Skip Navigation
acfbanner  
ACF
Department of Health and Human Services Administration for Children and Families
          
ACF Home   |   Services   |   Working with ACF   |   Policy/Planning   |   About ACF   |   ACF News   |   HHS Home

  Questions?  |  Privacy  |  Site Index  |  Contact Us  |  Download Reader™Download Reader  |  Print Print      

Office of Planning, Research & Evaluation (OPRE) skip to primary page content
Advanced
Search

Table of Contents | Previous | Next

Recruitment and Random Assignment

To recruit centers for the study, SRC staff sent information about the study, translated into Spanish and Haitian Creole, to the approximately 850 centers that serve subsidized children in the county. SRC staff and staff from the two central subsidy agencies then made follow-up telephone calls and screening calls to determine eligibility and interest. Abt, MDRC and SRC staff held informational meetings for center directors and staff to answer questions and explain the random assignment process. After eligibility was determined, 180 centers were randomly assigned to the four groups, allowing for some replacement of centers that dropped out before knowing their assignment16. Notification of assignment was provided at one large meeting to which center directors and teachers were invited. Four centers decided against participation after being reminded about the random assignment process and were replaced. Directors who reiterated their willingness to participate were asked to review and sign a Memorandum of Understanding that laid out their responsibilities and the responsibilities of the research team, and were then informed of their assignment. No centers refused their assignment. Over the course of two years, seven centers left the study. Five left because the center was closed or sold to an owner who chose not to participate; only two left because the director decided not to continue with the curriculum to which they were assigned. While, in spite of the incentives offered, teachers did leave and were replaced, our concern was about the attrition of centers, since they were the unit of random assignment. Center attrition, as we have seen, was very low and distributed quite evenly across the four groups.

Three classroom-level measures were used to assess the success of random assignment, that is, the equivalence of the four groups: a staff background questionnaire (collected for other purposes by the SRC), the baseline observation measures and the LAP-D assessments of children administered in Fall 2003. There were no significant differences between treatment and control groups. of differences. We therefore concluded that, in terms of measurable aspects of the classrooms, random assignment was successfully carried out. Exhibits A1-A4 in Attachment A provide a detailed comparison of the baseline characteristics of the four groups.



16 There are several reasons why the potential sample quickly diminished. The first and most important reason was lack of interest in participation. Many centers in Miami are either small, for-profit businesses or faith-based entities. While both groups are heavily represented in our sample, many Protestant faith-based centers use IBEKA, a religious literacy curriculum and are committed to it. Many small business owners did not want to participate in a government-sponsored study. Other reasons for reduction in the sample had to do with eligibility – centers were using High Scope or Creative Curriculum or another off-the-shelf curriculum. Because of the timing of our recruitment (late summer), some centers did not have a fully-enrolled 4-year-old classroom and could not assure us that they would. We also used zipcode information combined with a minimum number of subsidized children to eliminate middle-class centers that might have a few subsidized slots for low-income children. We were left with a pool of about 300 centers. We met with all the directors or owners from these centers, and in that process eliminated more of them. Sometimes a director was initially interested but really wanted one of the three curricula or didn’t want to risk being part of the control group. A few centers sent staff to meet with us because they were without a director; in these cases we were concerned that a new director would not honor the agreement to participate. (In the case of the small number of centers that dropped from the study because they were sold, the new owner was not interested in having us or the mentors in the center). We invited the remaining 200 to a final meeting to have the terms of the study explained once more, and 180 came to the meeting. All of these were randomly assigned to a group and people who opted out before they heard their assignment were replaced. We actually assigned 164 centers rather than 162 (B.E.L.L. began with two extra), and no-one who attended the meeting and agreed to random assignment was rejected. (back to footnote 16)

 

Table of Contents | Previous | Next