ABSTRACT
This chapter discusses how scientists use empirical results and statistical analyses to develop theoretical conclusions. Although such activities are an integral part of science, it is surprisingly difficult to relate theory to data, and it is often demonstrably done incorrectly. To keep the chapter relevant to the book, I discuss these issues in relation to a recent high-profile publication on autism that concluded that measurements on a visual search task among infants could predict emerging autism symptoms that appeared months later (Gliga et al. 2015). As shown in the chapter, a statistical analysis suggests that readers should be skeptical of the theoretical claims in Gliga et al. (2015). The reasons for this skepticism highlight general statistical issues that apply to investigations of autism and many other scientific topics.
