Scoring Interpersonal Violence Measures: Methodological Considerations.

ABSTRACT

BACKGROUND: Interpersonal violence, such as intimate partner violence, sexual assault, and adverse childhood experiences, is a significant global health concern. A major challenge to nurses and others working in the field of interpersonal violence deals with the complexity involved in measuring interpersonal violence. Numerous validated instruments exist; however, there is no standard approach for scoring these instruments. There is also a tendency to examine different forms of violence separately, not accounting for the known co-occurrence of violence. This has led to confusion as the interpretation of results often differs depending on the specific method used. OBJECTIVES: The purpose of this article is to summarize the major methods for scoring interpersonal violence measures and implications of each approach with a specific focus on co-occurrence. METHODS: The article begins with a summary of the primary goals of measuring interpersonal violence, major methods for scoring interpersonal violence measures, along with scoring challenges. We then provide a case exemplar examining the relationship between interpersonal violence and posttraumatic stress disorder symptoms to illustrate how scoring methods can affect study results and interpretation of findings. RESULTS: Our article shows that each scoring method provides a different picture of the distribution of interpersonal violence experiences and varies regarding the ease of interpretation. Scoring methods also affect interpretation of associations between interpersonal violence and other factors, such as having statistical power to detect significant associations. Accounting for the co-occurrence is critical for making accurate inferences by identifying potential confounding interactions between different types of violence. DISCUSSION: The application of different scoring methods leading to varying interpretations highlights the need for researchers to be purposeful when selecting a method and even applying multiple methods when possible. Recommendations are provided to assist researchers and providers when making decisions about the use of scoring methods in different contexts.