PON LEARNING that the National Reading Panel recommended certain researchbased practices, 1 a school district hired an expert to tell its teachers about those practices in a half-day professional development session. It then forced the teachers to comply by basing their year-end evaluations on whether they used the recommended practices. Does this sound familiar? It should, because this scenario is not unusual. School districts arrange for these inservice sessions because they assume that test scores will rise if teachers use researchbased practices and because influential educators such as Edward Kame'enui promote this use of professional development as "the logic model." 2 However, we argue that this model may send a counterproductive message to teachers. FIGURING OUT WHAT COUNTS Professional development is like any other form of teaching, in that learners always figure out what the teacher really wants. For instance, elementary students figure out what counts in classrooms by noting what teachers value. 3 If Compliance or Adaptation: WHAT IS THE REAL MESSAGE ABOUT RESEARCH-BASED PRACTICES? It's not whether teachers should use research-based practices, Mr. Duffy and Ms. Kear point out. Of course they should. The real question is whether teachers should feel free to adapt the practices they learn rather than comply strictly with expert guidance.

Summary Background Obesity disproportionately impacts children who are Black and Hispanic, those who live in rural communities and those who have low income. Predisposition to obesity early in life is profoundly impacted by feeding habits during the preschool years. The early care and education (ECE) setting impacts children's health by providing daily meals. Objective The goal of this study was to identify whether or not health disparities in foods and beverages served in ECE programs in Georgia exist based on socioeconomic, demographic and geographic variables. Methods A random sample was drawn from 3054 ECE programs across the state of Georgia. The likelihood of serving specific foods and beverages in ECE programs in the previous day was measured. Percentages and frequencies, logistic regressions, Spearman's rho and Odds ratio tests determined outcomes. Results A total of 974 surveys were returned. Data were stratified based on the income level of the participant families, race of enrolled children and geographic location of the ECE program. Disparities existed between programs based on race of enrolled children and geographic location. For example, although the odds of providing sweets increased by 0.6% as the percentage of Black children enrolled increased, the provision of healthier foods, such as the odds of providing fruits (P = .001), vegetables (P = .001) and protein (P = .001) also increased. However, after results were adjusted for covariates findings did not remain significant. Conclusions Future research focused on evaluating the foods and beverages provided in ECE programs and the relationship of how income, race and location are related may provide further understanding about the disproportionate childhood obesity rates in America.