Research in psychology and education tend to use large-N group designs that necessitate reporting of mean measures analyzed mostly with null hypothesis statistical testing (NHST), but sometimes with Bayesian, or the estimation approaches in inferential statistics. These approaches all render the average person or student as the the putative "subject" of interest in psychology and education, in addition to the myriad of serious problems, such as widespread replication failures, they have manifested. In reality, however, more often than not, it is the individual person or student who learns, not some nonexistent average person or student. A case is made that a largely ignored alternative to group designs is the Small-N or single-case designs that have a long and productive history in psychology, education, and medicine. They involve studying in-depth only a few subjects at a time under different conditions explored in some detail while observing systematic changes in behavior as those conditions change. In so doing, these designs not only focus on the individual, they reveal functional relationships between his or her behavior and the prevailing environmental conditions. In education, such environments range from the school, the classroom, or teacher (variables) to teaching methods, materials, and/or technology. Undoubtedly, adopting such designs more broadly in psychological and educational research would require a significant shift in how we approach asking questions, collecting data, analyzing and interpreting data, and making research and practice decisions. Not doing so, however, is like repeating the same mistake over and over and expecting a different result. [For the complete proceedings, see ED655360.]