Many survey researchers believe that the quality of answers to attitud
e questions may be improved by using no-opinion filters to remove thos
e responses that are not based on true opinions. The authors investiga
te whether reliability, one aspect of data quality, is greater for fil
tered questions. Split-ballot experiments from two telephone surveys o
f the Akron, Ohio metropolitan area are used to estimate the effect of
filtered questions on the reliability of items from Srole's anomia sc
ale, Rosenberg's self-esteem scale, and a scale of attitudes toward la
wyers. There is no evidence that filtered questions improve reliabilit
y, and in the case of self-esteem, reliability may be reduced on the f
iltered form.