ABSTRACT
This chapter explores some of the more sophisticated attempts to use aggregated data in delineating empirical generalizations on racial violence. It presents various studies that represent pioneering attempts to invent modes of assembling and interpreting data in generating or testing hypotheses about collective violence. The criteria of "property crimes with the use of violence" have remained somewhat more constant, and D. S. Thomas has shown that the numbers of these crimes correlate negatively with economic indices. The Hovland and Sears article of three decades ago attempted to test the frustration-aggression hypothesis by correlating selected economic indices with lynchings. Riots, as distinguished from lynchings and other forms of collective violence, involve an assault on persons and property simply. The chapter deals with Gurr's interesting comparative study of civil strife in 114 nations, done in an attempt to isolate the influence of such variables as persisting deprivation, history of past strife, perceived legitimacy of government, and a number of mediating variables.
