Testing and verifying nursing theory by confirmatory factor analysis

Abstract

Aim: This paper presents a discussion of the use of confirmatory factor analysis to test nursing theory.

Background: Theory testing is an important phase in nursing theory development. Testing of theory is intended to give more information about concepts and their usefulness in nursing practice. Confirmatory factor analysis is commonly used in instrument development in nursing science studies, but also in theory testing. However, there has been little discussion of its use in theory testing in nursing science research.

Data sources: Multidisciplinary methodological and research publications from 1990 to 2009 were used.

Discussion: The aim of confirmatory factor analysis is to test nursing theory that has already been established, i.e. researchers have an a priori hypothesis based on theoretical knowledge or empirical indications. Analysis is represented as three phases: preparation, model testing and reporting the results. Preparation involves data screening and preliminary analyses. Model testing is divided into model specification, model identification, model estimation, model evaluation and model modification. The results are reported with standardized regression coefficients of the items related to the latent variables, squared multiple correlations (R²) related to error terms and the model's goodness of fit indexes. Implications for nursing. Testing of theory is intended to give more valid information about the concepts and their usefulness in nursing practice.

Conclusion: Confirmatory factor analysis is a good method to test the structure of theory, for example to test the concepts built by concept synthesis or analysis. Tested theories are needed to develop nursing science itself.