| description abstract | Research criteria and methods are presented for performing and evaluating the growing discipline of solar-weather studies. A clear distinction among preliminary, exploratory, and confirmatory studies is presented, and it is shown that this distinction affects all dimensions and findings of a study. Particular attention is given to conceptual models, variables and data, sample designs, analyses, and formal ampliative inference issues under exploratory and confirmatory conditions. The utility of the proposed criteria and methods are illustrated by evaluating two recent solar-weather claims of strong association between solar flux and stratospheric temperature at the North Pole when partitioned by the phase of the quasi-biennial oscillation. The results of the evaluation indicate that these claims are only exploratory ones and are still in need of confirming evidence. Last, a reanalysis of the solar flux and temperature relation, using some of the proposed methods, indicates that the association actually changes with time, and thus, there appears to be other, presently unknown, factors that affect this potential solar-weather relation. | |
