Appendix K - Details for Question B2

Appendix K- Details for Question B1

Recruiting

This study is being proposed as part of the activities of a REL-Southeast sponsored research alliance, which includes education policymakers from Georgia and South Carolina. Once the study has received OMB clearance, districts which have expressed interest will be contacted by state alliance members and the study team. Districts will be provided with the approved proposal, MOUs (see Appendix D), the FAQ (see Appendix E), and consent forms (see Appendix A, F & G). In-person meetings will be held with LEAs. The plan is to begin contacting districts in January of 2014 (if OMB clearance is complete, or to begin as soon as it is completed) and to have all signed MOUs in place by June 2014.

Eight to ten districts will be recruited across the two states: 4-5 from each state. One district comprised primarily of rural schools will be recruited from each state, and a mix of urban school districts in small/medium and large town, to maximize the type of schools sampled within each state. The number of elementary schools in each state is provided in the urbanicity table below (Table F.1). Alliance members in both states will provide contacts for districts that would likely be interested in participating in the study. Districts that show interest in participating in the study and have at least 6 elementary schools with at least two 4^th grade teachers each will be selected for participation. It is expected that recruited districts will vary in size, with an average of approximately 10 participating elementary schools in 8-10 districts, for a total of 84 schools¹. A minimum number of schools was established to eliminate the need to recruit and manage relationships with more than six districts. Because rural districts are required in the plan, it may be necessary to include districts with as low as six eligible elementary schools. It is anticipated that some of the urban and county or suburban districts will have more than six eligible elementary schools. Limiting the study to 8-10 districts also serves to control the number of different DMI training groups (of up to 20 teachers each) that would need to be formed. A number of smaller districts scattered over a larger area would likely necessitate more DMI groups and more trainers.

Once the district leadership has committed to participate in the study, school leaders within the district (from schools with at least two 4^th grade teachers/classrooms) will be invited to a face-to-face informational meeting. If there are more schools interested than are needed, then a lottery will be used to select the participating schools. Subsequently, all schools which sign an MOU will be included in the study. The study, school MOUs, and their roles given possible assignment to either the DMI or control conditions will be carefully explained. School leaders will be asked to commit prior to random assignment only if they are willing to accept the 50% probability of receiving the DMI professional development. The study team has successfully used this approach in the past to recruit for two large-scale RCTs with 62 and 76 schools, respectively (Gersten, Dimino, Jayanthi, & Haymond, forthcoming; Rolfhus et al., 2012). Both studies exhibited very low school- and teacher-level attrition (<5%).

Once individual school leaders have signed an MOU, information regarding the study, along with informed consent forms will be shared with individual teachers in those schools. Informed consent from teachers will be acquired prior to pairing and random assignment of schools.

Table F.1 Summary of urban-centric locale designation of schools

State	Urban-Centric Locale Designation	Total Number of Schools
Georgia	City	707
	Town	136
	Rural	480
South Carolina	City	264
	Town	107
	Rural	295

References

Gersten, R., Dimino, J., Jayanthi, M., & Haymond, K. (Manuscript in preparation). The Impact of Participation in Teacher Study Groups: A Replication in First Grade Classrooms.

Rolfhus, E., Gersten, R., Clarke, B., Decker, L., Wilkins, C., and Dimino, J. (2012). An evaluation of Number Rockets: A Tier 2 intervention for grade 1 students at risk for difficulties in mathematics. (NCEE 2012-4007). Washington, DC: National Center for Education Evaluation and Regional Assistance, Institute of Education Sciences, U.S. Department of Education.

1 We need a minimum of 80 districts for adequate statistical power so we will recruit 84 districts to account for possible school-level attrition. We expect that the final number of schools after attrition will be 82.

K - 2