Repeating a sentence in one's native language is not a passive phonological copy of the input sentence. Rather, it involves comprehension and production, and therefore, difficulties in the comprehension and production of syntactic structure are manifested in difficulties to repeat the structure (Fattal et al., 2011; Friedmann, 2007; Lust, Flynn, & Foley, 1996). Specifically, when we compare sentences that are similar (same length and words) and differ only in the relevant syntactic feature, if a participant succeeds in repeating one structure but fails in the other structure, this might indicate a specific difficulty with the tested structure.
The PETEL sentence repetition test (Fattal et al., 2011; Friedmann, 2000; Friedmann & Novogrodsky, 2004; Friedmann & Szterman, 2011) includes subject and object which-questions, object relatives, object topicalization, sentences with verb movement, sentences with A-movement, embedded sentences with a sentential complement, and simple sentences without Wh-movement.
The sentences of the various types were presented in a random order. The test includes 70 sentences, all with four words (accusative markers, embedding markers, and prepositions are counted with the word following them). All the sentences derived by Wh-movement are semantically reversible. The simple sentences include either a transitive or an intransitive verb, and the verbs in the embedded clauses in the embedding condition are intransitive. In each of the sentences that include a transitive verb, the two NPs are of the same gender, to preclude a grammatical gender cue on the subject and object.
In the analysis of errors, structural errors are scored separately from lexical errors that do not affect the structure of the sentence. Structural errors are errors that change the linear order of elements in the sentences, thus changing its structure, the assignment of thematic roles or making it ungrammatical. Omissions and substitutions of function words were scored as structural errors. Lexical errors are errors in which the participant substitutes content words with semantically related words that do not appear in the sentence (e.g. substituting the noun cop with soldier, or the verb said with told) or omission, substitution or addition of adjectives, without changing the sentence structure.