Mark Guiberson, Barbara L. Rodríguez, Philip S. Dale; Classification Accuracy of Brief Parent Report Measures of Language Development in Spanish-Speaking Toddlers. Lang Speech Hear Serv Sch 2011;42(4):536-549. doi: 10.1044/0161-1461(2011/10-0076).
As a follow up to last month’s Research Tuesday article, I chose this research article as my second topic. Also in my workplace, we use the SPLS as our qualifying assessment for Spanish-speaking children, so I was curious to see what these researchers had to say.
The goal in this project was to evaluate the classification accuracy of three different parent report measures as they assess they language development of Spanish-speaking toddlers. The three parent report measures chosen were the Ages and Stages Questionnaire (Spanish ASQ; Squires, Potter, & Bricker, 1999), he short-form of the Inventarios del Desarrollo de Habilidades Comunicativas Palabras y Enunciados (INV–II; Jackson-Maldonado, Bates, & Thal, 1992; Jackson-Maldonado et al., 2003), and reported children’s 3 longest utterances (M3L–W). The children were also administered the Spanish Preschool Language Scale-4 (SPLS–4; Zimmerman, Steiner, & Pond, 2002) to assess concurrent validity of the parent report measures in comparison to the Expressive Language Subtest.
Who took part in this study? 45 Spanish-speaking families and their 2 year olds. 22 of the children had expressive language delays (ELD), and 23 of the children were typically developing in their language development (TD.)
What’s going on with these parent measures? Well, the Communication Subscale of the ASQ is six questions long. The short-form INV-II is a 100 word checklist with a question regarding combining words. The M3L-W is assessed by asking parents to write down the three longest utterances their child has produced (To calculate a score you add the number of words and then divide by three.)
And what did we learn? Tell me about the results!
- All three parent measures were significantly correlated with the SPLS-4. They showed concurrent validity (a type of evidence that is demonstrated when a test elicits similar results to a test which has already been validated).
- When researchers compared the test scores of the children with expressive language delays and those test scores of the children considered typically developing, children with ELD received significantly lower scores than the children considered TD. (…which makes perfect sense? I guess that’s good to know – probably wouldn’t be great if both groups scored similarly.) The biggest difference was noted on the M3L-W measure.
- The classification accuracy aspect of the project looked at sensitivity, specificity, negative predictive value, and positive predictive value. Sensitivity of the measures reveals how many of the children who had a dx of ELD, also tested as ELD. Specificity of the measures reveals the portion of children considered TD, who tested as TD. Negative predictive value (NPV) is the percentage of children with negative screening results who were accurately classified, and the positive predictive value (PPV) with positive screening results that were accurately classified. (This is all very confusing for me, I’m hoping as I read it becomes clearer.)
- The ASQ showed low sensitivity and NPV, but strong specificity and PPV – the ASQ was determined to be inadequate at detecting children with ELD (Possibly due to the fact that the questions are direct translations from English, and the ASQ was developed specifically for parental styles typical for Europe and America.)
- The INV-II had high sensitivity, specificity, NPV, and PPV – it “shows promise” for use as an expressive language screener
- The M3L-W had high sensitivity, specificity, NPV, and PPV – promising in terms of “pass/fail” screenings to determine need for in-depth assessment
- Since the INV-II and the M3L-W had similar results, the researchers performed another test called a “receiver operating characteristics” which revealed the M3L-W was “non-significantly” better and demonstrated stronger classification accuracy.
To sum it up, when screening toddler-age, Spanish-speaking children it is appropriate to use a vocabulary checklist as well as a parent report on MLU to gain clinical information prior to an in-depth evaluation. But don’t bother with the translated ASQ.
Stay tuned kids, for Research Tuesday in April!
Jackson-Maldonado, D., Bates, E., Thal, D. (1992). Fundación MacArthur: Inventario del desarrollo de habilidades comunicativas. San Diego, CA San Diego State University
Jackson-Maldonado, D., Thal, D. J., Fenson, L., Marchman, V. A., Newton, T., Conboy, B. (2003). MacArthur Inventarios del Desarrollo de Habilidades Comunicativas user’s guide and technical manual. Baltimore, MD Brooke
Squires, J., Potter, L., Bricker, D. (1999). Ages and Stages Questionnaire user’s guide. Baltimore, MD Brookes
Zimmerman, I. L., Steiner, V. G., Pond, R. E. (2002). Preschool Language Scale, Fourth Edition, Spanish Edition. San Antonio, TX Harcourt Assessment