In this chapter, the authors discuss the issue of research design in conducting inquiry on college impact and demonstrate the importance of longitudinal pretest-posttest designs in maximizing the internal validity of findings. They begin by discussing the strengths and weaknesses of different types of research design in the college impact literature (for example, cross-sectional, longitudinal with a pretest, and longitudinal without a pretest) and continue by presenting the results of a five-year review of research studies published in four major higher education journals: (1) "Journal of College Student Development"; (2) "Journal of Higher Education"; (3) "Research in Higher Education"; and (4) "Review of Higher Education". Within this review, they present the percentage of college impact studies published in these journals from 2004 to 2008 using the aforementioned research designs. They then offer an empirical example based on data from the Wabash National Study of Liberal Arts Education (the Wabash Study). This example demonstrates the extent to which the effects of an independent variable can be speciously over- or underestimated when a pretest is not used in research. The authors conclude with a discussion of the implications for conducting internally valid studies of college impact. (Contains 1 table.)