the breadth of resistance to civil rights in the North was limited. But if the hypothesis is
supported, then it would strongly substantiate the argument that both white backlash and
racial conservatism among Republicans did not emerge as a response to the unadvised
adoption of race-attentive policies like affirmative action in the late-1960s. It lend further
support to the argument they they emerged in earlier decades as a response to race-neutral
policies like fair employment practices as well as fair housing. The origins of electoral
realignment would have to be sought elsewhere.
MODEL
Based on the foregoing theoretical discussion, I examine empirically the impact of
white resistance and Republican control of state government on the passage of fair
housing legislation. I do so using discrete-time, event-history methods. For a variety of
reasons, I consider discrete-time methods superior to continuous-time methods for
assessing the passage of state legislation. The measurement of time is relatively inexact in
the study of legislation. When recording data on time of passage, it makes no difference
whether a law passed at the beginning or end of a given year. Both laws are counted as
having passed in the same year. At the same time, if one law passes in December 1956
and another law passes in January 1957, they are measured as having passed in entirely
different years even though they are separated chronologically by only a month. What
matters is not the precise time of passage but the legislative session with which the event
is associated. Furthermore, in the specific case of fair housing legislation, there are
several years in which more than one state passed a housing law. Discrete-time methods
handle these “tied events” far more capably than continuous-time methods. Most
Convention