Nonexperimental research is correlational in nature, vitally important for studies in psychology that often cannot directly manipulate variables of critical interest, such as exposure to traumatic experiences, as well as individual characteristics of intelligence, personality, or clinical diagnosis.
Correlational research cannot establish causality, yet these studies provide important evidence for defining a conceptual model describing and explaining how two or more variables are related.
Within a correlational analysis framework, the effects of mediating and moderating variables can be distinguished, and how each contributes to conceptual understanding of a research topic can be articulated.
Cofounding effects of third variables need to be controlled to avoid spurious correlations.
Descriptive statistics summarize and present data in a meaningful way but, importantly, do not allow for testing hypotheses or for generalizing results to the larger population. Inferential statistics allow for testing of hypotheses about a population from data collected from a sample.
