Alleged Child Sexual Abuse

Alleged Child Sexual Abuse

Alleged Child Sexual Abuse:

The Expert Witness and the Court

Clara Hellner Gumpert

Stockholm 2001

Dissertation for the degree of Doctor of Philosophy presented at Karolinska institutet in 2001.

Hellner Gumpert, C. (2001). Alleged child sexual abuse: The expert witness and the court. Stockolm, Sweden: Karolinska institutet, Department of Public Health Sciences, Division of Psychosocial Factors and Health.ISBN 91-628-4685-X

Key words: Child sexual abuse, expert witnesses, expert testimony, forensic evaluations, statement analysis, Formal Structure Analysis, Criteria-Based Content Analysis, Statement Validity Analysis.

Clara Hellner Gumpert, M.D.

Karolinska institutet, Department of Public Health Sciences, Division of Psychosocial Factors and Health.

 Clara Hellner Gumpert

ISBN 91-628-4685-X


Hellner Gumpert, C. (2001). Alleged child sexual abuse: The expert witness and the court.

Stockholm, Sweden: Karolinska institutet, Department of Public Health Sciences, Division of Psychosocial Factors and Health. ISBN 91-628-4685-X.

Background During the past decades, the evaluation of alleged sexual abuse has manifested itself as a major challenge for professionals working within the field of child maltreatment. A new role for psychologists and psychiatrists has been to give expert opinions regarding the credibility and reliability of child witnesses in legal proceedings. Although some aspects of evaluating suspected sexual abuse cases are close to traditional clinical work, other aspects necessitate new kinds of empirical and theoretical knowledge. Consulted experts have sometimes been accused of not meeting the new and different demands connected to such assignments. Professional efforts to determine whether a child has been the victim of sexual abuse or not has been the focus of an ongoing and heated debate in popular as well as scientific media. The overall aim of this study was to systematically explore how expert witnesses consulted in legal cases concerning alleged child sexual abuse have performed their task and collaborated with the courts.

Method The study was based on reviews of court files and written expert testimony (mainly focusing psychological expert testimony). Data from 800 court cases of alleged child sexual abuse was included (all cases from Swedish district courts during four different years; 1985, 1989, 1992 and 1997). A combination of quantitative and qualitative methods were applied.

Results The participation of expert witnesses decreased during the study period, from 25% of the cases (1985 and 1989) to 15% (1992) and 6% (1997). Logistic regression analyses showed that psychological expert participation was significantly associated with cases with young victims (< 7 years of age), cases with intra-familial abuse, and cases in which the suspect denied the charges (Paper I). The findings of the second study indicated that there have been communicative difficulties between expert witnesses and the courts. It appeared as if certain basic prerequisites for reaching a level of meaningful communication were not always achieved. Such prerequisites included a legally relevant focus for the expert’s evaluation, a mutual understanding of certain central concepts, and a thorough description of the expert’s reasoning, which would allow the court to review and analyse the expert’s conclusions.

The concept of ‘quality’ with regard to written psychological expert testimony (formal structure and content) was systematically approached in the third study, through the development of a reliable 12-item assessment tool (Structured Quality assessment of eXpert witness testimony [SQX-12], Paper III). In the fourth study, this assessment tool was applied on 121 (all available) expert witness reports from four different years, focusing the credibility or reliability of the child witness. Although the quality assessments according to the SQX-12 improved over the study period, the results indicated that testimony concerning alleged child sexual abuse failed to follow many of the recommended standards. Expert reports produced by professionals applying statement analysis usually scored higher on the SQX-12 than did those written by more clinically oriented professionals. Published guidelines appeared to have had limited influence on actual practice (Paper IV). Professionals applying statement analysis as an assessment tool varied with regard to what features of the child’s statements were emphasised and how such characteristics were evaluated. Special difficulties were connected to the interpretation of ‘low quality’ statements (Paper V).

ConclusionsThe results of the study illustrates various complex aspects of the evaluation of alleged child sexual abuse in a legal context. Different interventions are discussed and suggestions are given for further development of collaboration between psychological experts and the courts. In addition to an apparent need for improved quality with regard to written psychological testimony, other major issues concern how experts can provide the courts with legally relevant information, and how courts can make use of complex expert

knowledge and still uphold the integrity of the legal process.


Alleged child sexual abuse: The expert witness and the court


This thesis is based on the following papers, which will be referred to in the text by their Roman numerals. For copyright reasons the papers are not published on this website.

Paper I. Gumpert, C. H., Lindblad, F., & Johansson, A. (1999). Child sexual abuse: Expert testimony in Swedish district court. Child Maltreatment, 4, 343-352.

Paper II. Gumpert, C. H., & Lindblad, F. Communication between courts and expert witnesses in legal proceedings concerning child sexual abuse in Sweden: A case review. Accepted for publication (Child Abuse & Neglect).

Paper III. Gumpert, C. H., Lindblad, F., & Grann, M. (2000). A systematic approach to quality assessment of expert testimony in cases of alleged child sexual abuse. Psychology, Crime & Law. Accepted for publication.

Paper IV. Gumpert, C. H., Lindblad, F., & Grann, M. (2000). The quality of written expert testimony in alleged child sexual abuse: An empirical study. Psychology, Crime & Law. Accepted for publication.

Paper V. Gumpert, C. H., & Lindblad, F. Statement analysis in Sweden: A review of expert testimony regarding alleged child sexual abuse. Accepted for publication (Expert Evidence).






Expert testimony......

Psychological expert testimony and child sexual abuse......

The Swedish debate......

Assessing quality in forensic evaluations......






General perspectives......

Papers I, III and IV: Quantitative studies......

Papers II and V: Qualitative studies......

Ethical considerations......


Child sexual abuse: Expert testimony in Swedish district courts (Paper I)....

Communication between courts and expert witnesses in legal proceedings concerning child sexual abuse in Sweden: A case review (Paper II)

A systematic approach to quality assessment of expert testimony in cases of alleged child sexual abuse (Paper III)

The quality of written expert testimony in alleged child sexual abuse: An empirical study (Paper IV)

Expert testimony on child sexual abuse: A qualitative study of the Swedish approach to statement analysis (Paper V)


Major findings......


Future perspectives......

Concluding remarks......




CBCACriteria Based Content Analysis

CAPChild and Adolescent Psychiatry

FSAFormal Structure Analysis

ICCIntra-class correlation

OR Odds ratio

ROCReceiver operating characteristic

SVAStatement Validity Analysis

SQX-12Structured Quality assessment of eXpert witness testimony

VCValidity checklist

WPWitness psychologist

The term ‘psychological expert witness’ was used in relation to any expert witness consulted on psychological issues, regardless of whether the expert was a psychologist, psychiatrist or had other professional background.


No crime seems to evoke as much concern and controversy as child sexual abuse. Fergusson and Mullen (1999) named child sexual abuse as “one of the defining cultural themes of our age” (p. 1). Not only has child sexual abuse gained, by far, the most attention of all crimes against children (Jones & Finkelhor, 2001), but virtually every aspect of this crime has been a focus of debate and controversy (Hallberg & Rigné, 1995, 1999). Issues concerning the occurrence of child sexual abuse, the effects of sexual abuse on the child victim, the treatment of abuse victims, and the competence of the child witness are but a few of the topics that have been on the agenda.

From a legal perspective, criminal cases concerning suspected sexual abuse are among the most complicated (Diesen & Sutorius, 1999). Usually, there is no other evidence except the child’s statement. Numerous factors may influence the quality of this statement, a few of them are listed below.

  • The linguistic skills of children are less developed (Lamb, Sternberg & Esplin, 1998). The ability to understand relational terms, such as time, frequency, and duration - terms that may be crucial in a legal context – is dependent on the child’s age and developmental level (Lewin, 1995).
  • Not all children disclose experiences of sexual abuse spontaneously. The child may be reluctant or afraid of discussing experiences of abuse (Berliner & Conte, 1995; Smith, Letourneau, Saunders, Kilpatrick, Resnick, & Best, 2000; Svedin & Back, 1996).
  • Interviewing children for forensic purposes is difficult. A child may be hesitant to talk to an unfamiliar adult, or tense due to unfamiliarity with the situation connected to a formal interview. The interviewer may also be nervous or less experienced in interviewing children, which may affect the ability to establish rapport and encourage the child to freely recall what happened. In addition, the child’s account may be influenced by the type of questions posed by the interviewer (Ceci & Bruck, 1993; Cederborg, 1999; Cederborg, Orbach, Sternberg, & Lamb, 2000; Lamb et al., 1998).
  • The child may exhibit psychological symptoms or signs of emotional stress, which may influence the child’s response to a forensic evaluation. Alternatively, the child may not show signs of stress, which in turn may evoke doubt as to the origin of the allegations, since such signs or symptoms may be expected (Berliner & Conte, 1993).
  • There may be familial conflicts or other contextual circumstances that could be of relevance for evaluating the allegations of abuse (Bernet, 1993; Jones & McGraw, 1987; Mikkelsen, Gutheil & Emens, 1992)

One of the most heated arguments has been connected with professional efforts to determine whether a child has been sexually abused or not (e.g. Berliner & Conte, 1993). Professionals involved in evaluating suspected child sexual abuse have been accused of ‘overdiagnosing’ sexual abuse (Quinn, 1989), due to a lack of integrity and appropriate assessment procedures (Edvardson, 1996; Horner, Guyer, & Kalter, 1993). Others have claimed that the majority of cases remain undetected (Berliner, 1989). The evaluation of abuse allegations is a matter of concern to all professionals involved, not only law enforcement, but also representatives of mental health as well as child protective services. By far, the most delicate task for psychological professionals involves the evaluation of alleged victims of child sexual abuse for forensic purposes, when a psychologist or psychiatrist is asked to take on the role of an expert witness.

Expert testimony

An expert witness is defined as an individual who has been consulted by the court or by either of the parties to give an opinion on a matter that requires special knowledge (Edelstam, 1991). The purpose of expert witness participation in the Swedish legal process is that the expert shall provide knowledge that will increase the court’s competence. According to Diesen and Sutorius (2000), an expert witness statement is to be viewed as “an additional source of information in the court’s decision-making process” (p. 44). The recommendation is that expert witnesses be officially appointed, based on the assumption that a court appointment will guarantee the expert’s neutrality. Either one of the parties may also consult expert witnesses.

Contrary to legal rules in some other countries, an expert witness in Sweden is allowed both to present facts and to give an opinion on facts, such as the reliability of a witness statement or the credibility of a witness. Furthermore, a Swedish court has the right to consider and judge the relevance and strength of any circumstance related to the case (Diesen & Sutorius, 1999; Nyström, 1996). Thus, the issue of what evidence is permissible is usually not of immediate interest. Such matters are often discussed in countries adhering to the adversarial legal principle, such as the United States and Great Britain (Bala, 1994; Myers, 1993).

An issue of importance in Sweden as well as in other countries is whether or not the expert witness violates the court’s domain by expressing an opinion on legal issues. Even though a Swedish expert witness is allowed to give an opinion on facts, there is a broad consensus within forensic literature that the expert is not allowed to give an opinion on a suspect’s guilt (Diesen & Sutorius, 1999s; Insulander, 1996; Nyström, 1996; Wiklund, 1990). In the former national guidelines for the evaluation of suspected child sexual abuse (Socialstyrelsen, 1991), it was recommended that the opinion should be expressed in terms of whether the allegations seemed reasonable or not. It is stressed that the courts need to thoroughly review and analyse the expert’s reasoning (Gregow, 1996; Nyström, 1996; Wiklund, 1991, 1992a).

In a thesis dealing with the rules on and use of expert evidence in Sweden, Edelstam (1991) pointed out that it is reasonable to expect that the need for external knowledge in the courts will increase. The author referred to the overall trend in society towards greater complexity, and the constant development of new and better scientific methods. However, the difficulty to judge whether issues that involve psychological judgements require expert evidence (i.e. ‘special knowledge’) was especially mentioned.

Psychological expert testimony and child sexual abuse

In legal cases concerning child sexual abuse, the issue of psychological expert testimony has been particularly controversial. The role and mandate of the expert witness, as well as the reliability of the assessment methods have been discussed (Berliner & Conte, 1993; Burton & Myers, 1992; Melton & Limber, 1993; Wiklund, 1999). Some psychologists have asserted the benefits of their expertise. For example, the well-known German forensic psychologist Udo Undeutsch claimed that “in his area of specialty the expert is not merely an assistant [to the court] but a veritable master” (Undeutsch, 1982, p. 52). In contrast, the view during the 1990’s appears to be that the role of the expert witness is to educate and inform the court rather than to provide conclusions (Allen & Miller, 1995; Diesen & Sutorius, 1999; Insulander, 1996; McAnulty, 1993; Thomas-Peter & Warren, 1998). Closely connected to the question of expert role is the issue of the expert’s responsibility to recognise the limits of his or her knowledge and assessment procedures (Gregow, 1996; McAnulty, 1993; Myers, 1993), and the duty not to “engage in irresponsible speculation” (Schetky, 1991, p. 403).

Within the professional community of psychologists and psychiatrists, an area of debate has been the difference between performing a forensic as opposed to a clinical evaluation. Numerous authors have emphasised the importance of recognising this distinction (Borum & Grisso, 1996; Campbell, 1997; Heilbrun, Rosenfeld, Warren, & Collings, 1994; Skeem, Golding, Cohn, & Berge, 1998; Wiklund, 1999). In fact, some have even contended that mental health professionals should be prevented from testifying about their patients (Shuman, Greenberg, Heilbrun, & Foote, 1998). Part of this discussion emanates from the observation that many reports presented before the courts have failed to address the legally relevant issues (Borum & Grisso, 1996). Another reason why this distinction is emphasised is the failure of some experts to recognise the different responsibilities of a legal as opposed to a clinical decision-making process (Campbell, 1997; Horner, Guyer & Kalter, 1993). Some authors are of the opinion that it is inappropriate to combine a therapeutic relationship with a forensic evaluation task, due to the risk of identifying too strongly with the patient (Shuman et. al, 1998; Wiklund, 1999).

The Swedish debate

In Sweden, the controversy concerning a clinical as opposed to a ‘strict forensic’ perspective in proposed sexual abuse cases has been closely connected to the issue of whether statement analysis or child psychiatric/psychological evaluations constitutes the most appropriate evaluation procedures. Whereas the importance of experience from working with children has been emphasised by some authors (Nyström, 1992), others have criticised ‘child professionals’ for underestimating the risks that are involved in interviewing children (e.g. Edvardsson, 1996). Such risks could include, for example, the suggestibility of the child and the possible impact of the relationship between the child and the evaluator.

Statement analysis refers to the practice of analysing the content of witness statements as well as the context in which the statement was given. The hypothesis underlying statement analysis is that truthful statements differ significantly from unfounded or falsified stories. The statement is evaluated by applying a set of principles commonly referred to as ‘reality criteria’, assessing, for example, the amount of detail in the statement. The original version of this methodological framework was called Statement Reality Analysis (Undeutsch, 1982). According to Undeutsch, reality-criteria are “designed to constitute a guarantee that the statement is factual” (p. 43). Statement analysis has a long tradition in Sweden[1], due to the close collaboration between German and Swedish forensic psychologists (Holgerson, 1990; Trankell, 1971; Undeutsch, 1982).

Two attempts to further develop Undeutsch’s work have been made. In Sweden, professor Arne Trankell (1971) outlined the so-called Formal Structure Analysis (FSA). In the United States, Raskin and Esplin (1991) described the Statement Validity Analysis (SVA). Both of these applications of statement analysis can be described as combinations of different partial assessment approaches, of which the analysis of the statement constitutes one element. In addition, other information is incorporated into the assessment model. This may include, for example, data regarding the context in which the statement evolved, the quality of the interview, or data related to the individual who is assessed. However, the two assessment models differ in the sense that the FSA is based on a qualitative paradigm (hermeneutics1), and the SVA is constructed to be available for empirical, quantitative validation.

Several efforts have been made to test the reliability and validity of parts of the SVA (for a review, see Lamb, Sternberg, Esplin, Hershkowitz, & Orbach, 1997). The validity of the FSA is said to be guaranteed by the thorough documentation of each step of the analytic procedure, which makes it possible for the reader to judge whether the reasoning makes sense (Holgerson, 1990). In addition, the use of two “formal criteria” (see Paper V) are believed to serve as a guarantee that the evaluation will be performed “in a scientific way” (Holgerson, 1990, p. 133).