Duration and Severity
of Woman Battering
A Conceptual Model/Continuum
ALBERT R. ROBERTS
This study provided the data to develop a new five-level classificatory schema or continuum of the duration and chronicity of woman battering based on interviews with 501 battered women. Having no preconceived ideas of what this typology would actually look like, we were able to examine women's experiences and develop a framework based on the common themes presented by their interviews. Our goal is to describe different types of battering relationships with a greater level of detail to enhance our understanding of the variety of ways in which battering relationships occur and endure.
The subjects included 501 battered women and were comprised of four subsamples from a state women's prison; from two shelters for battered women; from three suburban police departments; and from a modified snowball sample. A snowball sample is a nonprobability method of sampling used when a population or group is very difficult to find by random sampling, and where members of the subsample recommend one another to the interviewer. Each interviewer and subject is asked whether they know of a person that fits the proposed characteristics of the potential sample. If they know such a person, then they are asked to furnish the first name and phone number of a potential interviewee who may be a neighbor, friend, relative, or former acquaintance. Snowball sampling is the best way to locate hidden groups such as drug abusers, incest survivors, and wealthy battered women who usually have no contacts with formal agencies and social institutions