The last couple of days I’ve been Newcastle, UK attending the child language seminar. I gave a keynote and I found myself thinking I was Sally Field, …they like me they really like me. I spoke on dynamic assessment and presented preliminary results from our diagnostic markers of language impairment study. The bottom line is that dynamic assessment appears to work to differentiate bilingual children with and without language impairment. like with previous studies of monolingual children, modifiability is the best predictor of language ability.
What’s exciting to me is the opportunity to talk with researchers from different parts of the globe who are interested in what dynamic assessment can provide. For example, Resa Nilipour and Yalda Kazemi both gave papers on measures they are developing for use in Iran. They each reported that there are no or very few tests available for SLPs to use. So, they are in the process of developing tests that will accurately identify Persian speaking children with language impairment. Until now, SLPs there use informal, clinician-developed measures and observations. I think that this is a great opportunity to develop a new test building on what we know about markers of language impairment, and building on the fact that markers could be different across languages. This is a HUGE improvement over previous approaches that blindly translated measures without regard to the structure of languages. There was another presentation on development of a screener for use in Brazil by Noemi Takiuchi. Unfortunately, I only caught the end of the talk, but from what I did hear and from the questions and comments it seems that they are making progress in developing a measure that is short and demonstrates differences between children with and without language impairment.
One conversation I was with Dr. Nilipour, who is working on a measure of grammar. He seems to be using a sensible approach of looking at different aspects of grammar via language samples derived from a story. He’s looking at use and accuracy of derivational morphology, verb inflections, verb agreement, MLU, NDW, among other things. He also has a short test that looks at many of the same grammatical forms in different ways. At this point he’s comparing performance by grammatical type, but I also suggested that he run item-level analyses, which will provide information about the best items for selection into a smaller set.
This represents only a smattering of the papers and posters on test development for languages other than English at the conference. It’s exciting to lean about this work that will fill a critical need.