked to child sexual abuse. Rather
more significance was attached to depression, parasuicide and recurrent stomachaches or head-
aches, possibly because they were seen to be indicative of a high degree of distress. The reality
of these perceived links needs to be tested more vigorously using Bayes' Theorem. The recent
interest in the association between anorexia nervosa and sexual abuse (Palmer, Oppenheimer,
Dignon, Challoner, & Howells, 1990) post-dates this study.
Since the majority of the wide variation in perceptions of the links between specific signs
and child sexual abuse was not associated with any of the respondent characteristics examined,
it is likely that individual factors, such as personal experiences and beliefs, were more influen-
tial. Because of the number of statistical analyses conducted, some of the significant associa-
tions found may have arisen by chance. It is important to note that the significant associations,
where present, were consistent in that greater interviewing experience, a higher number of
qualifications or membership of the medical/psychological or social work groups rendered the
respondents less likely to perceive the signs as sexual abuse indicators. These associations
were not accounted for by variations in respondent age, sex, or number of children.
It is difficult to separate the influence of professional group from that of the number of
qualifications and interviewing experience. However, length of time spent training and working
within a profession is considerably greater than that spent on courses or in interviewing
children, which may explain the relatively stronger associations with the respondents' profes-
sion. Each profession has its own distinctive ethos and objectives that may be unresponsive
to training experiences.
The lack of any significant association between beliefs about indicators of child sexual
abuse and specific training in child sexual abuse received cannot be overemphasized. Even
where training was received, it appears to have had no effect on respondents beliefs about the
indicators of child sexual abuse.

CONCLUSIONS

At the time of this study many investigatory interviewers had been inadequately trained in
child sexual abuse. This issue needs to be addressed by employers, managers, and professional
organizations. Furthermore, the training received had no measurable effect on beliefs about
indicators of child sexual abuse. Prospective studies are needed to evaluate the optimal training
needed by members of different professions to change their beliefs about child sexual abuse.

REFERENCES

Baker, A. W., & Duncan, S. P. (1985). Child sexual abuse: A study of prevalencein Great Britain. Child Abuse &
Neglect, 9, 457-467.
Boat, B. W., & Everson, M. D., (1988). Use of anatomical dolls among professionals in sexual abuse evaluations.
Child Abuse & Neglect, 12, 171-179.
Clwyd Area Child ProtectionCommittee(1988). Guidelines for the investigation of allegations of child sexual abuse.
Clwyd Social Services.
Einbender, A. J., & Friedrich,W. N. (1989). Psychologicalfunctioningand behaviorof sexuallyabused girls. Journal
of Consulting and Clinical Psychology, 57( 1), 155-157.
Finkelhor, D. (1980). Risk factors in the sexual victimizationof children. Child Abuse & Neglect, 4, 265-273.
 Training and beliefs of sexual abuse investigators 941
Finkelhor, D., & Hotaling, G. T. (1984). Sexual abuse in the national incidence study of child abuse and neglect: An
appraisal. Child Abuse & Neglect, 8, 23-33.
Gale, J., Thompson, R. J., Moran, T., & Sack, W. H. (1988). Sexual abuse in young children: Its clinical presentation
and characteristic patterns. Child Abuse & Neglect, 12, 163-170.
Glasgow, D., & Bentall, R. P. (1989). What do expert witnesses in child sexual abuse think they are doing? "'Diagnosis"
and the sexually accurate doll "test" as professional myths. The Liverpool Law Review, XI(I) 43-57.
Goldston, D. B., Turquist, D. C., & Knutson, J. F. (1989). Presenting problems of sexually abused girls receiving
psychiatric services. Journal of Abnormal Psychology, 98(3), 314- 317.
Gordon, B. N., Schroeder, C. S., & Abrams, J. M. (1990). Children's knowledge of sexuality: A comparison of
sexually abused and nonabused children. American Journal of Orthopsychiatry, 60(2), 250-257.
Hibbard, R. A., Serwint, J., & Connolly, M. (1987). Educational program on the evaluation of alleged sexual abuse
victims. Child Abuse & Neglect, 11(4), 513-519.
Hibbard, R. A., & Zollinger, T. W. (1990). Patterns of child sexual abuse knowledge among professionals. Child
Abuse & Neglect, 14(3), 347-355.
Livingston, R. (1987). Sexually and physically abused children. Journal of the American Academy of Child and
Adolescent Psychiatry, 26(3), 413-415.
Palmer, R. L., Oppenheimer, R., Dignon, A., Challoner, A., & Howells, K. (1990). Childhood sexual experiences
with adults reported by women with eating disorders: An extended series. British Journal of Psychiatry, 156, 699-
703.
Phillips, L. D. (1973). Bayesian statistics for social scientists. London: Thomas Nelson.
Quinn, K. M., & Schetky, D. H. (1989). Resolved: Child sex abuse is overdiagnosed. Journal of the American
Academy of Child and Adolescent Psychiatry, 28, 789-792.
Russell, D. E. H. (1983). The incidence and prevalence of intrafamilial and extrafamilial sexual abuse of female
children. Child Abuse & Neglect, 7, 133-146.
Sullivan, R., & Clancy, T. (1990). An experimental evaluation of interdisciplinary training in intervention with sexually
abused adolescents. Health and Social Work, 15(3), 207-214.
Vizard, E., Bentovim, A., & Tranter, M. (1987). Interviewing sexually abused children. Adoption and Fostering 11(1),
20-25.

R6sum6---Cette recherche s'est pench6e sur des intervenants qui m~nent des enqu&es sur les agressions sexuelles,
afin de voir s'il existe un lien entre, d'une part, les facteurs tels que leur profession, leur scolarit6, la formation qu'ils
ont reque concernant les agressions sexuelles et, d'autre part, les indices de maltraitance qu'ils consid~rent importants.
On a envoy6 un questionnaire ~ tousles intervenants professionnels qui s'occupent d'enqu&er sur les abus sexuels
dans une r6gion g6ographique d61imit6e. Panni les 136 envois, 106 questionnaires (77,9 p.c.) ont 6t6 remplis et remis
aux chercheurs; 60 venaient de personnes qui menaient des entrevues d'engu~te, parmi lesquelles huit professions
diff6rentes 6taient repr6sent6es. Leur formation variait beaucoup et c'6tait principalement les travailleurs sociaux et
le personnel des soins de sant6 qui 6taient les plus form6s, en contraste avec les policiers et le personnel s'occupant
des enfants en milieu r6sidentiei. Ce que les intervenants consid&aient des indices importants variait 6norm6ment
d'un intervenant h l'autre. Aucune des variables ou presque ne pouvait expliquer ces variances. Lh o~ on a not6 un
lien, les facteurs d'importance 6taient principalement ceux de la profession, suivi ensuite de la scolarit6 et du hombre
d'entrevues effectu6es. La formation portant sur les agressions sexuelles ne semblait avoir peu d'influence dur leurs
croyances au niveau des indices. Il faudrait pousser les recherches pour d6terminer quel est le niveau de formation
n6cessaire pour que les intervenants des diverses disciplines puissent modifier leurs croyances concernant les indices.

Resumen--El estudio tiene como objectivo examinar la relaci6n de los antecedentes profesionales, el nfimero de
titulaciones y el entrenamiento especifico en abuso sexual infantil con las creencias acerca de los indicadores de abuso
sexual infantil en profesionales que Ilevan a cabo entrevistas de investigaci6n de abuso sexual infantil. St remiti6 un
cuestionario a todos los profesionales que estuvieran implicados en la investigaci6n de casos de abuso sexual infantil
dentro de un firea geogr~ifica concreta. Se recogieron 106 (77.9%) de los 136 cuestionarios remitidos. Sesenta de ellos
pertenecian a sujetos dedicados a la entrevista de investigaci6n. Entre ellos, se encontraron ocho tipos de profesiones
diferentes. Se observ6 una considerable variaci6n en la formaci6n general y en la formaci6n especifica en abuso
sexual. Ambos tipos de formaci6n hablan sido recibidas m ~ frecuentemente por los m6dicos y trabajadores sociales
que por los trabajadores de instituciones residenciales y los oficiales de policla. Las creencias acerca de los indicadores
de abuso sexual infantil mostraron una notable variaci6n entre los sujetos que contestaron al cuestionario. La mayoria
de las variaciones no fueron explicadas por las variables estudiadas en los profesionales. Donde se produjo una
evidencia de asociaci6n, los antecedentes profesionales de los entrevistados tuvieron el efecto m~is importante, seguido
pot el nfimero de titulaciones y el ntimero de entrevistas de investigaci6n llevadas a cabo en cada afio de trabajo. La
formaci6n previa en abuso sexual infantil no tuvo efecto significativo en dichas creencias. Es preciso realizar m~s
investigaciones con respecto al entrenamiento que necesitan los miembros de diferentes profesiones para cambiar sus
creencias acerca del abuso sexual infantil.
942 R. I. Davey and J. Hill
APPENDIX

You may have come across some of the following behaviours in children. Please rate each
according to whether you feel they are:

"not indicative of sexual abuse", i.e. the behaviour commonly occurs in children who have
not been sexually abused and its presence would not normally arouse your suspicion that
sexual abuse is a possible cause;

"mildly indicative of sexual abuse", i.e. the behaviour commonly occurs in children who
have not been sexually abused but its presence would arouse your suspicion that sexual abuse
is a possible cause;

"strongly indicative of sexual abuse", i.e. this behaviour in a child indicates a strong
possibility that sexual abuse has occurred;

"convincing proof of sexual abuse," i.e. if this behaviour is present in a child then sexual
abuse has definitely occurred.

Not Mildly Strongly Convin-

Indica- Indica- Indica- cing
tire tive tive Proof

(The 22 signs) n Q Q Q