Factor Stability, Number of Significant Loadings, and Interpretation: Evidence from Three Studies and Suggested Guidelines
Factor stability is an important issue for Q methodological studies that seek to identify viewpoints in a population, since it is possible that the addition of significant loading can change the factor array and the consequent interpretation. This article examines change in the composition of factors as the number of significant loadings changes and the subsequent effects on interpretation. The literature does not give firm guidance on the appropriate number of significant loadings. Data from three case studies are presented showing the extent of changes in the distribution of items in the selected factor arrays that occur as the numbers of subjects and significant loadings increase, In some cases, changes in item position altered factor interpretation. The results show that a number of research dimensions affect factor stability, so there are no uniform rules to guide researchers. The general applicability of the results is discussed with suggestions relevant to the issue of factor stability.