Development of a New Measure of Adolescent Dating Aggression: National Norms with a Focus on Marginalized Youth

A serious problem with existing ADA research is that measures used to assess it are flawed. A new ADA measure is urgently needed. The ideal new measure will (a) reflect contemporary forms of abuse (e.g., cyber abuse) and relationship styles (e.g., multiple concurrent partners); (b) reflect
the input of youth historically underrepresented in ADA measurement development research (i.e., Black, Hispanic, Native American, and LGBTQ youth); (c) embed context of abusive acts in items, in
part to address the potential that gender-based differences might otherwise be masked (e.g., whether the act causes injury, fear, is experienced as fun horseplay); and (d) be appropriately tailored for different developmental stages of adolescence. Therefore, our goal is to modernize,
expand, contextualize, and otherwise improve the quality of ADA measurement items, so that valid inferences about youth experiences of ADA can be made more easily.,~
The proposed mixed methods study will generate a new measure of victimization and
perpetration (available in versions for youth 11-14, 15-18, and 19-21 years old), tested with a large nationally-representative sample (n=1,400), and short-form(s) of the measures that can be used for
rapid screening in healthcare or other settings--six instruments in total. The work will be organized in four phases. Phase 1: review of existing literature and previously collected qualitative data to generate a starting item pool; Phase 2: collect qualitative data from Black, Hispanic, LGBTQ and Native American youth and refine item pool via a Delphi process with advocates and research experts; Phase 3: pilot test and re-test instruments with a nationally-representative sample of 1,400 youth via an online survey to establish psychometrics; Phase 4: field test and validate short forms of the instruments with a healthcare sample (n=315) to create developmentally-appropriate screening tools for non-research settings. Phases 1 and 2 will take place in Year 1; Phase 3 in Year 2; and Phase 4 in Year 3. Classical Test Theory (CTT) and Item Response Theory (IRT) will be used to
establish psychometrics. A content-based analysis approach will be used for qualitative analyses. Domestic violence advocates, and research experts, will participate in a Delphi process to
strengthen the validity of the item pool prior to psychometric testing. In addition, the project will generate two quantitative datasets for archiving (the national 'data and clinical sample data). Multiple
scholarly papers from this experienced and ideally-prepared interdisciplinary team will also be published in criminology, public health, pediatric, and psychology peer-reviewed journals. ca/ncf