Recently, while working on a coauthored article on the psychometrics of a specific instrument, a question arose from my research team: Should we always include a theory in our research? Probably, the question came up because it was difficult for us to find a theoretical basis for a psychometric test (other than classic test theory and item response theory). Then, I remembered the struggles. The theory selection process for research projects, especially when research design and theory cannot be easily linked, challenges novice and experienced researchers alike.

I am tempted to support mandatory use of theory in research. First, theories provide systematic and comprehensive frameworks and/or directions for research, and the use of theory in research advances our science through theoretical deduction and induction. Furthermore, theory helps to ensure rigor: Criteria for scientific review relate to the theory that supports a study (e.g., major concepts, questions, study design, data analysis). Finally, the use of theory in a research proposal plays an important role in research funding decisions, and that is a huge incentive to use theory.

However, there is no one answer to the question about mandatory usage of theory in research because the answer depends on the situation (e.g., purpose of the research, topic, population, designs, and researchers' expertise). For instance, many studies do not need theory to support concepts/variables or research questions. The grounded theory method (Strauss & Corbin, 1990) prescribes researchers to start without a theory because the goal is theory construction through inductive reasoning. Phenomenological researchers also begin their research questions without a theory because they search for meaning in little known lived experience or they search for a new perspective on a well-known phenomenon (Cohen, Kahn, & Steeves, 2000).

We have benefitted from nurses' efforts to link theory to research and the results of these efforts-a variety of theories. Indeed, thought-about nursing theory has evolved along with our discipline throughout nursing history. Grand theories yielded to midrange theories, and situation-specific theories followed. Currently, theories from diverse perspectives coexist. Without recognizing it, we enjoy luxury in theories and take them for granted, which we should not. This may be the right time to reevaluate the use of theories in our research. What would be the criteria that we need to use in our selection of a theory for our research? What kinds of validation/evaluation/test process should we take before using a specific theory in our research? In what cases should we exempt the use of a theory in research, and how can we link theory to research in those cases? What guidance should we provide to our junior researchers for theoretical deduction and induction in their research? We may have answers for all these questions already, but we need to make collective efforts to clarify them for ourselves and the next generation of researchers.

Eun-OkIm PhD, MPH, RN, CNS, FAAN

Professor and Marjorie O. Rendell

Endowed Professor

School of Nursing

The University of Pennsylvania

[email protected]

References