Recent Research: Child Victim Interviews
Drinnan, Lora, Canadian Journal of Police and Security Services
Lamb, M.E., & Garretson, M.E. (2003). The effects of interviewer and child gender on the informativeness of alleged child sexual victims in forensic interviews. Law and Human Behavior, 27, 157-171.
It has been suggested that Hugo Munsterberg was one of the first social scientists to promote the empirical exploration of eyewitness memory accuracy (Wrightsman, 2001). Munsterberg (1908) argued that it was important for the courts and other legal entities to know whether a witness's recollections are objective reconstructions of past events, or reconstructions studded with associations and suggestions. Since the early work of Munsterberg, significant research has been conducted in eyewitness memory, creating a rich tapestry of areas of concentration within the field itself. One area that has received a significant amount of attention involves the processes surrounding the interviewing of witnesses, especially in regard to the differences between children and adult witnesses. Included among the findings to date in this area, is the fact that children and adults differ in terms of their level of vulnerability during the interview process, with children tending to be more susceptible to misleading questions than adults (Roebers and Schneider, 2000).
In their article entitled, "The Effects of Interviewer and Child Gender on the Informativeness of Alleged Child Sexual Victims in Forensic Interviews", Lamb and Garretson suggest that age and communicative abilities are important considerations when questioning a child victim in cases involving alleged sexual abuse. Specifically, the authors argue that the gender of the interviewer and the child can significantly impact the effectiveness of the interaction with regard to the accuracy of information. As evidence for this claim, the authors cite previous research indicating that male interviewers tend to be more verbally aggressive and argumentative (i.e., they tend to ignore, challenge, dispute, and interrupt) than female interviewers. Lamb and Garretson argue that this more aggressive approach may cause children to become more susceptible to suggestion and result in a less accurate account of events when compared to female interviewers who tend to be perceived as less authoritative and more supportive.
Based on the above research, Lamb and Garretson argue that supportiveness appears to be a key factor in a child victim's perception of authority and that this can significantly decrease the level of suggestiveness and subsequently improve the accuracy of responses to misleading questions. In short, the authors suggest that forensic interviewers who are perceived as more supportive tend to elicit more information from a child. In addition, this information tends to be more accurate because the supportive interaction encourages the child to respond with their perception of the event rather than to suggestive remarks that may be made by the interviewer.
In addition to the above, Lamb and Garretson also suggest that the gender of the interviewer and interviewee is especially important in situations where best practice guidelines and protocols are not employed. For the most part, best practice guidelines maintain that the interviewer should build a rapport with the child in order to facilitate disclosure and informativeness. In addition, guidelines suggest that interviewers should utilize open-ended utterances that allow the child to provide details without interference or suggestion from the interviewer. As an example, the authors refer to the NICHD (National Institute of Child and Human Development) protocol (Orbach, Hershkowits, Esplin, & Horowitz, 2000) which encourages building a rapport with the child at the initiation of the interview in order to create a more relaxed atmosphere. The protocol also encourages the interviewer to explain the 'ground rules' to ensure that the child understands the importance of telling the truth, including asking for clarification if a question is confusing and saying "I don't know" when appropriate. In order to demonstrate the kind of information that is expected of the child, the NICHD protocol also suggests that the interviewer first guide the child through a recent unrelated event. The structured interview protocol also promotes the use of open-ended questions such as "What happened next?", and disparages the use of option-posing and suggestive questions such as "Did he hurt you?" and "Where did he touch you?"
Given the findings of previous research, Lamb and Garretson hypothesized that children would respond better to female interviewers by providing more informative responses as measured by the number of 'forensically relevant details' in their statements. In addition, the authors predicted that, because children are more comfortable with female and same-gender interviewers, interviews conducted by females on girls would result in the highest number informative responses. The authors also hypothesized that the behaviour of the interviewer, as measured by the number and type of utterances made during the interview, would significantly impact the number of forensically relevant details obtained from the child. Specifically, it was predicted that male interviewers would elicit more details through the use of option-posing and suggestive statements, as opposed to directive and invitation statements, because children tend to be more vulnerable to suggestive statements posed by authoritative figures. The final supposition made by Lamb and Garretson was that the gender-of-interviewer effect would be diminished or eliminated when interviews were conducted following structured protocol guidelines.
In order to investigate the above hypotheses, Lamb and Garretson (2003) conducted an archival analysis of 672 forensic interviews obtained from Britain, Israel, and the United States. Of the 672 interviews, 474 were conducted by females and 198 were conducted by males. In addition, 350 of the interviews were conducted according to the NICHD structured interview protocol while 322 of the interviews were conducted without an identifiable protocol. The children interviewed were categorized into three age groups (i.e., 4 - 6, 7 - 9, 10 - 14) and all interviews were reviewed in order to identify four types of utterances made by the interviewer: invitations, directive questions, option-posing questions, and suggestive statements. Invitations were classified as open-ended questions that allowed the child to provide details about the event in question. Directive questions included requests for additional information about topics previously mentioned by the child. Option-posing questions were defined as requests for conformation or denial to a statement made by the interviewer. Finally, suggestive statements were categorized as utterances that imply something not mentioned by the child.
As hypothesized, the results of the Lamb and Garretson study indicate that the gender of both the interviewer and the child significantly impact the type and amount of information obtained during forensic interviews only when a structured interview protocol was not employed. Specifically, female interviewers who did not follow the NICHD protocol appeared to cultivate greater compliance from both boys and girls, as measured by the number of new details evoked by their prompts, than male interviewers who did not employ the NICHD protocol. Interestingly, however, female interviewers who did not use a protocol also tended to treat boys and girls differently, posing more invitations and suggestions to boys than girls. In contrast, analysis of the data suggested that male interviewers who did not employ a structured interview protocol did not treat boys and girls differently in any identifiable way. In addition, results indicated that when the NICHD structured protocol was not used, girls dispensed more details in response to directive questions posed by female interviewers than male interviewers. Importantly, however, older girls (aged 10 - 14) provided more details in response to option-posing questions posed by male interviewers. Finally, when the NICHD protocol was not used, younger children (ages 4 - 6) dispensed more details in response to suggestive utterances posed by interviewers of the opposite gender.
In contrast to the above, when the NICHD structured interview protocol was utilized, the gender-of-interviewer effects were either significantly diminished or eliminated completely. For example, older children divulged more details in response to directive questions posed by female interviewers in non- protocol interviews, but did not deviate in their responses to the various utterances made by male and female interviewers in protocol interviews. Similarly, although younger children provided more details in response to option-posing questions made by female interviewers in non-protocol interviews, when protocol interviews were employed there was no significant difference in response levels between female and male interviewers.
Overall, Lamb and Garretson's results appear to be consistent with their hypotheses. Specifically, the authors report that their analyses indicate that the gender of the interviewer and child does affect the informativeness of the interview when a structured protocol is not used. That is, children provide more forensically-relevant details about the alleged event to female interviewers than male interviewers. This difference, however, is diminished or eliminated when interviewers follow the NICHD structured interview protocol. In addition, it appears that the gender-of-interviewer effects are not as dramatic as previous research suggests. Simply put, Lamb and Garretson expected to find more extreme differences in both the method used and the information obtained by male and female interviewers. Although differences were found, they were much less pronounced than expected.
The results of the study also appear to support the hypothesis that female interviewers tend to evoke more detail from children than male interviewers, however, further analysis of the data provided some interesting additional findings. Specifically, although female interviewers were better able to draw out details from children than male interviewers, they also tend to pose more invitations and suggestions to boys than to girls and, as such, the quality and accuracy of information received from boys was reduced due to its suggestive nature. In other words, although the female interviewers succeeded in obtaining more information from children than male interviewers, this was done through the use of invitation and suggestion which causes the information obtained to be less reliable.
Although the Lamb and Garretson study provides some thought-provoking results, it is not without limitations. One of the most significant concerns with this study is that it was conducted using archival analysis, a non-experimental form of research involving the collection and examination of previously recorded events and data. That is, the interviews analyzed by Lamb and Garretson were not conducted under their scrutiny and, as such, the authors of the study did not have the ability to control the interview processes (e.g., type of room used for the interview, time of day interview was conducted). As with all archival research, this situation leads to questions regarding potential confounding variables that cannot be controlled for through this research avenue. Future research on factors affecting the interview process may benefit from observing the interview process while it is occurring, alleviating the impact of confounding variables.
A second limitation to this study revolves around the fact that, although the findings are based on a collection of "forensic interviews", there is no mention of who conducted the interviews (i.e., psychologists, social workers, police officers) or where the interview was conducted. In order to conclusively generalize that interviews be conducted according to structured protocol guidelines, interviews from all forensic settings need be analyzed. Due to the fact that a police station offers a significantly different setting than a psychologist's office, it is possible that these settings would evoke differential responses from children regardless of the interviewer. These different reactions, in turn, could lead to differences in the amount and accuracy of the information obtained. It may be argued that further research is warranted on the impact of interviews conducted in different settings by different professionals in order to determine what type of interview process is best suited for each type of interview setting.
A third limitation of this study is the lack of random sampling that offers the ability to generalize results to a broader audience. For example, due to the fact that the majority of interviewers and interviewees in this study were female, it is difficult to generalize these results to interviewers of both genders. As such, although this study does indicate that the gender of both interviewer and interviewee impacts the interview process as well as the information secured from the interview, more research involving both male interviewers and interviewees needs to be conducted in order to generalize the findings to the public at large.
A final issue that bears mentioning with regard to the research on the impact of gender on interviewer effectiveness revolves around the distinction between sex and gender. That is, social scientists have argued that there is a clear distinction between an individual's biological sex (i.e., male or female) and their socially-defined gender. As such, in some situations it is possible to have a female who "acts" as a male would (according to the stereotypical characteristics proscribed by society) and a male who "acts" like a female would. Lamb and Garretson, like many others who have published on this topic, do not appear to differentiate between the concepts of sex and gender and, as such, do not appear to investigate whether the impact is due to the gender or sex of the interviewer. That is, in their paper, the authors suggest that male interviewers are stereotypically more aggressive than female interviewers and therefore tend to be perceived as more supportive by children. One might argue, however, that a male interviewer who employs a supportive style might be more effective than a female interviewer that employs an aggressive style. Future research should attempt to tease out this distinction.
In sum, this study indicates that using a structured interview protocol, such as the NICHD, minimizes or eliminates gender-of-interviewer effects thereby improving the quality of information obtained during the interview process. Although the findings of the present study provide a solid foundation on which to base interviews as well as investigation into the interviewing process, further investigation is clearly required. Until this future research is conducted, police service personnel should be cognizant of the fact that gender-of-interviewer effects appear to be decreased through the use of structured protocol interviews.
Munsterberg, H. (1908). On the witness stand. New York: Doubleday, Page & Co.
Orbach, Y., Hershkowitz, I., Esplin, P., & Horowitz, D. (2000). Assessing the Value of Structured Protocaols for Forensic Interviews of Alleged Child Abuse Victims. Child Abuse & Neglect, 24, 733-752.
Roebers, C. M., & Schneider, W. (2000). The impact of misleading questions on eyewitness memory in children and adults. Applied Cognitive Psychology, 14, 509-526.
Wrightsman, L.S. (2001). Forensic psychology. Toronto, Ontario: Nelson Thomson.
* Comments concerning this paper can be addressed to: Lora Drinnan, Department of Psychology, University of Regina, Regina, Saskatchewan, Canada, S4S 0A2.
Lora Drinnan *
University of Regina…
Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: Recent Research: Child Victim Interviews. Contributors: Drinnan, Lora - Author. Journal title: Canadian Journal of Police and Security Services. Volume: 1. Issue: 2 Publication date: June 2003. Page number: 144. © 2006 Meritus Solutions, Inc. COPYRIGHT 2003 Gale Group.
This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.