—The analyses of Drs Baumeister and Bacharach as reported in Intelligence1 are flawed by their use of correlational analyses to draw inferences regarding causality and by selectively analyzing smaller, post-hoc subgroups to support their conclusions. Their current criticism of the 3-year, 5-year, and 8-year data is based on a conclusion derived from a correlational analysis of a selected cohort of the combined intervention and follow-up groups drawn from data deposited with NAPS. This approach violates the design of the study as a multisite randomized trial, ignores the specific characteristics of each assessment point, and biases their sample by including only those children seen at all time points. At each age, we followed the intention-to-treat principle used in randomized controlled trials to minimize bias: every available randomized child was analyzed without any post-hoc exclusions. As seen in Table 2 in our article, the sample seen at 8 years was nearly
McCarton CM. The Infant Health and Development Program: Results at 8 Years-Reply. JAMA. 1997;277(16):1278–1279. doi:10.1001/jama.1997.03540400028021
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: