Persistent disregard for the inadequacies of null hypothesis significance testing and the viable alternative of observation-oriented modeling

Observation-Oriented Modeling is an approach to data conceptualization and analysis that challenges researchers to seek scientific, rather than statistical, inferences from their research. As such, it represents a radical alternative to Null Hypothesis Significance Testing and other methods currently employed by psychologists. In an effort to neutralize the reticence among practitioners of long-established methods to explore new techniques and ways of thinking, it is shown in this chapter how concepts in OOM can be used to expand upon or supplant contemporary conceptualizations of effect size, causality, measurement, and inference. By reviewing previous published studies, we show how OOM provides the tools to (1) focus on persons rather than aggregate effect sizes, (2) use all four of Aristotle’s species of cause, (3) avoid unwarranted measurement and statistical assumptions, and (4) seek explanatory rather than statistical inferences. Finally, we demonstrate how these conceptual and analytical tools are used in a novel analysis of a recently published paper on inter-identity amnesia in persons diagnosed with dissociative identity disorder