형태소 오류 치료에 대한 사전 학습의 효과

Children are treated for an emergent grammatical form that is used correctly at least 60% during three pre-treatment probe sessions. This form is treated until children generalize it's use an average of 90% or more across 3 probe sessions. Their treatment target is then switched to a grammatical form that is used less than 30% correct across 3 pre-treatment probe sessions.

The treatment procedure uses 24 treatment doses embedded into child-friendly activities (crafts, book reading, play). Each each consists of a clinician eliciting use of a targeted grammatical form, a child's attempt (correct or incorrect) to use that form, and an immediate clinician repetition of the child's attempt (a recast), correcting any ungrammatical elements as needed. Clinicians must also attempt to gain the child's attention before providing the recast. Treatment sessions are approximately 30 minutes in duration.

Children are treated for a grammatical form that is used accurately less than 30% of the time during 3 pre-treatment probe sessions. This form is treated for the entire treatment period

The treatment procedure uses 24 treatment doses embedded into child-friendly activities (crafts, book reading, play). Each each consists of a clinician eliciting use of a targeted grammatical form, a child's attempt (correct or incorrect) to use that form, and an immediate clinician repetition of the child's attempt (a recast), correcting any ungrammatical elements as needed. Clinicians must also attempt to gain the child's attention before providing the recast. Treatment sessions are approximately 30 minutes in duration.

Clinicians elicit ten uses of the trained grammatical form using toys, activities, and vocabulary with-held from treatment sessions. This is done on each of three days at the end of the treatment period. The percent use calculated as (#correct/10)*100 is calculated over three days at the end of treatment, and these values are averaged for each child. Note that only performance on the hard target is considered in the analysis, even for children previously treated on an easier target. This tests the hypothesis that prior treatment of an easy target speeds learning of a hard target, compared to starting treatment with a hard target. The easy target is not compared to the hard target because, by design, this target to a high level of use, whereas learning levels for the hard target are free to vary.

Clinicians elicit ten uses of the trained grammatical form using toys, activities, and vocabulary with-held from treatment sessions and the percent correct use is calculated as (#correct/10)*100.

This analysis tests the hypothesis that the treatment in general (collapsed across treatment arms) leads to gains in the use of treated grammatical forms compared with no treatment of grammatical forms. Measures reporting single-participant treatment effect size are typically reported with a Unit of Measure in treatment effect size d. Single-subject treatment effect size d metric reflects gains in morpheme use from baseline levels to the end of treatment, for each individual child. This is calculated for each of the Target and Control morphemes as "Generalization d = ((end treatment percent correct mean - pretreatment percent correct mean)/ (end treatment standard deviation)). Note that the percent correct means are based on the average percent correct use for three sessions pretreatment and during the 6th week of treatment.