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

Average Impacts Across All Sites

Exhibit ES.3 reports average impacts for school years 2004-05 and 2005-06.4 All impact estimates are regression-adjusted to control for a linear specification of the rating variable each site used to select its Reading First schools as well as selected teacher and/or student background characteristics used in the analysis. The impacts have been estimated using multi-level models to account for the clustering of students within classrooms, classrooms within schools, and schools within sites. In Exhibit ES.3, 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 Reading First schools absent Reading First funding and are calculated by subtracting the impact estimates from the Reading First schools' actual mean values. Impacts were estimated for each study site and averaged across sites in proportion to their number of Reading First schools in the sample. Average impacts thus represent the average study school. On average:

  • Reading First did not improve students' reading comprehension. The program did not increase the percentages of students in grades one, two, or three, whose reading comprehension scores were at or above grade level. In each of the three grades, fewer than half of the students in the Reading First schools were reading at or above grade level.


  • Reading First increased total class time spent on the five essential components of reading instruction promoted by the program. The program increased average class time spent on the five essential components of reading instruction by 8.56 minutes per daily reading block in grade one, and by 12.09 minutes per daily reading block in grade two. This implies a weekly increase of three quarters of an hour for grade one and one hour for grade two.


  • Reading First increased highly explicit instruction in grades one and two and increased high quality student practice in grade two. The program increased the percentage of class observational intervals spent on the five dimensions of reading instruction that involve highly explicit instruction by 3.65 percentage points in grade one and by 6.98 percentage points in grade two. The program also increased the percentage of class observational intervals spent on the five dimensions of reading instruction that involve high quality student practice by 3.67 percentage points in grade two. There was virtually no observed change in grade one.


  • Reading First had mixed effects on student engagement with print. The program reduced the percentage of students engaged with print by a statistically significant 8.42 percentage points in grade two. The impact on student engagement with print in grade one (4.63 percentage points) was not statistically significant.
Exhibit ES-3. Estimated Impacts on Reading Comprehension, Instruction, and Percentage of Students Engaged with Print: Spring 2005, Fall 2005, and Spring 2006

Top

4 Exhibit ES.3 and all other tables indicate whether findings are based on the full study sample or specific subgroups. Where appropriate, each exhibit also includes an "Exhibit Reads" section that walks readers through the exhibit by highlighting the first row or line of information presented.