Having noted these generalities, let us turn now to some more specific
considerations of teachers' views of disadvantaged speech, bearing in mind
as we do so the assumption made here of the linguistic validity of all dialects.
Some more useful American research has been done by Williams and his
colleagues, working with black, white and Mexican-American children (see,
generally, Williams, 1976). In a number of studies it was found that when
white and black teachers evaluated children on the basis of their speech,
along a number of semantic-differential dimensions, two underlying factors
emerged (see also above). One of these, labelled "confidence-eagerness",
reflected such things as the perceived confidence and social status of the
child. The other factor, "ethnicity-non-standardness", was also associated
with judgements of social status and, as well, with perceptions of ethnicity
and the standardness-non-standardness of the child's speech. These findings
were expanded when Williams et al. (1972) asked white and black teachers
to evaluate low- and middle-status black, white and Mexican-American children in Texas. First, the teachers were requested to provide semanticdifferential
evaluations of the three ethnic groups as presented to them via a
simple written label - this presumably being a measure of teachers' general
and overall stereotypes. Second, teachers evaluated videotaped samples o(
children's speech and, third, they were asked to estimate each child's classroom
achievement.