Table 1.

Comparison of Epistemological Assumptions of Yin, Stake, Merriam, and Reilly (Inspired by Patton¹⁵)

Methodologist	Epistemological Position	Reality Assumptions	Knowledge Assumptions	Methodological Assumptions
Robert K. Yin, PhD	Postpositivism	Reality exists, but we cannot know it perfectly. Reality is governed by natural, causal, or other laws, but they cannot be absolutely known.	Knowledge is the result of the combination of experimentations leading to a closer approximation of actual mechanisms.	All methods have their limits, so both quantitative and qualitative methods are needed to generate empirical evidence and test plausible rival hypotheses.
Robert E. Stake, PhD	Constructivism	Multiple socially constructed realities are built through interactions with others and human lived experiences.	Human experience can be known, and each human knows his/her own experience of the real.	Qualitative methods need to capture the diversity of realities through a deep understanding of peoples’ perspectives and experiences regarding a specific situation or phenomenon.
Sharan B. Merriam, MEd, EdD	Constructivist Pragmatism	Reality is constructed through meanings and understandings developed socially and experientially	Human experience can be known, and each human knows his/her own experience of the real. The finality of knowledge is to address concrete problems and provide answers or direction to progress. The truth will be what works in practice.	All methods are considered. What is important is to guide the research process by the principles of ethical and scientific rigor, and to clarify hypotheses.
Rosemary C. Reilly, PhD, Med	Transformative Paradigm	Reality is the product of critical interpretation that aims to transform the social world in order to emancipate marginalized groups or communities.	The knowledge is ideologically oriented, and the focus is empowering in its goal.	Qualitative method is privileged. Participants to the research are experts into the underlying causes of the issues within their social world.