This chapter highlights design and analysis considerations relevant to many experiments and quasi-experiments in developmental psychology. Here are the first two paragraphs:

“A major function of data analysis is to facilitate inference about psychological and behavioral process. Analytic tools help us learn about underlying processes. But, textbooks typically do not highlight this key function of data analysis, instead focusing on the multitude of problems one must address when analyzing data such as correcting the p-value for multiple tests or using the right error term in an F test or discussing what goodness-of-fit measures to report in a paper. Textbooks tend to ignore the nonstatistical issues that can interfere with making solid inferences from data, even when one computes a p-value correctly.”

“In this chapter we present data analysis in a different light. We highlight the role of data analysis in making solid inferences about psychological, behavioral, and developmental processes from experimental and quasi-experimental designs. As researchers, we conduct studies because we want to learn something we did not already know (though sometimes studies are conducted for other reasons such as gathering information to make predictions). The goal is to ensure that the design and data analytic procedures we use do not interfere with our ability to learn about underlying developmental processes. The chapter highlights how design decisions and data analytic procedures work together in service of illuminating our understanding of development, which sometimes goes beyond the practice of computing p-values in a test of significance.”

Gonzalez, R., Yu, T., & Volling, B. (2011). Analysis of experimental and quasi- experimental data: Pinpointing explanations. In Handbook of Developmental Research Methods, B. Laursen, T. Little, & N. Card (Eds). Guilford Publications: New York, 247-264.

Download the chapter

Publisher’s book page