5
HYPOTHESES
My research is meant as half-confirmatory and half-exploratory.
1. The confirmatory part answers the question whether there is any systematic pattern
relating tolerance choices and social background measures. The existence of meaningful (and
mainly, robust) patterns reinforces the standpoint of the “pluralistic intolerance” thesis and casts
doubts on the possibility of education alone to lead to higher levels of tolerance. Underlying the
attitudes then, we should suspect the effect of perceived interests,
3
and the will to protect them
even at the price of exclusion of certain groups from civil liberties.
2. The exploratory research question is: what is the pattern and how strong is it?
I hypothesize that less intolerance toward atheists and communists is related to lower-
income background, liberal/ Democratic options, and urban environment, while less intolerance
toward racists and militarists better associates with higher-class background and conservative/
Republican options. (I do not expect opinions on homosexuals to be related to either, with the
exception of urbanism.)
DATA AND OPERATIONALIZATION
The basic test is a regression analysis of the joint 1998 and 2000 GSS data. My
dependent variable is the tolerance battery (questions 76abc through 81abc, see Codebook,
Appendix1). I recoded the answers to range from 0 to 3 for each target group. Thus I worked
3
At least in realistic conflict interpretations this would be the underlying cause. Sniderman et al.
(2004) have applied some social identity-type explanans, too, and found evidence that fearing
the fate of the own national culture has a substantial impact on intolerance toward immigrants.
The authors do not think that the two explanations are mutually exclusive. More radically, I
surmise that social identity factors are but a specific subgroup of interest-related factors.
Convention