Nelson L K, Weber-Olsen M
J Speech Hear Disord. 1980 Nov;45(4):549-63. doi: 10.1044/jshd.4504.549.
The Elicited Language Inventory (ELI) (Carrow, 1974b) was administered in standardized fashion and in a modified procedure with contextually supported cues to eight language-delayed and eight normal-speaking children in Brown's (1973) Stage II of morphological development. Additionally, grammatical morpheme use under the two ELI presentation conditions was compared with use of the same morphemes in spontaneous speech for two language-delayed and two normal-speaking subjects. Results of both comparisons favored modified use of the ELI with contextually cued items when sampling children's expressive speech. A significantly greater number of morphemes was imitated and provided in obligatory contexts under modified ELI conditions. Significantly fewer total errors were produced by all subjects when ELI items were presented with contextual cues. Moreover, children's performance on contextually cued ELI items was a significantly better predictor of their grammatical morpheme use in spontaneous speech than performance on the standard version of the ELI. Except for total errors, subject group differences were not significant; however, the older language-delayed children performed consistently better than the stage-matched normal-speaking children on all imitative response measures.
