The Children's Embedded-figures Test (Karp & Konstadt, 1963) was developed as a measure of field-dependence for use with children aged from 5 to 12 yr. It is unspeeded and presents S with two series of tasks in which a basic shape is to be located within a complex perceptually embedding context. Internal consistency reliability estimates presented in manual (Witkin, et al., 1971) for children aged from 7 to 12 yr. range from .83.to .90. The only reported retest data for younger children are from a study by Dreyer, Nebelkopf, and Dreyer (1969) of 46 middle-class children. The time interval was 6 mo., children being 5% yr. at initial testing. The Pearsonian correlation was .87, suggesting that the CEFT is a reliable instrument for use with children in age range examined (Witkin, et al., 1971). The purpose of present study was to obtain reliability data for children from a working-class population and over a somewhat longer time period. The Children's Embedded-figures Test was administered to an original group of 53 kindergarten children as part of a larger study of relationship between field-dependence and perceptual egocentrism (Bowd, 1974). Children ( 18 girls, 29 boys) of those originally tested were re-tested 10 mo. later. The standard procedure was followed. The mean age of children when first tested was 5.11 yr., mean socioeconomic status was 47.7 (SD 5.53) on Blishen's (1958) scale. The retest correlation between scores over 10-mo. period was .80 ( 9 < .01). Field-dependence failed to correlate significantly with either sex or socioeconomic status for present sample, and mean score remained unchanged at 6.59 after 10 mo. The score appears rather less stable for 5to 6-yr.-old age group than may have been implied by Dreyer, Nebelkopf, and Dreyer's ( 1969) study which employed only a 6-mo. interval.
Read full abstract