19
shift from consideration of the ‘least-liked’ group to a ‘moderately-disliked’ group
resulted in the extension of less tolerance. In fact, the change produced a decrease of
tolerance in the amount of -.285 among these survey participants. Although this finding
is not surprising, it is important in that it underscores the point that tolerance decision-
making seems reliant upon the specifics of the situation under consideration.
Further, in this context, it is important to note that the responses given by the
respondents to the feeling thermometer rating disagreement with the ‘least-liked group’
were highly varied. The scores given to the ‘least-liked’ group in this survey ranged from
1 to 20, a highly varied finding given that the ‘disagreement’ portion of the scale only
ranges from 1 to 50. This result, in combination with the finding that tolerance decisions
are not general but group-specific, calls into question the assumption made by Sullivan,
Piereson and Marcus, that the ‘least-liked group’ technique yields a standardized, or
‘content-controlled’ measure.
Rather, this variation in the answers of survey participants lends inferential
support for the hypothesis of this study, that tolerance decisions are more appropriately
envisioned as a continuum decision-making conducted by each citizen. Contemporary
information may determine the placement of a particular decision on that continuum, a
view consistent with the broader theory presented in With Malice Toward Some. This
finding also suggests a fertile area of future research, particularly helpful to the assembly
of a more comprehensive model of tolerance decision-making. As models two and three
(below) suggest, the process by which American society contemplate various groups
organized in society may not be identical.
Convention