Giusto, M., & Ehri, L. C. (2019). Effectiveness of a partial read-aloud test accommodation to assess reading comprehension in students with a reading disability . Journal of Learning Disabilities , 52 (3), 259–270. https://doi.org/10.1177/0022219418789377

This journal article reports on the study also reported by a dissertation by Giusto in 2015.

The accommodations of oral delivery—live/in person, as well as prompting, were investigated. Three conditions were examined: (a) prompting (pacing) only, (b) prompting (pacing) and oral delivery, live/in person—called "partial read-aloud with pacing—and (c) no accommodations provided. Partial oral delivery of a reading test is provided when a test administrator or proctor presents test instructions and item stems—but not reading passages—aloud; test-takers read the text passages silently for themselves.

[This journal article reports on the study, with the same participants, that has been reported in Giusto (2015).] A total of 82 students in grade 3 from two schools and a summer program in Queens, NY (U.S.) participated. Some participants (n=28) were selected due to having average reading comprehension skills and poor decoding skills; the remaining participants (n=54) served as a comparison group because they had average reading comprehension and average decoding skills. The hypothesis being examined was that the partial read-aloud with pacing (PRAP) accommodation differentially benefits students with adequate reading comprehension skills but poor decoding skills. Demographic data including age, gender, race/ethnicity, home language (all were English proficient), and socioeconomic status, were also reported; sample characteristics were consistent with those of the three participant sites.

Reading comprehension achievement was measured by the norm-referenced Gates MacGinitie Reading Comprehension Test, Fourth Edition, Form S, Grade 3 (GMRT; MacGinitie, MacGinitie, Maria, & Dreyer, 2002). Participants' observed test response times were recorded in number of words read silently and words read aloud. For screening/identification, several norm-referenced test scores were gathered: Word Identification Subtest and Word Attack Subtest on Woodcock Reading Mastery Test-Revised (WRMT-R; Woodcock, 1987); receptive vocabulary on the Peabody Picture Vocabulary Test, Fourth Edition (PPVT-4; Dunn & Dunn, 2007); and Understanding Spoken Paragraphs Subtest and Understanding Concepts and Spoken Directions Subtest on the Clinical Evaluation of Language Fundamentals, Fourth Edition (CELF-4; Semel et al., 2003).

Students with reading disability pertaining to decoding scored significantly higher on reading comprehension with the partial read-aloud with pacing accommodation than with either the pacing-only support or without accommodations. Further, the pacing-only and non-accommodated conditions did not yield significantly different scores from one another. The students without decoding difficulties (i.e., without disabilities) did not score significantly differently across the three testing conditions. Additionally, the varying linguistic backgrounds among students in both participant groups did not demonstrate an effect on the mean scoring patterns; that is, linguistic background was not a factor in the differences reported previously. Correlations among the reading comprehension test and the screening tests indicated that "the paragraph comprehension task [in the pre-test] may be a better way to measure language comprehension as it explains variance in reading comprehension of text" (p. 65). The researcher also reported anecdotal observations of participants while completing the reading comprehension test under different conditions. Students in both participant groups showed no attention (possibly unfocused) or seemed frustrated with waiting to move on to another test section, in the pacing-only condition. Finally, the researcher indicated that the partial read-aloud accommodation with pacing support addressed concerns indicated in other research and practice about invalidating the reading comprehension construct. In other words, not reading the text passages yet providing read-aloud for the remaining parts of the test was demonstrated to differentially benefit students with decoding skills difficulties. Limitations of the study were reported, and future research directions were suggested.