Understanding rape survivors' decisions not to seek help from formal social systems

Health and Social Work, May, 2009 by Debra Patterson, Megan Greeson, Rebecca Campbell

Data Analysis

Data analysis proceeded in two phases. First, two analysts developed open codes (Strauss & Corhin, 1990) that captured key thematic content in the survivors' narratives. In the second phase, we used Erickson's (1986) analytic induction method for data analysis, which is an iterative procedure for developing and testing empirical assertions in qualitative research (see also Ryan & Bernard, 2000). In this approach, an analyst reviews all of the data sources multiple times, with the goal of arriving at a set of assertions that are substantiated based on a thorough understanding of all of the data. The next task is to establish whether each assertion is warranted by going back to the data and assembling confirming and disconfirming evidence. The analyst must look for five types of evidentiary inadequacy: (1) inadequate amount of evidence, (2) inadequate variety in the kinds of evidence, (3) faulty interpretative status of evidence (that is, doubts about the accuracy of the data due to social desirability bias), (4) inadequate disconfirming evidence (that is, no data were collected that could disconfirm a key assertion), and (5) inadequate discrepant case analysis (that is, no cases exist that are contrary to a key assertion) (Erickson, 1986). Assertions are revised or eliminated on the basis of their evidentiary adequacy until a set of well-warranted assertions remain. For this project, two analysts worked separately through Erickson's method so that the second analyst could provide independent verification of the assertions. Each analyst independently developed a preliminary list of assertions. The first analyst then expanded the list and tested the assertions against the data, refined, and in some instances eliminated them. Once the first analyst had completed what she thought was a well-warranted set of assertions and assembled confirming evidence, the second analyst then crosschecked those assertions against the data. Consensus was reached by the two analysts. Both examined the final list of assertions against the data and found them to be well-supported by the data.

RESULTS

Almost all of the survivors in this study identified multiple reasons that factored into their decision not to seek help (see Figure 1). First, more than three-quarters of the survivors anticipated that these systems would have rejected them personally and their expressed need for help. Some survivors discussed having had feelings of shame and unworthiness, which made them feel that they were unworthy of help. Similarly, some did not think they qualified for assistance. Second, more than three-quarters of the survivors did not believe formal social systems could have helped them. Survivors believed that these systems would not have been able to enhance their well-being or protect them from further harm by their perpetrators. Third, more than three-quarters of the survivors anticipated that formal social systems would have been harmful to their well-being. Survivors described how not seeking help was a form of self-protection against system personnel and processes they had perceived as hurtful. Each of these types of barriers to help seeking was identified by the majority of survivors in this study and should not be considered mutually exclusive groups.