112
bibliographic search was either neutral or favorable regarding the
LAS. Some of the criticisms of earlier forms of the test, such as
no alternate test forms or test-retest information (Berman-Benedek,
1979), have been corrected. Argulewicz and Sanchez (1982) find that
the LAS is . one of the more comprehensive and psychometrically
sound measures of language assessment" (p. 285). Mace-Matluck and
Dominguez (1981) conclude that the LAS I ". . underestimated the
first language ability of the children" (p. 3) and that teacher ratings
were a more accurate measure.
De Avila and Duncan (1981) devote 55 pages of their technical
manual to the issues of reliability and validity. They conclude that
. . oral language proficiency as assessed by
the LAS could be viewed as a necessary (albeit,
not sufficient) precondition for school achieve
ment. The data presented which addressed the
discriminate validity of the test showed the
test to be extremely capable in discriminating
limited from proficient native speakers, (p. 116)
In their 1982 study, De Avila and Duncan find that observational
data used in addition to the information gained through administration
of the LAS can increase the reliability of the language assessment.
They conclude that further research is needed utilizing mean length of
utterance (MLU) measures to determine "... the exact markers' which
go to make up the global score" (p. 20). The value of this procedure
will be to enhance both the reliability and validity of the assessment
procedure (De Avila & Duncan, 1982).
In reviewing five widely used oral language assessment instruments
for bilingual children, Bordie, Bernal, Bradley, Christian, Galvan,
Holley, Leos, Mace-Matluck, Matluck, Natalicio, Oakland, and Richard