Premium
Early language screening in City and Hackney: the concurrent validity of a measure designed for use with 2½‐year‐olds
Author(s) -
LAW J.
Publication year - 1994
Publication title -
child: care, health and development
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.832
H-Index - 82
eISSN - 1365-2214
pISSN - 0305-1862
DOI - 10.1111/j.1365-2214.1994.tb00392.x
Subject(s) - test (biology) , concurrent validity , context (archaeology) , cutoff , test validity , receiver operating characteristic , measure (data warehouse) , psychology , predictive validity , identification (biology) , scale (ratio) , screening test , pilot test , developmental psychology , psychometrics , medicine , computer science , applied psychology , data mining , machine learning , pediatrics , geography , paleontology , physics , botany , cartography , archaeology , quantum mechanics , internal consistency , biology
Summary This paper reports data relating to the development of a screening test for language impairment in 2½‐year‐oId children. The screening test itself has previously been described. The results of a pilot study and a larger community study are reported. In all, 34 children were included in the pilot study and 1015 in the community study. The reference test selected was the Reynell Developmental Language Scales and the cutoff adopted 1·5 standard deviations below the mean for either the expressive or the receptive scale of the test concerned. The pass mark on the screening test was ascertained using receiver operating characteristics (ROC). The validity is reported given the cut‐off identified. The specificity, sensitivity and both positive and negative predictive abilities are reported for both the pilot and the subsequent study. The application of the test is discussed in the context of the current debate about early identification.