Addendum to OMB Supporting Statement. Feb 2011

Addendum

1.  What was the specific problem with access to DoDEA schools?

DoDEA simply would not allow our research team access to the schools for data collection. Therefore, non-research professionals such as teachers and/or other school personnel were left with the task of administering the surveys. This typically yields substandard results.

2.  Provide information about similar work in the past with public schools. Provide assurance that it’s a reasonable assumption that research staff will be given authorization to conduct data collection.

The research team has conducted SOS evaluations with over 30 schools in several parts of the country in the past 5 years. In no case has access to students for data collection purposes been denied. We are very familiar with the processes involved in submitting IRB proposals to participating schools to allow our research personnel to collect data on site. We do not anticipate any problems with this aspect of the study.

3.  Provide more information about the sampling design. Will treatment and control schools be matched? Will there be any geographic clustering in the sampling in order to make data collection more convenient? Will evaluation schools be sampled from schools that already implement the SOS program? Give more background in how sample will be collected.

To ensure that the experimental groups do not have widely varying student populations, schools agreeing to participate will be matched based on student demographics and then randomly assigned to treatment and control groups. We do hope to achieve some geographic clustering to promote efficiency of data collection; fortunately, there is natural geographic clustering to high impact schools in the US that will assist in this effort. To enable the largest number of schools possible to participate in the evaluation, we will not exclude those already using the SOS program but in those cases will confine the evaluation to those students who have not been previously exposed to the program, e.g., the freshmen class in traditional high schools. The analytic strategy described in our OMB Supporting Statement is well-suited to this sampling design, and will allow us to adjust for differences in the composition of the treatment and control groups that may emerge despite the use of matching in the assignment process.

4.  Send previous SOS publications.

Previous publications and reports are included with the submission.

5. Was fidelity to the intervention considered?

Yes, fidelity to the training was considered. However, the training materials and implementation protocol are very straight forward. In pilot studies, trainers had no problem with implementation. Furthermore, the costs of additional staff to conduct this type of evaluation would be prohibitive and significantly reduce the number of students who could learn these skills were these funds redirected.