Use of factor analysis in Journal of Advanced Nursing: literature review

Abstract

This paper reports a review analysing the use of factor analysis in papers in Journal of Advanced Nursing. Factor analysis is a multivariate statistical method for reducing large numbers of variables to fewer underlying dimensions. There are several methods of factor analysis with principal components analysis being the most commonly applied. Factor analysis has been used by researchers in nursing for many years but the standards for use and reporting are variable. Papers using factor analysis in Journal of Advanced Nursing were retrieved from 1982 to the end of 2004. The search term 'factor analysis' was used in the CINAHL database and applied specifically to Journal of Advanced Nursing in December 2004. Retrieved papers were included in the review if they came from Journal of Advanced Nursing and used factor analysis as part of the method of the reported study. One hundred and twenty-four papers were retrieved as a result of the initial search criteria of which 116 were from Journal of Advanced Nursing. Screening of papers for the use of factor analysis left 100 papers for review. Principal components analysis was the most commonly used method of factor analysis; Eigenvalues greater than one was the most commonly applied criterion for selecting the number of factors followed by orthogonal rotation to achieve simple structure. The majority of papers did not report the whole factor solution and there were papers that did not specify anything beyond the fact that they carried out factor analysis. Confirmatory factor analysis was rarely used and exploratory methods other than principal components analysis were also rarely used. Factor analysis is quite commonly used in nursing research reported in Journal of Advanced Nursing. While some papers are exemplary there is room for improvement in the reporting of all aspects of factor analysis.