Do some school-age children have no language? Some problems of construct validity in the pre-LAS Espanol

Bilingual Research Journal, Summer 2002 by MacSwan, Jeff, Rolstad, Kellie, Glass, Gene V

It must be emphasized that the Pre-LAS Espanol and other native language tests used in the United States are intended to assess children's oral language ability in the native tongue, not literacy. To avoid labeling non-literates as semilingual or alingual, we should carefully distinguish language, an integral part of every person's identity, from literacy, a technological development of use to some but not all individuals and human societies (for discussion, see MacSwan, 2000; MacSwan & Rolstad, in press; Wiley, 1996).3

However, some test makers include items on their oral language instruments that assess aspects of language use that are specific to academic culture-and, in some cases, items or subparts that are not specifically related to language ability at all. Doing so in the context of oral native language assessment, and characterizing the results as an index of native language ability, enormously privileges the educated classes and recalls the classic critique of prescriptivism (Labov, 1970).

Surely there is a specific character to the way language is used in many academic contexts, and it is reasonable to suspect that the use of this academic register-or "Discourse," as Gee (1996) prefers-will co-occur with success at school and mastery of academic content. A serious problem arises, however, when we construe this domain of language use as "more complex" or "developmentally superior" to the language children use natively at home or on the street, and consequently assess school language as indicative of "higher language proficiency." The Pre-LAS Espanol, like many commercially available language assessments, purports to identify children's Spanish-speaking ability, categorizing them as "fluent Spanish speaker," "limited Spanish speaker," or "non-Spanish speaker."

This way of conceptualizing language proficiency-in which academic registers are viewed as richer, more complex, and so on-not only lacks reasonable empirical and theoretical support (Martin-Jones & Romaine, 1986; Wiley, 1996; MacSwan, 2000; MacSwan & Rolstad, in press), but it also implies that entire cultures and communities that have rejected literacy and formal schooling have a "less complex" language than those in the "literate" world. The problem arises because culture-specific and language-specific characteristics are arbitrarily selected as defining characteristics of the abstract psychological construct of language ability-observed in all human communities, everywhere.

Below we consider the validity of the Pre-LAS Espanol in connection with these concerns. The six distinct parts of the test, which will be the focus of subsequent discussion, are described in Table 2.

Validity of the Pre-LAS Espanol

A psychological test such as the Pre-LAS Espanol is purported to measure a psychological trait-in this case, Spanish language proficiency. DeAvila and Duncan (1982) developed the LAS family of tests according to a view of language as consisting of four aspects: phonology (the sound system), the lexicon (vocabulary), syntax (principles that govern word order), and pragmatics (the use of language to satisfy particular goals) (cited in Guerrero & Del Vecchio, 1996).