Skip Navigation
Reading First Impact Study: Interim Report
NCEE 2008-4016
April 2008

Back Arrow List of Exhibits         Back to Report Return to Report

Exhibit ES-3.  Estimated Impacts on Reading Comprehension, Instruction, and Percentage of Students Engaged with Print: Spring 2005, Fall 2005, and Spring 2006

Actual Mean
with Reading First
Estimated Mean without Reading First Impact Statistical
of Impact (p-value)
Reading Comprehension
Percent Reading At or Above Grade Level
Grade 1 45.4 42.2 3.15 (0.260)
Grade 2 38.9 38.8 0.12 (0.965)
Grade 3 37.9 40.1 -2.22 (0.383)
Number of minutes of instruction in the five dimensions combined
Grade 1 59.41 50.85 8.56* (0.003)
Grade 2 59.53 47.44 12.09* (<0.001)
Percentage of intervals in five dimensions with Highly Explicit Instruction
Grade 1 29.78 26.13 3.65* (0.023)
Grade 2 31.55 24.57 6.98* (<0.001)
High Quality Student Practice
Grade 1 19.21 18.35 0.86 (0.559)
Grade 2 18.78 15.11 3.67* (0.012)
Percentage of Students Engaged with Print
Grade 1 46.92 42.29 4.63 (0.216)
Grade 2 49.72 58.14 -8.42* (0.030)
The complete Reading First Impact Study (RFIS) sample includes 248 schools from 18 sites (17 school districts and 1 state) located in 13 states. 125 schools are Reading First schools and 123 are non-Reading First schools. For grade 2, one non-RF school could not be included in the analysis because test score data were not available.
Impact estimates are statistically adjusted (e.g., take each school's rating, site-specific funding cut-point, and other covariates into account) to reflect the regression discontinuity design of the study.
Values in the "Actual Mean with Reading First" column are actual, unadjusted values for Reading First schools; values in the "Estimated Mean without Reading First" column represent the best estimates of what would have happened in RF schools absent RF funding and are calculated by subtracting the impact estimates from the RF schools' actual mean values.
A two-tailed test of significance was used, and where applicable, statistically significant findings at the p ≤05 level are indicated by *.
EXHIBIT READS: The observed average percent of first-graders reading at or above grade level with Reading First was 45.4 percentage points. The estimated average percent without Reading First was 42.2 percentage points. The impact of Reading First on the percent of first grade students reading at or above grade level was 3.2 percentage points, which was not statistically significant at the p<.05 level (p=.260).
Sources: RFIS SAT 10 administration in the spring of 2005 and 2006, as well as from state/district education agencies in those sites that already used the SAT 10 for their standardized testing (i.e., FL, KS, MD, OR); RFIS Instructional Practice in Reading Inventory, spring 2005, fall 2005, and spring 2006; RFIS Student Time-on-Task and Engagement with Print, fall 2005 and spring 2006.