Editorial Perspective: Questioning kids: applying the lessons from developmentally sensitive investigative interviewing to the research context

Is a cup sadder than an orange? Your answer to that question – and whether you attempt to answer at all – might depend on whether you are a child. Question–answer exchanges are fundamental in interpersonal communication, and how children are questioned has crucial implications in clinical, medical, legal, and educational contexts. Of utmost importance to psychological scientists, children are routinely questioned in research contexts to study issues ranging from the effects of exposure to violence, to coping with cancer treatment, to the effectiveness of interventions for anxiety and depression, to perceptions of gender differences in intellectual ability. As such, children are questioned for research purposes across various areas of psychology including clinical, developmental, health, cognitive, and social as well as the related fields of psychiatry, social work, nursing, and medicine. Until now, the lessons learned and consensus built from decades of research on children's eyewitness testimony have been applied to improving legal, clinical, and forensic practice. However, how we talk to children as researchers also affects whether we produce rigorous science, and in light of this research, we believe we should pause for critical self-reflection. Poor questioning or ineffective contextualizing of our questions may result in, at best, incomplete, and at worst, misleading, information. We propose that scientists studying children should use a large but siloed body of research on developmentally sensitive investigative interviewing to develop methodology that best assesses children's experiences, knowledge, symptoms, behaviors, etc. We assert that scientists studying children can be more intentional about: (a) how we phrase questions, (b) how we contextualize questioning, and (c) how we describe methodology. In this piece, we use the term ‘children’ to refer to youth who are approximately preschool age (i.e., 3-year-olds) through middle childhood (i.e., approaching adolescence) as this range encompasses where the bulk of the research concerning developmentally sensitive investigative interviewing has focused. How we phrase questions – from the overall format or syntax down to the specific words – matters profoundly. Although researchers likely appreciate the importance of avoiding leading questions, even with adults, there is a great deal of variability in how leading or suggestive questions are defined, and subtle nuances in questioning can exert large effects on what children say. In investigative interviewing, where children are often asked about adult wrongdoing, it is common to distinguish between closed-ended questions, or those that can be answered with a single word or detail, and open-ended questions, or those that request elaboration (Stolzenberg, McWilliams, & Lyon, 2017a, 2017b). Closed-ended questions can lead to ambiguous, incomplete, and/or inaccurate responses from children, particularly when compared to open-ended questions (e.g., Tell me about X; Stolzenberg et al., 2017a, 2017b; see Lamb, Brown, Hershkowitz, Orbach, & Esplin, 2018, for a review). Especially problematic are questions that can be answered with a ‘yes,’ ‘no,’ or other proffered option as provided by interviewers in a ‘forced choice’ or ‘option-posing’ question. The classic example from the investigative interviewing research involves asking children whether sexual touch was over or under their clothes, a question that neglects other possibilities (e.g., both, clothes were not on, etc.) and has significant implications for the charges that may result from a child's statement. In nonforensic contexts, seemingly innocuous conversational norms such as adding ‘Do you know’ at the beginning of a question stem (e.g., Do you know if it was light outside?) to communicate politely can significantly decrease children's accuracy (Evans, Stolzenberg, Lee, & Lyon, 2014). Even the inclusion of a single word, such as using ‘tell’ instead of ‘ask,’ can shift a child's response from accurate to inaccurate, because it presupposes intent not provided by the child, and ineffectively capitalizes on children's developing use of language. For example, an interviewer may wish to inquire about possible coaching attempts with a child by asking, ‘Did your mom tell you what happened?’ Children may assent to ‘tell’ as an appropriate replacement for ‘ask’ (‘Did your mom ask you what happened?’), thus profoundly changing the meaning of the exchange (Stolzenberg et al., 2017b). A significant breakthrough in investigative interviewing was the discovery of how much new information can be gleaned via cued invitations (You said X. Tell me more about X.), a method that avoids introducing interviewer input and relies on children's own words to scaffold their responses (Lamb et al., 2018). When the research questions allow, scientists should use open and cued invitations to question children. One reason that option-posing questions are exceptionally problematic is highlighted by another robust finding in the investigative interviewing research: Children rarely request clarification or admit ignorance in response to adults’ questions (Malloy, Katz, Lamb, & Mugno, 2015). In fact, they will even answer highly complex, bizarre, or nonsensical questions like the one we posed at the beginning of this article, despite rightly identifying them as ‘silly’ questions (Waterman, Blades, & Spencer, 2000). However, children are far more likely to say ‘I don't know’ to such nonsense questions when they are more open- (e.g., Where do circles live?) than closed-ended, presumably because it is more difficult to generate a relevant response than it is to simply parrot back one of a limited number of proffered options, none of which may accurately capture children's thoughts, feelings, behaviors, or experiences (Waterman, Blades, & Spencer, 2004). When specific categories of information are needed, scientists should consider using nonleading wh- questions (what, who, where, when, why, or how) over option-posing questions. Children's failure to exercise their ‘don't know’ or ‘don't understand’ response options stems in part from the power imbalance inherent in adult–child questioning. Thus, it is crucial for researchers to contextualize the questioning by establishing ‘ground rules’ (see Brubacher, Poole, & Dickinson, 2015, for a review). Children are unaccustomed to serving as informants and may be used to delivering perfunctory responses to routine questions on commonly discussed topics posed by adults who already know the answers. While adults understand that a fundamental aspect of communication is signaling misunderstandings, the investigative interviewing literature has made clear that children do not feel thus obligated, do not know how, or do not feel comfortable doing so. In short, just because children respond to your question–and provide a seemingly reasonable response–does not mean that they understand it, a sobering realization from a data collection perspective. Children should be adequately prepared for any questioning which includes being informed about its purpose, being told that it is okay to say ‘don't know’ and ‘don't understand,’ given explicit permission to correct or disagree with the researcher, and building rapport, especially if questioning involves sensitive topics (Brubacher et al., 2015). Such practices may also help facilitate a more fully informed assent process when youth and other vulnerable populations (e.g., the elderly, individuals with intellectual disabilities) are invited to participate in research, while simultaneously maximizing their reporting of accurate and complete information. Scientists must be more thorough in describing research methodologies, informing others about what was said prior to and during questioning. Relatedly, our data analysis and reporting strategies should reflect the fact that questioning is an inherently reciprocal and dynamic exchange rather than a one-sided opportunity to ‘extract’ data from children. Undoubtedly, how researchers phrase and contextualize questions must depend on individual research questions and goals. There may, in fact, be circumstances in which it is ideal to pose option-posing questions or skip methods that build rapport or establish ground rules for children's responses. However, it is best if these crucial methodological decisions are made intentionally with regard to the age of children involved and with awareness of the potential impact on study results. Furthermore, such decisions should be discussed in the peer-reviewed publishing process, as they can affect the validity and replicability of collected data. Given the extent to which scientific data are collected from children via question–answer exchanges with adults, and the potential for the resultant research to affect the lives of children and families and public health more generally, the consequences of inadequately or inappropriately questioning children may be dire. So, are we suggesting that scientists avoid questioning children altogether? No. Generalizing from studies on adults to children may be ineffective or downright dangerous, especially in clinical or medical contexts. Furthermore, there are some areas of knowledge that may not be meaningfully acquired without relying on children's direct reports. Consistent with the United Nations Convention on the Rights of the Child, we believe in giving children a voice, especially when sharing their views, opinions, and experiences with researchers can lead to scientific breakthroughs; appropriately tailored prevention and intervention strategies; and empirically informed public policies. Furthermore, we do not wish to diminish the potential importance of these issues among adolescent or adult research participants who may be similarly, although perhaps more moderately, affected by how inquiries are phrased and the more general research context. However, we assert that these concerns are of exceptional importance in research involving children where there are developmental considerations (e.g., cognitive, socioemotional, and linguistic) that may constrain children's effective participation in research and affect best practices for collecting data. The success of science comes from our ability to establish robust and reliable findings. Thus, to combat the concerns raised here, we suggest that researchers studying children facilitate complete and accurate responding from them by considering the findings from investigative interviewing research when designing, conducting, and reporting studies, as well as when training future researchers or evaluating the merits of publishable data. L.C.M. and S.N.S. contributed equally to drafting and editing this article. The authors have declared that they have no competing or potential conflicts of interest.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.