Hierarchy and Reliability of the Preschool Language Scales-Fifth Edition: Mokken Scale Analysis.

Purpose This study examined the psychometric properties of the Preschool Language Scales-Fifth Edition (PLS-5 English) among preschool children from low-socioeconomic status (SES) families. Method The PLS-5 was administered individually to 169 3- to 4-year-old children enrolled in Head Start programs. We carried out a Mokken scale analysis (MSA), which is a nonparametric item response theory analysis, to examine the hierarchy among items and the reliability of test scores of the PLS-5 Auditory Comprehension (AC) and Expressive Communication (EC) scales. Results The PLS-5 EC items retained a moderate Mokken scale with the inclusion of all the items. On the other hand, the PLS-5 AC items formed a moderate Mokken scale only with the exclusion of five unscalable items. The latent class reliability coefficients for the AC and the EC scale scores were both above .90. Several items that violated the invariant item ordering assumption were found for both scales. Conclusions MSA can be used to examine the relationship between the latent language ability and the probability of passing an item with ordinal responses. Results indicate that for preschool children from low-SES families, it is appropriate to use the PLS-5 EC scale scores for comparing individuals' expressive language abilities; however, researchers and speech-language pathologists should be cautious when using the PLS-5 AC scale scores to evaluate individuals' receptive language abilities. Other implications of the MSA results are further discussed.

Language performance of low-income African American and European American preschool children on the PPVT-III.

PURPOSE: The performance of low-income African American preschoolers (36 to 52 months old) on the Peabody Picture Vocabulary Test--Third Edition (PPVT-III; L. M. Dunn & L. M. Dunn, 1997) was examined to provide a norm for assessing the performance of this population and to explore the link between socioeconomic status (SES) and language scores on the PPVT-III. METHOD: Four hundred and eighty-two African American and 52 European American children in a comparison group were individually administered the PPVT-III. RESULTS: On average, African American children performed approximately 1.5 SD below the expected mean based on national norms. Using standard cutoff scores, the PPVT-III identified more children as having language delays than did other measures of language abilities. Socioeconomic factors were related to PPVT-III scores, indicating that the degree of disadvantage within children with low SES was related to language abilities. Maternal education level, marital status, and the number of children in the household were uniquely associated with children's performance on the PPVT-III. CLINICAL IMPLICATIONS: The importance of supporting language development in preschool children from low-income families is discussed.

Problem behaviors of low-income children with language delays: an observation study.

Children from low-income families are at increased risk for significant behavioral and language problems. Early identification of these problems is essential for effective intervention. The purpose of the present study was to use multiple behavioral assessments to examine the behavioral profiles of sixty 3- and 4-year-old children from low-income families enrolled in Head Start programs and to compare the behavior characteristics of 32 children with language delays with those of 28 children with typical language development. Teachers completed the Child Behavior Checklist/Caregiver-Teacher Report Form/2-5 (CTRF; T. M. Achenbach, 1997) and the Social Skills Rating System (SSRS; F. M. Gresham and S. N. Elliott, 1990), and children were observed in the classrooms during structured and unstructured activities. Children with language delays exhibited more problem behaviors and poorer social skills on some of the observational measures than did children with typical language development, as predicted, but not on all.

The performance of low-income, African American children on the Preschool Language Scale-3.

This study examined the performance of 701 low-income African American preschoolers (36 to 52 months old) on the Preschool Language Scale-3 (PLS-3, I. L. Zimmerman, V. G. Steiner, & R. E. Pond, 1992). On average, African American children performed about 1 SD below the expected means for their ages on both the Expressive Communication and Auditory Comprehension subscales. Independent sample t tests showed no significant differences between African American children and a comparison sample of 50 European American children. Item analysis was used to examine the potentially problematic items of the PLS-3 for each age cohort of children. We found that 6 items appeared to be particularly difficult for the African American sample. The findings suggest that the PLS-3 is generally an informative language test for African American preschoolers; however, scores should be interpreted with caution.