This experiment investigated teacher judgments of functionally equivalent speech samples obtained from middle and lower socioeconomic status black and white children. Fifty-six female preschool and primary-grade teachers (9 black and 47 white) enrolled in an early childhood education graduate course rated and ranked taped picture descriptions. These descriptions were equivalent on features described but were different in speech patterns used to describe them. A Friedman analysis of variance yielded significant differences (p < .05) in both rating and ranking scores assigned to the four groups. Post hoc analyses indicate that the differences occurred between both the social class and racial groups. Results indicate that middle socioeconomic status children are rated and ranked more positively than lower socioeconomic status children and that white children are rated and ranked more positively than black children.Although children who speak nonstandard dialects may not experience difficulty in comprehending standard English (Hall & Turner, 1974;Quay, 1975), their nonstandard speech may indirectly hinder their academic success. That is, nonstandard speech may negatively influence teacher evaluations of the speaker's abilities, resulting in differential treatment of the nonstandard-speaking child. A first step in ascertaining whether this situation occurs is the examination of teacher evaluations of the performance of children whose speech patterns differ from middleclass speech patterns, when such performance is known to be equivalent for all speakers.Previous research findings have suggested that teachers consider the speakers of nonstandard dialects to be less adequate in their speech than speakers of standard English (Naremore,

Three age groups (7-, 8-, and 9-year-olds) and three socioeconomic status (SES) groups (lower, middle, and upper) were assessed on four cognitive variablesclassification, class inclusion, conservation of number, and conservation of substance-and on communication encoding effectiveness. It was shown that in addition to age and SES, general cognitive ability accounted for a significant proportion of the variance in communication encoding.Communication performance is assessed by presenting sets of identical arrays of stimuli (such as pictures) to dyads of children. The encoder (speaker) must describe a target stimulus so that the decoder (listener) can select it from the array. Both age-related and social-class differences in communication ability have been documented, with the performance of younger children inferior to that of older children (Glucksberg, Krauss, & Higgins, 1975) and the performance of lower socioeconomic status (SES) children inferior to that of middle-SES children (Baldwin, McFarlane, & Garvey, 1971;Quay, Mathews, & Schwarzmueller, 1977). Although some attempts have been made to explain these differences, there has been no empirical identification of particular variables that actually determine or mediate the age or social-class communication variation (Glucksberg et al., 1975). The findings of both age-related differences (Piaget & Inhelder, 1966) and social-class differences (Overton, Wagner, & Dolinsky, 1971) in cognitive skills suggest that age and SES may affect communication encoding primarily through cognitive level.

No abstract

This study evaluated the effects of age, the dyad's attention to and verbal labeling of target and nontarget stimuli, and listener instructions to ask questions on referential communication in 48 five-year-old and 48 eight-year-old dyads. A MANOVA for dyad activities indicated both age and dyad-activities differences. The dyads that located and verbally labeled the critical attributes were superior to the dyads that simply located the critical attributes and to control dyads on speaker performance. A MANOVA for listener instructions to ask questions yielded no age differences on the number of questions asked and no interaction between age and instructions to ask questions. It indicated that listeners instructed to ask questions and control listeners differed on the number of questions asked but not on communication accuracy.

No abstract