Second grade good readers (N=46) and fourth grade good and poor readers (N=46; N=48) were subjects in a study that examined linguistic cue usage by good and poor readers in a Southern metropolitan area. A secondary purpose was to compare the standard and phoneme cloze procedures with oral miscue analysis. The subjects were identified as good or poor readers on the basis of their percentile scores on the Comprehensive Test of Basic Skills and a teacher rating scale, and then randomly assigned to one of three test forms: intact for the oral analysis, standard cloze, and phoneme cloze using a fifth-word deletion pattern that included the first phoneme of the deleted word. Responses were scored on their semantic and syntactic appropriateness. Oral reading errors were also analyzed on their graphic similarity to the text word. The two principle findings of this study were that (1) the syntactic and nonsense error linguistic variables discriminated between good and poor fourth grade readers, and (2) the phoneme and standard cloze procedures were not as effective as oral reading miscue in predicting group membership, with fourth grade readers using the syntactic and nonsense variables as the discriminant dimensions. The second grade readers were significantly different from both fourth grade groups on the percentage of synonym errors, but the two fourth grade groups were not significantly different from each other. (HOD)