The purpose of this study was to obtain a measure of the reliability of the informal reading inventory (IRI) when administered to selected sixth grade students. Houghton Mifflin's Pupil Placement Tests were administered to 36 sixth grade girls. The raw scores of the students on the pupil Placement Tests were divided into two equivalent halves, and a relationship was obtained by employing the split-half method of correlation. From the reliability of the divided tests, the self-correlation of the whole test was estimated through the use of the Spearman-Brown prophecy formula. Fisher's Zr transformation was used to obtain confidence intervals for r at the .05 level. Null hypotheses were tested for significance on each subtest according to their approximation to the .90 correlation coefficient criterion. The conclusions derived included: (1) the IRI is a statistically reliable instrument for girls at the sixth grade level, (2) subtest grade levels do not appear to be reliable for testing; (3) selective grade levels are more dependable than others for reliability predictions; and (4) the average difference between the instructional level of the IRI and the Metropolitan Achievement Test reading level was approximately three years. (Author/WR)