Assessment of Language Skills in Rural Preschool Children
Smith, Tina T., Myers-Jennings, Corine, Coleman, Thalia, Communication Disorders Quarterly
This study examined the extent to which linguistic variation in the English language might have been affecting the performance of 160 rural preschool children in Orangeburg, South Carolina, on four commonly used standardized tests. The children, ages 3-0 to 5-11 years, were administered the Patterned Elicitation Syntax Test, the Test of Auditory Comprehension of Language--Revised, the Test of Language Development--2 Primary, and the Test of Pragmatic Skills. Results revealed that when dialectal variations were not considered, the performance of children in this study differed from that of the normative population on tests that assessed grammatical morphemes.
Perhaps the most difficult task facing speech--language pathologists is the unbiased assessment of participants from culturally and linguistically diverse populations. It has long been argued that the primary factor contributing to the over-representation of minorities enrolled in special education programs is the cultural bias inherent in most standardized, norm-referenced tests (Chinn & Hughes, 1987; Gollnick & Chinn, 1990; Mercer, 1973). As a consequence, judicial action (Diana v. California State Board of Education, 1970; Larry P. v. Wilson Riles, 1972) and subsequent federal legislation (P.L. 94-142, 1975; P.L. 99-457, 1986) have mandated that standardized tests used to identify participants in need of special services be culturally and linguistically relevant and nondiscriminatory. Ironically, the same legislation that mandates culture-fair assessment of ethnic minorities requires the use of standardized, norm-referenced testing to qualify participants for federally funded special education services (P.L. 94-142, 1975).
To date, most attempts to create more culture-fair language tests have utilized one of two approaches: (a) renorming existing tests on various ethnic minorities and (b) adjusting the scores for dialectally sensitive items on standardized tests for speakers of a nonstandard-English dialect (Leonard & Weiss, 1994). While the first approach has clearly been the method of choice, a large number of standardized measures remain prejudiced against non-mainstream English speakers, particularly working class African Americans who use African American English. According to Taylor and Payne (1994), "little has been done to improve tests and other evaluation procedures for handicapped children, especially those with communicative handicaps, to make the tests linguistically and culturally valid" (p. 93). Although P.L. 94-142 mandates nondiscriminatory testing, making this a reality is not an easy task. Vaughn-Cooke (1983) examined and evaluated the following alternative approaches to traditional testing:
1. Standardizing existing tests on non-mainstream English speakers
2. Including a small percentage of minorities in the standardization sample when developing a test
3. Modifying or revising existing tests in ways that will make them appropriate for non-mainstream speakers
4. Utilizing a language sample when assessing the language of non-mainstream speakers
5. Utilizing criterion-referenced measures when assessing the language of non-mainstream speakers
6. Refraining from using all standardized tests that have not been corrected for test bias when assessing the language of non-mainstream speakers
7. Developing a new test which can provide a more appropriate assessment of the language of non-mainstream speakers. (p. 29)
Vaughn-Cooke found both strengths and weaknesses for each approach. For example, she pointed out that if a test had been developed to assess one dialect (e.g., Standard English), renorming it to assess speakers of a different dialect (e.g., African American English) "will not make the test valid or appropriate" (p. 30). Thus, it appears that ensuring that one receives a nonbiased clinical evaluation that is culturally and linguistically fair presents a challenge. …