Been a Heavy Life: Stories of Violent Men

4. Research Methods When Research Is Being Researched

4 RESEARCH METHODS WHEN RESEARCH IS BEING RESEARCHED Early on I was helped in my methodological decision making by the conventions of qualitative sociology. Some very general tools for working can be taken for granted. For example, the qualitative sociologist typically uses nonprobability sampling. If the researcher intends to conduct interviews, the interview format tends to be open-ended. The analysis usually emphasizes the perspectives of those whom one is studying, but also incorporates one’s own perspectives into analysis and documentation, the latter being the reflexive position described in chapter 3. At the design stage, then, I had the welcome feeling that my research methods were for the most part handed down to me. This impression turned out to be illusory, as I had few guidelines to follow when it came to analyzing the interview’s shaping influence on the narratives of my participants. I encountered few precedents for a methodological process that would yield a close and deep examination of the “how” of talk, and not just the “what” (Holstein and Gubrium 2000). I constructed my own process (Presser 2005). Thus, this chapter follows a trajectory from fairly derivative to more creative decision making. W H E N R E S E A R C H I S B E I N G R E S E A R C H E D . 47 Accessing Violent Men My research participants were referred to me by agencies and organizations working with convicts, ex-convicts, and homeless people. Thirteen men were from Harrison, New Jersey (near New York City); ten were from Cincinnati; two were from New York City; one was from Philadelphia; and one was on death row during the research, in a state withheld to disguise his identity.1 Participation was voluntary, but only one potential research participant declined to participate. I also presented tape recording as optional, but all of my participants consented to being tape recorded. All of my research participants were males seventeen years of age and older who had been convicted of committing at least one violent crime or—though never convicted—reported committing at least one violent crime. Most of the participants (twenty-five of the twenty-seven) were in fact convicted of a violent crime: this comes as no surprise, since I had asked my contacts to refer me to men with such a conviction history. I reasoned that being labeled violent might cause some havoc to one’s identity. But two of the men referred to me had not, after all, been so convicted. Table 1 provides select sample information: pseudonym, age at first interview, race (racio-ethnic group, based on self-report), and violent offenses based on self-reports and official records. In parentheses are self-reported crimes of which subjects were not convicted. Multiple counts of a charge are not noted. My working definition of violent crime was an act intended to cause and causing harm to another person. The violent crimes committed by my research participants were assault, robbery, rape, and/or murder. Many of the men had committed more than one of these offenses. Some men told me that they had committed additional violent crimes unbeknownst to law enforcement officials. Only two of my research participants denied having committed the violent charges of which they were convicted. Several more men disputed offense specifics and not the fact that they had violated another person.2 Originally, I wanted to expand the taken-for-granted meanings of violence by including persons who were responsible for harming people in their capacity as professional decision makers. Such persons, if they are called “offenders” at all, are commonly called white-collar offenders. My early efforts to include them failed. Though I tapped a fairly wide network of acquaintances and friends working in criminal justice, law, and business, I was unable to find persons willing to be interviewed, conceivably because of outstanding civil 48 . B E E N A H E AV Y L I F E liability risks. Consequently, my sample consists of offenders traditionally defined as violent, with direct bodily harm a basic aspect of that definition. The decision to sample men exclusively was made early on. It was influenced by scholarly work on masculine identity construction and violent offending (Connell 1995; Gilgun and McLeod 1999; Messerschmidt 1993, 2000). I was also inspired by the feminist call (J. Allen 1989; Naffine 1997) to understand crime for its sex specificity...