I've decided to review research papers each month in the year 2002. These will be articles that come to my attention that I think should be read by bilingual clinicians because I believe they are important and could affect clinical practice.
I always get excited about research and enjoy talking to people about their work and research problems. Research isn’t easy and I’m always impressed with a new article that has answered a few questions that logically we knew the answer but didn’t have the evidence for our colleagues who want data. I’m referring to an article by Restrepo and Silverman in the November, 2001 issue of American Journal of Speech-Language Pathology. The title is "Validity of the Spanish Preschool Language Scale-3 for Use with Bilingual Children". I hope that you’ll read this paper.
Abstract
"This study evaluated the validity of the Spanish edition of the Preschool Language Scale-3 (I.L. Zimmerman, V.G. Steiner, & R. E. Pond, 1993). As a preliminary step, the authors reviewed the test to determine whether it met psychometric criteria established by McCauley and Swisher (1984) and Hutchinson (1996). Examination of the test’s psychometric characteristics revealed that the test met only 4 out of 10 criteria proposed by McCauley and Swisher and none of the additional criteria from Hutchinson. A problems were evident in the test’s norming and in the lack of reliability and validity data. The authors then investigated whether, despite the test’s psychometric shortcomings, it was useful for the assessment of Spanish/English-speaking children. Results revealed that the children studied performed approximately 1.5 SD below the mean. Moreover, the children’s performance on the subtests did not reflect an even progression of item difficulty, indicating limited evidence of construct and content validity."
I know that the PLS-3 is a popular test for evaluating English and Spanish-speaking children. I’ve told my students over the years that it had to be the worst one on the market that could be used with Spanish-speaking preschoolers. I’ve tried them all. Before you read the article I hope you’ll take the time to read the Spanish manual for the PLS and take a look at the norms that the authors report for Spanish speakers. This will prepare you for the article and give you a greater understanding and appreciation for the critique of the test and then the presentation of the results and discussion.
These authors’ item analysis of the PLS-3 (Spanish) found problems with the content of specific items. They indicated that the children’s performance could have been affected by lack of familiarity with the test format, difficulty understanding poorly translated items, difficulty with the vocabulary used as the level of difficulty increased or decreased with translation, and scoring that penalized children’s culturally appropriate responses. I’m not sure what we’re assessing when we do use the PLS-3 Spanish version. I just hope that its not used exclusively to determine whether a child is language impaired. I understand that Psychological Corp is developing a new Spanish version of the PLS. I hope that this new test will be different and take into consideration the criteria for good test construction.
Restrepo and Silverman stress throughout the article that parent report and a language sample has been found to be the best methods of determining whether a language disorder exists in a Spanish-speaking child. They refer to 4 measures that have been reported as reliable: number of grammatical errors per T-unit, parental concerns, family history of speech/language problems, and mean length of T-unit. The reference to how T-units are computed is in the article.
Restrepo and Silverman also used a parent questionnaire to help identify parent concerns. The article had the English translation. You can probably write Restrepo at the University of Georgia at Athens for the Spanish version. Her dissertation from the University of Arizona had a lengthy Spanish questionnaire that may also be useful to you. The article’s shorter version focuses on the child’s use of the Spanish language and doesn’t ask about family history of communication disorders. I always ask for the familial history of speech and language disorders.
The article is "must reading." I’ll probably use it as a tutorial with my students just to get them to learn how to argue the limitations and/or merits of a test that is used with Spanish-speaking children that are not theoretically constructed for Spanish-speakers and normed for this population.
I hope that this will be a healthy, prosperous, and happy New Year for you. Happy reading!
