Improving Global Measurement of Intimate Partner Violence Open Access

Abstract

Intimate partner violence (IPV) is a pervasive global health issue with negative impacts on the well-being of victims, their children, and society. Increased awareness of these health and social impacts has led to increased efforts to measure and address IPV, including an international target to reduce IPV against women and girls in the Sustainable Development Goals. However, significant limitations in current IPV measurement necessitate improvements to commonly-used survey instruments and analytic techniques. This dissertation focuses on addressing specific limitations in IPV measurement using large, population-based survey data. These limitations are (1) measurement error due to self-report; (2) systematic bias due to underreporting; (3) lack of consensus in IPV domains and outcome construction; (4) lack of equivalence across countries; (5) small item sets with limited construct validity. Chapter 3 addresses issues (1) and (2) using multiple overimputation and multidimensional bias analysis, while Chapters 4 & 5 address issues (3), (4), and (5) using latent modeling techniques, including factor analysis and item response theory. In the first of these investigations, multidimensional bias analysis established reasonable bounds of sensitivity and specificity for general health survey measures of IPV, with sensitivities in some surveys as low as 10%-20% under the assumption of high specificity. Multiple overimputation generally produced IPV prevalence estimates similar to those from the violence-focused surveys, except where these estimates were low and highly discrepant between general and violence-focused surveys. The latter two investigations, in lower-to-middle and high-income countries, respectively, reveal that IPV is measured as a primarily unidimensional construct with nuanced domains, including controlling behaviors, physical assault, sexual abuse and coercion, and emotional abuse. While not globally equivalent, common quantitative scales generally measure IPV comparably across countries within geographical regions. Current data on IPV can be used more effectively by applying bias analytic techniques and by incorporating non-physical acts into prevalence estimates, where available. To improve global IPV measurement in the future, internal validation using anonymous methods or clinician interviews, mixed-methods refinement of scales via psychometric validation and cognitive interviewing, and community-based participatory research to facilitate flexible standardization and inform our understanding of the factors influencing reporting and underreporting are needed.

Table of Contents

CHAPTER 1: INTRODUCTION.. 1

1.   Intimate Partner Violence and Its Consequences. 1

2.   Sustainable Development Goal 5 and IPV Measurement. 1

3.   Measurement of IPV in Historical Context. 4

4.   Ongoing Issues in IPV Measurement. 6

5.   Potential Solutions and the Future of IPV Measurement. 10

CHAPTER 2: BACKGROUND & SIGNIFICANCE. 24

CHAPTER 3: Addressing measurement error in IPV self-report data using multiple overimputation and multidimensional quantitative bias analysis. 33

CHAPTER 4: Measurement structure and regional invariance of the Demographic and Health survey intimate partner violence items: A comparative confirmatory factor analysis. 57

CHAPTER 5: Measurement structure, invariance, and item functioning of intimate partner violence scales in high-income countries: Results from 30 population-based surveys. 86

CHAPTER 6: OVERALL DISCUSSION & CONCLUSIONS. 114

About this Dissertation

Rights statement

• Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.

School	Laney Graduate School
Department	Global Health and Development
Degree	Ph.D.
Submission	Dissertation
Language	English
Research Field	Psychology, Psychometrics Health Sciences, Epidemiology Health Sciences, Public Health
Keyword	intimate partner violence invariance testing factor analysis measurement bias analysis
Committee Chair / Thesis Advisor	Cari Jo Clark, Emory University
Committee Members	Angela M. Bengtson, Emory University Timothy L. Lash, Emory University Regine Haardörfer, Emory University

Last modified

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	Improving Global Measurement of Intimate Partner Violence ()	2026-05-14 14:10:08 -0400	Press to Select an action Download

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions
	Supplemental File 1 (Chapter 3) (Tables for Chapter 3)	2026-05-14 14:10:18 -0400	Press to Select an action Download
	Supplemental File 2 (Chapter 4) (Tables for Chapter 4)	2026-05-14 14:10:35 -0400	Press to Select an action Download
	Supplemental File 3 (Chapter 5) (Tables for Chapter 5)	2026-05-14 14:10:58 -0400	Press to Select an action Download