The centerpiece of the study design is the use of random assignment to create a group of teachers exposed to comprehensive teacher induction (treatment) and an equivalent group exposed to the district's usual set of induction services (control). The study design allows us to measure and compare outcomes for these two groups to estimate the impacts of comprehensive induction relative to the services teachers receive from their district's prevailing induction program. We used surveys and school records to measure the background of the study teachers, their receipt of induction services and alternative support services, their attitudes, and the key outcomes of student achievement and teacher mobility.
We selected 17 school districts to participate in the study. District selection was based upon factors such as district size and poverty, whether the district was already implementing a comprehensive teacher induction program, and district willingness to participate in the evaluation. The selected districts, which were spread across 13 states, served low-income students, with every district in the study having more than 50 percent of its students qualifying for the federal School Lunch Program. We then assigned each district to one of the two providers of comprehensive induction, either ETS or NTC, based primarily on district preferences. Nine districts participated in the ETS program; eight districts participated in the NTC program. The preference-based method of assigning districts to providers does not allow for and should not be used to make direct comparisons of one provider to the other.
IES later expanded the treatment to include a second year of services for a subsample of the districts, in effect creating two studies: one for districts that received one year of services (during the 2005-2006 school year), and the other for districts that received two years (during the 2005-2006 and 2006-2007 school years). In the two-year districts, teachers who had been assigned to the treatment group were offered continued services for a second year. The goal of this expansion was to enable the study to address its main research questions separately for one-year and two-year comprehensive induction programs. Policymakers are interested in both models of service delivery because they are both viable policy options for future implementation.
We used convenience sampling to select the districts to receive a second year of the treatment; we selected the districts based upon factors such as whether the mentors who had been trained within the district by ETS or NTC were available for a second year and whether the group of districts selected for a second year would include approximately one-half of the total number of teachers participating in the evaluation. Dividing the sample in this way does not allow for and should not be used to make direct comparisons between the districts that received one year of treatment and districts that received two years of treatment, but instead allows us to investigate the effectiveness of one-year programs separately from that of twoyear programs.
In this Year 2 impact report, unlike the Year 1 impact report (Glazerman et al. 2008), we present findings separately for the set of 10 districts that received one year of treatment ("one-year districts") and the other set of 7 districts that received two years of treatment ("two-year districts"). Both sets of findings are based on data collected through two years of the study. When appropriate, however, we compare outcomes from the first year of the study to outcomes from the second year of the study within the one-year districts and within the two-year districts.
Within each district, a subset of elementary schools participated in the study. As noted above, we randomly assigned these elementary schools to either a treatment group, which was offered comprehensive teacher induction, or a control group, which took part in the district's usual teacher induction program. The final sample size included 418 schools across the 17 districts.
Within each study school, we selected all eligible teachers, defined as beginning teachers who met certain criteria: taught in an elementary grade (K-6); were new to the profession; and were not already receiving induction support from a teacher preparation or certification program. Under these criteria, the 252 schools in the one-year districts contained 561 eligible teachers, and the 166 schools in the two-year districts contained 448 eligible teachers. For the student achievement analysis, we limited the collection of student test score data to teachers meeting another set of eligibility criteria, including teaching a self-contained classroom in a tested grade and subject. This resulted in the collection of reading test scores for 139 teachers and math scores for 123 teachers in the one-year districts, and of reading scores for 96 teachers and math scores for 95 teachers in the two-year districts.3
Eligible teachers in a school were either all exposed or all not exposed to treatment, a method known as cluster random assignment. Cluster random assignment was necessary because varying the types of induction services available in the same school building could result in contamination of the control group. Therefore, we assigned all eligible teachers to treatment or control status based on the school where they were expected to teach at the point of random assignment.