Beyond integration: a coherence and transparency framework for mixed methods research in the era of generative AI

Mixed methods research (MMR) has developed sophisticated approaches for combining qualitative and quantitative inquiry through design, integration, joint displays, and meta-inference. However, integration alone does not ensure methodological quality. A mixed methods study may demonstrate identifiable points of integration while retaining inconsistencies between its philosophical assumptions, theoretical framework, research questions, design, methods, analyses, and resulting meta-inferences. At the same time, the expanding use of generative artificial intelligence (GenAI) introduces new analytical intermediaries and potentially new points of integration, making methodological transparency, provenance, and human epistemic accountability increasingly consequential. In this conceptual analysis, we examine the relationships among methodological pluralism, integration, fit, alignment, coherence, ecological validity, and transparency and propose the Mixed Methods Coherence and Transparency Framework (MMCTF). The framework conceptualizes mixed methods quality across five interconnected domains: epistemic coherence, theoretical and design coherence, integrative coherence, inferential and ecological coherence, and methodological transparency. We distinguish coherence from both alignment and convergence, arguing that methodologically productive mixed methods inquiry may generate confirmation, expansion, or theoretically informative discordance. We further conceptualize GenAI not merely as an analytical tool but as a potential mediator of mixed methods integration whose influence must remain traceable and subject to human validation. The MMCTF is proposed as a conceptual architecture for supporting researchers, reviewers, and editors in designing and evaluating methodologically coherent and transparent mixed methods research, particularly as human and AI-supported analytical processes become increasingly intertwined in psychological research.