| Summary: | The purpose of this study was to examine the validity of the DELV-ST by comparing it to two other screeners, the Fluharty-2 and the Washington-Craig Language Screener (WCLS). The participants were 73 African American Pre-K and Head Start children, aged four- to five-years-old.
Fail rates were higher than what has been reported in the literature. They were highest for the Fluharty-2 (57%), lower for the DELV-ST (52%), and lowest for the WCLS (46%); however, there were no statistical differences in the fail rates by screener. Approximately 54% of the children passed or failed all screeners. Unfortunately, the remaining 46% failed one or two of the screeners, with 91% of the children failing the first or second screener given and only 9% failing the third. Thus, order or practice effects seemed to contribute to the findings. Indeed, when the overall fail rate was recalculated using the results of the third screener, the fail rate was lower at 33%.
Fail rates did not vary statistically by the childrens gender, caregiver education, and use of nonmainstream English. The childrens gender, caregiver education, and use of nonmainstream English were also independent of their screening performance, except the childrens listener judgment dialect rating was positively correlated to their MCLUw, their caregiver education was positively correlated to their Receptive Language Quotient scores from the Fluharty-2 and the Wh-Question scores from the WCLS, and their age was negatively correlated to their DELV-ST error scores and positively correlated to their Fluharty-2 scores and the Wh-Question scores from the WCLS.
There was also some evidence of convergent and divergent validity among the screeners; however, not all tests of these relationships provided evidence for the validity of the screeners as suggested by the test developers.
Together, these findings suggest that the DELV-ST is comparable to the Fluharty-2 and the WCLS for screening low-income, AA children. Future studies are needed to evaluate the predictive validity of the three screeners and further investigate the role test practice may play in low-income, AA childrens screening results.
|
|---|
