Abstract
Background: Disorders in the acquisition of morphological agreement phenomena, e.g. case marking, are characteristic of German children with specific language impairment (SLI). Although the success of individual therapy has already been documented, there are general doubts about the success of grammar facilitation in the classroom.
Aims: To investigate whether children with SLI, who were trained with the new program ‘Context‐Optimization’ during classes, achieved significant improvements in the acquisition of case and whether this therapeutic approach was more effective than the traditional methods used in a control group.
Methods & Procedures: Forty‐five classes of 19 special language schools participated in the study. Within 12 weeks the experimental group, consisting of 63 children, received 12 hours of therapy during their lessons. The therapeutic measures were implemented directly by their class teachers, who are additionally qualified as school speech and language pathologists. The case marking skills of the 63 children constituting the control group were also trained during classes or within individual or group therapy, although their teachers used other methods than Context‐Optimization.
Outcomes & Results: The experimental group participants increased their case‐marking skills significantly, even after a short‐term intervention, and these effects remained stable. It is evident that their improvements were independent of any risk factors such as deficits in the perception and processing of language, problems in gender agreement and bilingualism. Context‐Optimization therapy in the classroom is effective for both accusative and dative, the prominent therapy goal, while traditional methods are only effective for accusative.
Conclusions: A critical analysis of the data reveals two important implications for future interventions: (1) the reduction of language material to ensure proper use of gender during therapy sequences; and (2) the adaptation of therapeutic steps to the learning tempo of the pupils.