Assessing the value of structured protocols for forensic interviews of alleged child abuse victims

OBJECTIVE: To evaluate the effectiveness of a structured interview protocol (NICHD Investigative Interview Protocol) operationalizing universally recommended guidelines for forensic interviews. METHOD: The NICHD Investigative Protocol was designed to maximize the amount of information obtained using recall memory probes, which are likely to elicit more accurate information than recognition memory probes. Forensic investigators were trained to use the NICHD protocol while conducting feedback-monitored simulation interviews. The utility of the protocol was then evaluated by comparing 55 protocol interviews with 50 prior interviews by the same investigators, matched with respect to characteristics likely to affect the richness of the children's accounts. The comparison was based on an analysis of the investigators' utterance types, distribution, and timing, as well as quantitative and qualitative characteristics of the information produced. RESULTS: As predicted, protocol interviews contained more open-ended prompts overall as well as before the first option-posing utterance than non-protocol interviews did. More details were obtained using open-ended invitations and fewer were obtained using focused questions in protocol interviews than in non-protocol interviews, although the total number of details elicited did not differ significantly. In both conditions, older children provided more details than younger children did. CONCLUSION: The findings confirmed that implementation of professionally recommended practices affected the behavior of interviewers in both the pre-substantive and substantive phases of their interviews and enhanced the quality (i.e., likely accuracy) of information elicited from alleged victims.     

Assessing the effectiveness of the NICHD investigative interview Protocol when interviewing French-speaking alleged victims of child sexual abuse in Quebec

ObjectivesThe study was designed to assess the effectiveness of the flexibly structured NICHD Investigative Interview Protocol for child sexual abuse (CSA) investigative interviews by police officers and mental health workers in Quebec. The NICHD Protocol was designed to operationalize “best practice” guidelines and to help forensic interviewers use open-ended prompts to facilitate free recall by alleged victims.MethodA total of 83 interviews with 3- to 13-year-old alleged victims were matched with 83 interviews conducted by the same interviewers before they were trained to use the Protocol. Interviews were matched with respect to the children's ages, children–perpetrator relationships, and the types and frequency of abuse. Coders categorized each of the prompts used to elicit information about the abuse and tabulated the numbers of new forensically relevant details provided in each response.ResultsInterviewers used three times as many open-ended prompts in Protocol interviews than in non-Protocol interviews, whereas use of all other types of questions was halved, and the total number of questions asked decreased by 25%. Protocol-guided interviews yielded more details than comparison interviews. The mean number of details per prompt increased from 3 to 5 details when the Protocol was used. Even with young children, interviewers using the Protocol employed more invitations to elicit forensically relevant details.ConclusionsFrench-speaking investigators using the NICHD Protocol used open-ended prompts rather than focused questions when interviewing alleged victims. In addition, these interviewers needed fewer questions to get relevant information when using the Protocol.Practical implicationsA French version of the NICHD Protocol is now available to police officers and social workers who investigate the alleged sexual abuse of young children in French-speaking countries. This French version allowed trained interviewers to increase the use of invitations and reduce the use of more focused and risky questions. When the number of questions was controlled, more central details and more details in total were obtained in Protocol interviews, because the average prompt elicited more detailed answers in Protocol interviews. However, learning to use the NICHD Protocol required extended training and continued feedback sessions to maintain the high quality of interviewing.     

The relationship between within-interview contradictions and eliciting interviewer utterances.

OBJECTIVE: To determine whether interview practices associated with inaccurate reporting in laboratory analog contexts were also associated with inaccurate information in actual forensic contexts. METHOD: The forensic interview of a 5-year-old girl, an alleged victim of sexual abuse, was analyzed to explore interview practices associated with the retrieval of contradictory information. Content analyses of the child's responses focused on: (1) new informative details about the reported incidents; (2) contradictory details; (3) "central" and "peripheral" details; and (4) the types of utterances used to elicit each detail. RESULTS: The results illustrate how risky option-posing and suggestive utterances can be, as most (90%) contradicting details were elicited using option-posing and suggestive utterances and almost all (98%) of the contradicted and contradicting details were central, containing crucial information concerning the investigated allegation. No contradictory details were elicited in response to open-ended invitations. CONCLUSION: The findings demonstrate that poor interviewing practices can be associated with high levels of internal contradiction and should be avoided by forensic interviewers. To avoid contaminating children's reports and increase the likely accuracy of the information retrieved, moreover, interviewers should elicit as much information as possible using open-ended utterances, which tap free-recall memory.     

Investigative interviews of child witnesses in Sweden

OBJECTIVE: To evaluate the structure and informativeness of interviews with 4- to 13-year-old alleged victims of sexual abuse in Sweden. METHOD: Seventy-two alleged victims of sexual abuse were interviewed by six experienced officers from one police district in Sweden. Our evaluation focused on the structure of the interviews, the distribution and timing of the investigators' utterance types, and the quantity and quality of the information provided by the children. RESULTS: Content analysis revealed that the interviewers relied primarily on option-posing and suggestive questions--together, these comprised 53% of their utterances--when interviewing the alleged victims. As a result, most of the details (57%) obtained from the children were elicited by option-posing and suggestive utterances. Only 6% of the interviewers' utterances were open-ended invitations, and these elicited only 8% of the information obtained. CONCLUSION: The reliance on option-posing and suggestive prompts may have reduced the accuracy of the information obtained, thereby interfering with the investigations, and reducing the forensic admissibility of the children's statements. This suggests a continuing need in Sweden, as in other countries, for interview practices that enhance the quality of information provided by young victims.     

Investigative interviews of alleged sexual abuse victims with and without anatomical dolls

Verbal and nonverbal responses by alleged victims of child sexual abuse were coded for length, amount of information, and the manner in which they were elicited by the interviewer. In 16 of the interviews, anatomical dolls were employed for the purposes of demonstration, whereas they were not used in another eight cases matched with respect to other characteristics of the children and the alleged events. Children interviewed with dolls provided an equivalent number of details and spoke as many words in the substantive portion of the interview as did children interviewed without dolls, and interviewers in the two groups used similar probes to elicit information. However, the average responses by the children were significantly longer and more detailed when dolls were not used. Children gave longer and more detailed responses to open-ended invitations when dolls were not used. Caution is necessary when interpreting these findings.     

Interviewing youthful suspects in alleged sex crimes: a descriptive analysis

OBJECTIVE: To introduce and evaluate a structured interview protocol designed for investigative interviews of youthful alleged perpetrators of child sexual abuse. METHOD: Seventy-two alleged perpetrators ranging from 9 to 14 years of age (M = 12 years) were interviewed by 1 of 13 experienced youth investigators, employed by the Israeli Ministry of Labour and Social Affairs, about incidents that had been reported by alleged victims. All interviews were conducted as part of the investigators' regular work and followed the structured interview guide appended to this article. RESULTS: Interviewers questioned older and younger children similarly, but addressed fewer invitations, directive questions, and option-posing prompts to suspects who denied the allegations than to those who partially or fully admitted them. The total number of details provided by the suspects did not vary depending on their age or whether or not they fully or partially admitted the allegations. In both cases, more information was elicited using invitations rather than suggestive or option-posing prompts. CONCLUSION: Contrary to expectations, suspects who at least partially admitted their involvement provided considerable amounts of information and were very responsive to free recall prompts, although interviewers used more risky (potentially error-inducing) prompts when interviewing suspects rather than alleged victims.     

Differences between accounts provided by witnesses and alleged victims of child sexual abuse

OBJECTIVES: To determine whether child witnesses of sexual abuse were more or less informative about the alleged incidents than alleged victims when interviewed similarly. METHOD: Twenty-six alleged victims of child sexual abuse (aged 5 to 14 years; M=9.8 years) and 26 children who had witnessed but not experienced similar events were interviewed by experienced youth investigators about the alleged abuse. Children in the two groups were matched with respect to their age, relationships with the alleged perpetrator, and seriousness of the alleged offenses. All children were interviewed using the NICHD investigative interview protocol. RESULTS: Witnesses and victims provided similar amounts of information about the incidents of abuse. Interviewers used more open-ended invitations and elicited more information using open-ended prompts from witnesses than from victims, whereas they used more risky (including suggestive) prompts when interviewing victims. DISCUSSION: These results confirm that young children can be informative witnesses about events that they have either experienced or witnessed.     

Effects of age and delay on the amount of information provided by alleged sex abuse victims in investigative interviews

A total of 145 children of 4 to 5, 6 to 7, 8 to 9, and 10 to 12 years of age were interviewed within 3 days, 1 month, 1 to 3 months, or 5 to 14 months after allegedly experiencing a single incident of sexual abuse. The proportion of substantive investigative utterances eliciting new details from the children increased with age and decreased after delays of more than 1 month. Age (but not delay) was also associated with the length and richness of informative responses to individual investigative utterances of all types. Children were more likely to provide new details in response to option-posing and suggestive prompts. As in previous field studies, interviewers employed few open-ended prompts, and thus only 5% of the information obtained was elicited using free-recall prompts.     

Dynamics of forensic interviews with suspected abuse victims who do not disclose abuse

OBJECTIVES: The present study was designed to explore structural differences between forensic interviews in which children made allegations and those in which children did not make allegations. METHODOLOGY: Fifty forensic interviews of 4- to 13-year-old suspected victims of abuse who did not disclose abuse during the interview were compared with the same number of forensic interviews of alleged victims who made allegations of sexual or physical abuse. Only cases in which there was substantial reason to believe that abuse had taken place were included in the study. Audiotapes of the interviews were examined with a focus on interviewer utterances and children's responses during the pre-substantive rapport-building, episodic memory training, and 'getting the allegation' phases of the interviews, which all employed the NICHD Investigative Interview Guide. FINDINGS: Forensic interviews which yielded allegations of child abuse were characterized by quite different dynamics than interviews with children who did not make allegations. When interviewing non-disclosers, interviewers made less frequent use of free recall prompts and offered fewer supportive comments than when interviewing children who made allegations of abuse. Children who did not disclose abuse were somewhat uncooperative, offered fewer details, and gave more uninformative responses, even at the very beginning of the interview, before the interviewers focused on substantive issues and before the interviewers themselves began to behave differently. CONCLUSIONS: A premature focus on substantive issues may prevent children who are not responsive in the episodic memory training phase from disclosing abuse. Identifying reluctant disclosers and making more extensive efforts to build rapport before substantive issues are broached, or interviewing such children in more than one session, may help suspected victims disclose their experiences.     

Witness recall across repeated interviews in a case of repeated abuse

In this illustrative case study we examine the three forensic interviews of a girl who experienced repeated sexual abuse from ages 7 to 11. She disclosed the abuse after watching a serialized television show that contained a storyline similar to her own experience. This triggered an investigation that ended in successful prosecution of the offender. Because this case involved abuse that was repeated on a weekly basis for 4 years we thus investigated the degree to which the child's narrative reflected specific episodes or generic accounts, and both the interviewer's and child's attempts to elicit and provide, respectively, specific details across the 3 interviews collected in a 1 month period. Across the 3 interviews, the child's account was largely generic, yet on a number of occasions she provided details specific to individual incidents (episodic leads) that could have been probed further. As predicted: earlier interviews were characterized more by episodic than generic prompts and the reverse was true for the third interview; the child often responded using the same style of language (episodic or generic) as the interviewer; and open questions yielded narrative information. We discuss the importance of adopting children's words to specify occurrences, and the potential benefits of permitting generic recall in investigative interviews on children's ability to provide episodic leads. Despite the fact that the testimony was characterized by generic information about what usually happened, rather than specific episodic details about individual occurrences, this case resulted in successful prosecution.     

Impact of interviewers’ supportive comments and children's reluctance to cooperate during sexual abuse disclosure

In the field of child sexual abuse (CSA) disclosure, many studies have been conducted on the impact of interviewers’ questioning style, but few have examined the impact of interviewers’ supportive comments on children's cooperative and reluctant disclosure of substantive details. This field study used a sample of children ranging from 4 to 13 years of age who have all disclosed CSA. The first objective was to examine if the interviewer's and the child's comments during CSA interviews would vary as a function of the use of the National Institute of Child Health and Human Development (NICHD) Protocol. The second objective was to identify the strongest correlates of the proportion of details disclosed by the children during forensic interviews. A total of 90 matched NICHD Protocol and non-Protocol interviews done by the same interviewers were audio-taped, transcribed, and coded using verbal subscales. The goal was to explore if differences exist between the interviewers’ supportive and non-supportive comments as well as children's cooperative and reluctant statements during investigative interviews conducted prior to or after the NICHD Protocol training. Results of a MANCOVA showed that the use of the NICHD Protocol had no influence on interviewers’ and children's demeanors. A hierarchical multiple regression analysis controlling for relevant variables (e.g., child's age and NICHD Protocol) showed that children's reluctance and interviewers’ non-support were associated with a lesser proportion of details. Overall, these results indicate that in order to promote detailed disclosure of CSA, interviewers should decrease their non-supportive comments and learn to deal more effectively with children's reluctance during forensic interviews. As such, protocols and training should encourage investigative interviewers to devote more time identifying early signs of children's verbal reluctance and to understand the negative impact of non-supportive comments on the disclosure of substantive details.     

The effect of multipart prompts on children's testimonies in sexual abuse investigations

ObjectiveThe current study aimed to explore the frequency and effects of multipart prompts on the testimonies of children who were alleged victims of sexual abuse and were interviewed using the National Institute of Child Health and Human Development (NICHD) Investigative Protocol. The effects of the multipart prompts were studied by considering the type of prompt given to the children and examining the richness of the children's testimonies (e.g., the number of words and the number of forensic details) and the ways the children contended with these prompts (e.g., which demand they answered, whether they signaled misunderstanding).Method71 Israeli children aged 4–9 years were interviewed after a complaint of single incident of sexual abuse by a perpetrator who was not a family member. All of the interviews that met the specified criteria and were conducted within a specified period were included in this study. Two raters identified simple versus multipart prompts and analyzed the children's responses.ResultsThe results clearly showed that multipart prompts were used in most interviews, regardless of the child's age. An average of 5.58 multipart prompts per interview was given. The effects of the multipart prompts were destructive and harmed the length and the richness of the children's testimonies. Children of all ages failed to signal their lack of understanding of multipart prompts, and 24% of their responses were unintelligible. When the children did produce a relevant and substantive answer, they primarily responded to the last demand in the multipart prompt and rarely provided an answer to both demands.ConclusionsThe study clearly indicates that even well-trained investigative interviewers present inappropriate multipart prompts to children. The findings contribute to the existing knowledge about the adverse effect that multipart prompts have on children's narratives, indicating that children of all ages provided poorer testimonies in response to multipart prompts. The systematic knowledge accumulated in both laboratory and field studies indicates that it is necessary to eliminate the use of multipart prompts by updating existing practical guidelines and training courses.     

The development of communicative and narrative skills among preschoolers: lessons from forensic interviews about child abuse

This study examined age differences in 299 preschoolers' responses to investigative interviewers' questions exploring the suspected occurrence of child abuse. Analyses focused on the children's tendencies to respond (a) at all, (b) appropriately to the issue raised by the investigator, and (c) informatively, providing previously undisclosed information. Linear developmental trends characterized all types of responding. When the types of prompts were considered, 3- to 4-year-olds responded slightly more informatively to specific (directive) recall prompts than to open-ended prompts whereas children aged 5 and older were more responsive to open-ended recall prompts. The findings suggest that even 3-year-olds can provide information about experienced events when recall processes are activated, although the ability to provide narrative responses to open-ended recall prompts only becomes reliable later in development.     

Young children's references to temporal attributes of allegedly experienced events in the course of forensic interviews

Developmental differences in references to temporal attributes of allegedly experienced events were examined in 250 forensic interviews of 4- to 10-year-old alleged victims of sexual abuse. Children's ages, the specific temporal attributes referenced, and the types of memory tapped by the interviewers' questions significantly affected the quantity and quality of temporal references produced. The findings documented age-related increases in 4- to 10-year-olds' references to temporal attributes, using the appropriate relational terminology, both spontaneously and in response to temporal requests. More references to temporal attributes were elicited from recall than from recognition memory, highlighting spontaneous reporting capabilities. Implications for theories concerning the developing understanding of temporal concepts and for the design of effective, age-appropriate, forensic interview techniques are discussed.     

Improving the Measurement of Cognitive Skills Through Automated Conversations

Open-ended, short-answer questions, referred to as constructed responses (CR), allow students to express knowledge and skills through their own words. While CRs can reduce the likelihood of guessing correct answers, they also enable students to provide errant responses due to a lack of knowledge or a misunderstanding of the question. Conversation-based assessments (CBAs) require constructing responses to open-ended prompts (similar to CRs), but also leverage natural-language processing to provide adaptive follow-up prompts that target particular information. This work describes an investigation into potential benefits of CBAs as they are compared to a CR approach. Results from 632 middle schoolers indicate that, when compared to CRs, the CBA items allowed 41% of students to provide a more complete response and improve their score.     

Struggles and Successes in Constructivist Jewish Early Childhood Classrooms

This study examined the experiences of teachers in a Jewish early childhood center implementing constructivist theory and pedagogy through a Reggio Emilia-inspired model. Constructivist practices were described through interviews, surveys, classroom documentation, and observations. The data indicated that areas that were initially deemed as successes were hard to maintain and seemingly overshadowed by challenges. These areas included understanding constructivist theory, using open-ended materials, and relying on open-ended questions to facilitate knowledge. Furthermore, teachers struggled when comparing information gained by children from classes using a more traditional educational approach. Insights in each area are provided to support Jewish and general studies constructivist teachers.     

Subjective Priors for Item Response Models: Application of Elicitation by Design

Bayesian methods incorporate model parameter information prior to data collection. Eliciting information from content experts is an option, but has seen little implementation in Bayesian item response theory (IRT) modeling. This study aims to use ethical reasoning content experts to elicit prior information and incorporate this information into Markov Chain Monte Carlo (MCMC) estimation. A six‐step elicitation approach is followed, with relevant details at each stage for two IRT items parameters: difficulty and guessing. Results indicate that using content experts is the preferred approach, rather than noninformative priors, for both parameter types. The use of a noninformative prior for small samples provided dramatically different results when compared to results from content expert–elicited priors. The WAMBS (When to worry and how to Avoid the Misuse of Bayesian Statistics) checklist is used to aid in comparisons.