The 43 students in the study are drawn from nine urban, public elementary schools in the Northwest.
Students were in second and third grade, with 25 students in second grade and 18 in third grade. Of the 43 students in the analysis, six were identified as limited English proficient, six were identified as receiving special education, and 32 were classified as Title I. The students in the sample scored between the 10th and 37th percentile on the Woodcock Reading Mastery Test-R/NU.
The intervention was a supplemental phonics-based instruction program delivered by a paraeducator four days a week in 30 minute doses over a period of 15 weeks. All instruction is scripted with activities including modeling, guided practice, and independent practice in letter-sound correspondences, decoding, sight word reading, and spelling. The first 10 weeks consisted of (1) 15 minutes of phonics instruction per session, focusing on single-letter and two-letter spelling patterns for the first five weeks and two-letter spelling patterns for the second five weeks, and (2) 15 minutes of oral passage reading per session, in which students read 50- to 60-word passages featuring words with taught spelling patterns. The oral reading passages increased in length, depending on student mastery level, to 80- to 120-word "Quick Reads" program passages selected to match grade-level. The exact content covered varied by student based on the pretest, as the paraeducator skipped or moved quickly through material in which the student had demonstrated mastery. Every ten lessons a mastery test was given to further tailor instruction to student needs. The final five weeks was solely oral reading practice using a repeated reading instruction method, provided the student had demonstrated "adequate mastery" of covered phonics skills. This section was truncated to 2 weeks for Phase II.
The comparison condition (the late treatment group) is business-as-usual regular classroom instruction. In the second phase of the study the control group also received the treatment. The first posttest is given in March after completion of the early training, but before the start of the late training. At the time of the second posttest, both groups had received the treatment, and so the only contrast of interest is the March posttest.
Support for implementation
Seven of the 11 paraeducators delivering the intervention were new-hires to their school, though only one had no prior experience in early reading instruction. All were women, two were minority, and had an between 12 and 17 (average of 15) years of education. The study researchers provided 3 hours of training in instructional procedures prior to the start of the intervention. Continued training during the intervention included bi-weekly site visit totaling an approximate 60-90 minutes of additional training. Treatment fidelity at each school was assessed by one of five researchers using a five-point rating scale on 21 instructional criteria. Each paraeducator was rated an average of 16 times. The ratings ranges from 2.5 to 4.0, with a mean of 3.8.