Programs that Work
Partners in Reading
Children Succeeding in School
Students performing at grade level or meeting state curriculum standards
Age of Child
Early Childhood (0-8)
Type of Setting
Type of Service
Type of Outcome Addressed
Cognitive Development / School Performance
Evidence Level (What does this mean?)
The Partners in Reading program (PIR) is a school-based tutoring program for beginner readers. PIR is designed to increase the number of books that students read independently, to enhance the difficulty level of the books that students read, and to improve students' ability to recognize and pronounce words correctly. Tutors help students read and reread instructional texts that are appropriate to a student's reading level, provide students with immediate and frequent feedback, and employ activities to encourage students' use of sophisticated word recognition strategies and to motivate students to learn. PIR was implemented in a single school that requested help in meeting the needs of its struggling readers, because the number of these students exceeded the resources of the school's Reading Recovery (RR) program (the regular tutoring program).
PIR targets struggling first-grade readers in schools with limited financial and personnel resources.
Miller (2003) evaluated the PIR program by studying first graders from 15 classrooms in a Title I elementary school in the coastal Carolinas. Approximately 65 percent of the schoolís students were racial minorities, primarily African-American, and 75 percent received free or reduced-price lunches. The sample was assembled in two groups over two consecutive years, and consisted of 54 PIR students, 62 RR students, and 58 control students. During the first year (Group 1 of two groups), 12 first-grade classrooms containing 263 students took part in the study. Nineteen first-grade students out of this first cohort were selected for tutoring. During the second year (Group 2), an additional three first-grade classrooms were added for a total participation of 15 classrooms consisting of 334 students. Thirty-five students out of the second cohort were selected for tutoring. Students in the treatment groups were selected based on their scores on tests of word recognition and spelling from the Howard Street Training Manual (Morris, 1992). Scores on these two assessment measures were compared with teachersí rankings of student reading ability. Students whose scores were at the bottom third of the measures as well as being in the bottom third of the teachersí rankings were selected for tutoring.
The selection of students for PIR occurred after the RR students were selected (because RR teachers were able to provide tutoring before the PIR tutors were ready to start). The control group included students who did not receive any special assistance from PIR, RR, or special education, and whose pretest scores on the word recognition and spelling measures were within one standard deviation of the PIR and RR studentsí average scores. (Subsequent analysis indicated that the three groups were statistically equivalent in word recognition and spelling at the start of the school year.)
From the initial sample, 18 of the 54 PIR students (33 percent), 15 of the 62 RR students (24 percent), and 29 of the 58 control students (50 percent) left the school prior to testing at the end of second grade. No attrition analysis was conducted to determine whether the remaining students in the three groups were equivalent at the outset of the study. Due to the high number of students who left the study sample, an analysis of data on Metropolitan Achievement Test scores for the combined group was done at the end of second grade.
Key Evaluation Findings
At the end of first grade, students were compared on their performance on word recognition and developmental spelling tests from the Howard Street Training Manual. At the end of second grade, the studentsí performance was compared using the Metropolitan Achievement Test for reading and language arts, which includes subtests on language, word recognition, vocabulary, and comprehension.
Word Recognition and Spelling at the End of First Grade
- At yearís end, students in PIR (19 students) and RR (30 students) were statistically equivalent and significantly superior to the control group (29 students) on word recognition scores.
- Similarly, PIR and RR students were statistically equivalent and significantly superior to control group students on scores of developmental spelling.
- At yearís end, students in PIR (35 students) and RR (32 students) were equivalent and significantly superior to control group students (29 students) on the measure of word recognition.
- PIR and RR students scored similarly on the developmental spelling test, with both groups significantly outscoring the control group.
- Significant differences were found among the three groups on the word recognition subtest, with PIR and RR students scoring higher than control students.
- Differences were also found among the groups on the comprehension subtest, with PIR students significantly outperforming control students, but no differences were found between PIR and RR students.
- No significant differences were found among the groups on the vocabulary or language subtests.
Public and private elementary schools
The school discussed in the Miller (2003) study utilized professional development funds to implemented PIR training.
Each tutor was assigned to multiple students and completed a progress chart for each student every six weeks. The charts included information on the number of tutoring sessions, the number of books a student had read, the reading difficulty level of each book, word-sort activities, and notes about a studentís successes or difficulties. Every six weeks, classroom teachers received a note from the tutor in the form of a greeting card, which stated the number of books each student had read and the associated reading difficulty levels.
Students received at least four 40-minute tutoring sessions each week. Each session consisted of two phases: First, a student read a book that was familiar from either a previous session or from a nightly homework selection. Second, tutors allowed the student to select one book from among several new books that were at a reading level slightly above the studentís instructional level. After selecting a book, the student would choose to read it independently, read it along with the tutor, or have the tutor read it to him or her. Students were encouraged to read the books independently.
The classroom tutors attended two half-day workshops prior to the start of the school year, where they learned how to organize a typical tutoring lesson.
Issues to Consider
This program received a "promising" rating. For both Group 1 and Group 2, PIR students significantly outscored control group students on measures of word recognition and spelling at the end of first grade. Results from the Metropolitan Achievement Test at the end of second grade indicated that PIR students had outscored control students on word recognition and reading comprehension.
An area of concern regarding the results of the Miller study is the high rate of student attrition prior to the end of second grade. Given that 33 percent of PIR students, 24 percent of RR students, and 50 percent of control students left the study prior to testing on the Metropolitan Achievement Test, it is possible that the remaining students were no longer directly comparable. In other words, because an analysis was not conducted to determine the pre-study comparability of these three groups, it may be that they differed significantly from each other in baseline characteristics due to the differential in their dropout rates.
It should be noted that the sample sizes in the Miller study were quite small, and that the program has been evaluated in only a single site. Therefore, the Miller study results may not be applicable to schools having a student body with markedly different socioeconomic characteristics.
Results from this study suggest that the Partners in Reading program may be an effective way for financially challenged schools to help struggling first-grade readers improve their language skills. Although no significant differences in achievement scores were found between students in the PIR program and those in the schoolís regular Reading Recovery tutoring program, the PIR program is much less expensive to implement.
The coastal Carolinas
Dr. Samuel D. Miller
P.O. Box 26170
School of Education
University of North Carolina
Greensboro, NC 27402-6170
Ph: (336) 334-3445
Fax: (336) 334-4120
No resources were available as of fall 2004.
Miller, Samuel D., "Partners-in-Reading: Using Classroom Assistants to Provide Tutorial Assistance to Struggling First-Grade Readers,"
Journal of Education for Students Placed at Risk,
Vol. 8, No. 3, 2003, pp. 333-349.
Morris, D., Case Studies in Teaching Beginning Readers: The Howard Street Tutoring Manual, Boone, N.C.: Fieldstrip, 1992.