Advertisement

Fear and Its Relationship to Crime, Neighborhood Deterioration, and Informal Social Control

  • Stephanie W. Greenberg
Part of the Research in Criminology book series (RESEARCH CRIM.)

Abstract

Two important models of the fear of crime have emerged in recent years—the victimization model and the social-control model (Lewis & Salem, 1980, 1981). According to the victimization perspective, a high crime rate leads to a high victimization rate, which leads to a high level of fear in anticipation of being victimized. The social-control model hypothesizes that the deterioration of social control, or the perception that this has occurred, is the source of fear, more than the objective risk of victimization. Several studies have found that the availability of social support or resources to deal with neighborhood problems alleviated fear, particularly in highly threatening residential environments (see e.g., Greenberg, Rohe, & Williams, 1984a; Skogan & Maxfield, 1980; Taub, Taylor & Dunham, 1984). These resources include local networks and community involvement at the individual level, and social cohesion and availability of community organizations at the neighborhood level.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Reference

  1. 1.
    For a detailed empirical and theoretical overview of these three models see Greenberg, Rohe, and Williams (1984a,b).Google Scholar
  2. 2.
    “Worry” is a four-item index that expresses relatively vague, nonspecific anxieties about crime, such as feeling uneasy about hearing “footsteps behind me at night” m the neighborhood. The method of scale construction and the scale reliability are found m Greenberg, Rohe, and Williams (1984b, Appendix A).Google Scholar
  3. 3.
    An attempt was made to use rates of personal, property, and nonindex crimes, but the high degree of multicollinearity among these variables (r >.68) precluded their simultaneous inclusion in the same equation.Google Scholar
  4. 4.
    For details of factor analysis and scale construction, see Greenberg, Rohe, and Williams (1984b; Tables A3–A5).Google Scholar
  5. 5.
    The adjustment in R2 refers to the correction made for the number of predictor variables relative to the number of cases.Google Scholar
  6. 6.
    As a general rule, nonlinear multivariate statistical techniques will achieve the same results as standard linear regression as long as the distribution of the dichotomous dependent variable falls between 25% and 75% (Cohen & Cohen, 1975). In the case of safety satisfaction, 73% stated they were somewhat or very satisfied with the safety of their neighborhood, which falls just within the acceptable range.Google Scholar
  7. 7.
    As in the previous analysis, dichotomous dependent variables (low-crime, known strangers, real home, community association membership, good investment perceived housing quality) were also analyzed by the discriminant technique. There were no differences between regression and discriminant analysis in either sign or ranking of predictor variables. For those who are interested, simple correlations among the variables used in this analysis appear in Appendix B in Greenberg, Rohe, and Williams (1984b).Google Scholar

Copyright information

© Springer Science+Business Media New York 1986

Authors and Affiliations

  • Stephanie W. Greenberg

There are no affiliations available

Personalised recommendations