Innovative Methods of Gathering Survey Data on Violence Against Women

• DOI: https://doi.org/10.1108/S1521-613620190000024008
• Published date: 26 August 2019
• Pages: 69-84
• Date: 26 August 2019
• Author: Walter S. DeKeseredy

69

CHAPTER 4

INNOVATIVE METHODS OF

GATHERING SURVEY DATA ON

VIOLENCE AGAINST WOMEN

Walter S. DeKeseredy

ABSTRACT

Purpose – This chapter presents some innovative ways in which researchers can

collect survey data on various types of violence against women.

Methodology/approach – The suggestions made here are drawn from over 30

years of national, international, and local survey research.

Findings – The methods described in this chapter minimize underreporting,

produce theoretically relevant data, and have meaningful policy consequences.

Originality/value – The research techniques reviewed here have made many

important contributions to the eld and the data they uncovered have helped

raise public awareness about one of the world’s most compelling social

problems.

Keywords: Violence against women; surveys; research methods;

feminism; gender; questionnaires

INTRODUCTION

Over 30 years of rigorous survey research conclusively shows that women

around the world are abused by men in “numbers that would numb the mind

of Einstein” (Lewis cited in Vallee, 2007, p. 22). Furthermore, the rates of

Methods of Criminology and Criminal Justice Research

Sociology of Crime, Law and Deviance, Volume 24, 69–84

Copyright © 2019 by Emerald Publishing Limited

All rights of reproduction in any form reserved

ISSN: 1521-6136/doi:10.1108/S1521-613620190000024008

70 WALTER S. DEKESEREDY

various types of violence against women in Canada and the United States have

not dropped over the past 20 years despite progressive legislative reforms, the

ongoing efforts of feminist coalitions’ lobbying and education initiatives, and

other major steps to enhance women’s health and safety (DeKeseredy, Hall-

Sanchez, & Nolan, 2018; Senn et al., 2014). These two claims are buttressed by

small- and large-scale surveys that used the methods suggested in this chapter.

The techniques recommended here are not exhaustive, but are superior to

those used in mainstream crime victimization and self-report surveys (e.g.,

the National Crime Victimization Survey) that are not specically designed to

capture rich quantitative and qualitative data on male-to-female physical and

sexual assaults, sexual harassment, stalking, image-based sexual abuse, and other

gendered harms that exist on what Kelly (1987, 1988) refers to as the “continuum

of sexual violence.” Perfect male-to-female violence surveys are not possible, but

good ones can and should be done. The methods reviewed in this chapter are

important steps toward achieving this goal.

BROAD OPERATIONAL DEFINITIONS

Dening violence against women is much more than “a matter of semantics.”

Major debates over denition are not trivial; they seriously affect how data are

gathered and the quality and quantity of social support services for women who

are beaten, sexually assaulted, and abused in other ways by current or former

male partners or acquaintances (DeKeseredy, 2011). Further, denitions are used

politically as tools in social struggles. Together with poverty, unemployment, ter-

rorism, and other social problems, violence against women is a highly politicized

topic of social scientic inquiry, and denitions of this harm reect this reality

(DeKeseredy & Schwartz, 2011; Ellis, 1987).

Some feminist scholars contend that only physical behaviors, such as beatings

and forced vaginal penetration, that meet criminal code standards should be

accurately measured because this is the way to “mainstream” gender into the

measurement of violence against women (Walby et al., 2017). However, as

Myhill (2017) puts it, while accurate measurement of behaviors that only involve

physical contact is commendable, “such an approach will paradoxically obscure

the gendered nature of abuse in other respects” (p. 38). Before addressing this

point, it is rst essential to focus on some additional problems by using narrow

legalistic denitions.

Consider Tjaden and Thoeness’ (2000) US National Violence Against Women

Survey (NVAWS). It was introduced to respondents as a crime and personal

safety survey, one that excluded a broad range of hurtful behaviors exempt from

the purview of criminal law (DeKeseredy & Schwartz, 2011). Not surprisingly,

then, the NVAWS uncovered a very low incidence rate (1.8% in the previous

12 months) of women victimized by one or more of these acts committed by an

intimate partner: rape, physical assault, rape and/or physical assault, and stalking.

The problem is that unless women clearly label hurtful behaviors as “criminal” in