Exploring adolescents’ perceptions of a self-report measure on violence against children: a multi-country study in Romania, South Africa, and the Philippines

Abstract

This thesis explores how adolescents perceive, interpret, and experience answering self-report measures on violence against children. In doing so, this study tests the International Society for Prevention of Child Abuse and Neglect (ISPCAN) Child Abuse Screening Tool (Children’s Version), known as the ICAST-C, with adolescents aged 10-17 years in Romania, South Africa and the Philippines. This study contributes to a nascent body of research that recognises the complexity of measuring violence against children, particularly in relation to testing self-report measures with the target population. It contributes to the literature on psychometric properties of self-report measures and is broadly situated within a public health approach to the prevention of violence against children. This study aims to enhance the evidence base on self-report violence measures by testing the ICAST-C. It uses the construct of content validity and cognitive processing theories to understand the particular ways in which adolescents interpret items, recall experiences, and generate an accurate answer that reflects their experiences. Consequently, it contributes to the development of robust measures that are capable of monitoring violence against children. In-depth cognitive interviews and arts-based methods were used with adolescents in Iasi and Cluj-Napoca, Romania; Buffalo City, South Africa; and Metro Manila, Philippines, between 2018-19. Data were generated from transcripts of interviews, observation notes, completed forms, and drawings completed by participants. Ethical approval was received from the University of Edinburgh, Babes-Bolyai University, University of Cape Town, and Ateneo de Manila University. The findings of the study show that while the ICAST-C is generally perceived to be acceptable, relevant and comprehensible, cognitive interviewing revealed deeper issues with clarity of items and response options. Factors pertaining to adolescents’ past experiences, perceived benefits and burdens, efforts exerted, and features of the ICAST-C impacted how they perceived and comprehended the measure. Adolescents’ observations highlighted issues with interpreting the intent and wording of items, and contexts and behaviours that are considered, or not considered. In addition, adolescents’ processes of recall reveal potential challenges in generating an internal response that is also linked with the options presented on the ICAST-C. While viewed as broadly appropriate, certain items were perceived as inappropriate in specific contexts e.g. sexual abuse items in the Philippines. These evaluations are formed not only due to contextual differences but also item interpretations that diverged from the intended measurement objective. Adolescents shared insightful and valuable feedback on the ICAST-C and offered suggestions for its use. Adolescents’ emotions while answering the ICAST-C reflect complex processes. They experienced several emotions rapidly and simultaneously, and thus this process needs to be understood as dynamic rather than static. Emotions are produced due to several reasons, such as comprehension, past experiences, beliefs, and expectations about their performance. The study highlights the value of adolescents being able to discuss their experiences in a safe setting with trained interviewers afterward. It also highlights the wide-ranging meanings adolescents assign to the ICAST-C such as a protective tool, an instrument that enables disclosure, and a measure that creates awareness. Adolescents bring their everyday selves and experiences with them into the study, which highlights the value of asking about topics other than violence. This study discusses key meanings adolescents create in the research interaction. The findings of this study emphasise the value of using qualitative methods to understand the psychometric property of content validity, as well as testing self-report measures with the target population. This study presents several recommendations for using the ICAST-C in future studies as well as for conducting research on violence with children and adolescents, such as conducting pilot studies in contexts using cognitive interviewing methods to ascertain content validity, planning for and creating opportunities for disclosure, including a post-completion interview while conducting surveys on violence, and expanding the scope of the ICAST-C to include other topics and more positive questions. The findings also highlight adolescents’ experiences of answering questions on violence, and in so doing, acknowledge the value of accommodating all emotions and experiences that may come up during research.