I got an email from a clinician in New York who is using the Preschool Language Scale-4 Spanish version and wanted to know my experiences with this new test and whether I thought it would be appropriate with the Caribbean children he was evaluating.
I told the clinician that I don‚t have much experience with
the new version since moving to Missouri but my first reaction
was to say that it probably isn't appropriate. I wasn‚t sure
of the norming population for this test and assumed that if the
group norms are from another Spanish speaking group, it probably
wouldn‚t be familiar to the children he was seeing in New
York. I decided to take a look at the PLS-4 Spanish.
The PLS-4 Spanish is an individually administered test used to
identify monolingual or bilingual Spanish speaking children who
have a language disorder or delay. The age range is from birth
through 6 years-11 months. Administration time is 20 to 45 minutes
depending upon the age of the child. The test content is different
from English but there are shared test tasks and content. Early
milestones may be identical but the content changes as the age
group gets older. PLS-4 Spanish yields norm-referenced test scores
for Auditory Comprehension andExpressive Communication subscales,
as well as a Total Language Score (standard scores, percentile
ranks, and age equivalents).
The PLS-4 Spanish and English editions are not identical in terms
of components, test tasks, scoring, or research design. You cannot
directly compare scores from the Spanish edition and the English
edition of PLS-4 because of significant differences in the Spanish
and English Standardization samples. There are 2 standardized subscales, Auditory
Comprehension and Expressive Communication; the Supplemental Measures
are not standardized. These include the Articulation Screener,
Language Sample Checklist, and Cuestionario para los Padres. The
English and Spanish versions have different norming populations.
The tests have different numbers of items, similar but not equivalent
concepts, morphemes, and sentence structures. The tasks are not
necessarily of the same difficulty on the two tests.
Although there are some general similarities across the 2 editions
the differences include the following: 1) Components; Examiner's
manual, record, form, picture manual, Spanish-edition specific: 2) Test
items/item placement; Minor differences in placement/order to
age 3:0, with more significant shifts beginning at age 3:6: 3)
Scoring; subitems sometimes are different (e.g., Plurals on the
English edition: babies, horses; Spanish edition: limones, ratones)
based on data from Spanish speakers: 4) Research; Demographics
reflect U.S. census figures for Hispanic children; separate norms
based on data for Spanish speakers; PLS-4 Spanish sample is comprised
of bilingual children. If you are testing a child new to the US
and/or living in a completely Spanish-speaking environment, you
may choose to compare the child's score to means and standard
deviations reported for children living in a Spanish-speaking
country.
Demographic characteristics of the PLS-4 Spanish are the following.
1188 children (ages 2 days through 6 years-11 months) were in
the norming group compared to 1534 for the English version. They
came from the Northeast (62), North Central (16), South (597),
West (513). Therefore, 50 and 43 percent of the population came
from the South and West, respectively. In reporting the dialects
represented, 965 were of Mexican background, 75 Southwest USA,
33 Puerto Rico, and 29 Cuba. The rest of the sample came from
Guatemala (6), Honduras (2) , El Salvador (6), Nicaragua (3), Ecuador (2) Costa
Rica (2), Panama (5), Dominican Rep. (7), Peru (11), Colombia
(9), Venezuela (1), Argentina (7), and Other (23). Two participants
didn't provide this information. Sample is varied but not too
many from each group with the exception of Mexico.
The manual has a section "Scoring Bilingual Children's Responses".
This is important to read. Bilingual children may respond in both
English and Spanish. Therefore, mark the child's responses correct
if the child meets the "Pass" criteria for the task.
The vocabulary used in the PLS-4 Spanish test stimuli is typical of vocabulary used in the Southwest. Dialectal variations are listed on the test and these can be substituted. For children who are from a different region than those vocabulary items listed, the manual suggests that you interview the parent before the evaluation and substitute those words.
Do not reword test stimuli to simplify them for a child. Changing
the sentence structures is not acceptable because it compromises
the difficulty of test items. The tasks are ordered by difficulty
so changing the sentence structure changes the order and test.
If a child code switches, the child's response may be correct
for the task. The examples given include: The child has substituted
a word in English for one in Spanish; The child uses an English
grammatical form that is parallel to a Spanish form; the response
meets pass criteria for the task.
The "Development of Test Tasks" section of manual is
worth reading. There was an extensive attempt to get representative
test stimuli. There was literature review, clinician feedback,
reviews by a panel experts, and then they had a "try-out"
data collection phase. The purpose of this last phase was to evaluate
the new items. There was a 2nd review by another panel. There
was a study to develop scoring rules for the new open-ended tasks
on the Expressive Communication subscale and to refine the scoring
rules developed for the PLS-3 Spanish test tasks. Finally, the
tasks underwent a number of statistical analyses. The items and
tasks were not randomly selected or thought about during lunch.
Page 120 of the manual states "If you deviate from the standardized
testing procedures and stimuli (e.g., if you improvise on theadministration
directions to make them more appropriate for the child you are
testing), you will compromise the comparison of the child's scores
to the normative data. Changing the task or directions and adding
or omitting cues or practice tasks for a particular child changes
the difficulty of the test tasks and, in this case, invalidates
the use of the normative scores..Further, comparing the scores
of a child you have tested to the sample of age-level peers is
valid only if the characteristics of the sample are comparable
to the child you are testing (pg. 121)."
Page 13 of the manual provides "guidelines for Evaluating Spanish-speaking Children."They're general recommendations and worthwhile reading. Chapter 7 of the manual presents the reliability and validity studies of the PLS-4 Spanish. It is enough to say in this brief review that the reliability and validity evidence was good.
The bottom line, is this a test that I would recommend. Yes, there has been tremendous work on the test items, there were attempts to include clinicians in the development, and the work was extensive in developing a Spanish test that will be helpful to clinicians when other variables are used in combination with the test scores.
Never use a test alone to qualify a child into special services.
The manual recommends alternative assessment procedures, observations,
and interviews with informants. The manual makes it very clear
on page 140, that there are other factors that should be taken
into account when evaluating a bilingual child. These include:
the impact of culture on assessment; acquisition of a second language;
normal language development in bilingual children; factors influencing
the retention or loss of L1 skills; characteristics of language
loss; characteristics of children who may have a language disability;
code switching; non-standardized assessment practices; test adaptations/medications;
and finally, use of interpreters.
Use the test, just don't make the evaluation simple.
Visit us on Facebook
