SOCIOKAIROS Companion Paper: Computational Heuristics for Reflexive Sociology

© Víctor Hugo Pérez Gallo – University of Zaragoza – [email protected] SOCIOKAIROS: A Heuristic Computational Assistant for the Formulation of Sociological Research Problems University of Zaragoza – SOCIOKAIROS Research Suite v15 (2025) Dr. Víctor Hugo Pérez Gallo Department of Sociology, University of Zaragoza, Spain Email: [email protected] License: Creative Commons Attribution 4.0 International (CC BY 4.0) Abstract This paper presents SOCIOKAIROS, a heuristic computational assistant designed to support the formulation and operationalization of sociological research problems. Unlike traditional CAQDAS systems focused on empirical data coding, SOCIOKAIROS acts as a reflexive epistemic mediator that guides the user through conceptual reasoning—from identifying dependent and independent variables to defining indicators and hypotheses in a context-sensitive manner. The system integrates theoretical frameworks from the sociology of knowledge (Berger & Luckmann, 1966; Bourdieu, 1979; Goffman, 1974) and critical epistemology (Gramsci, 1971; Feenberg, 2017), encouraging computational reflexivity that transforms technology into a tool for thought rather than mere data processing. Experimental implementations across Spain, Cuba, and Brazil demonstrate its pedagogical impact in strengthening scientific reasoning and the articulation between theory and method. Resumen (Español) Este trabajo presenta SOCIOKAIROS, un asistente heurístico computacional orientado a la formulación y operacionalización de problemas científicos en sociología. A diferencia de los programas CAQDAS, centrados en el análisis empírico, SOCIOKAIROS funciona como un mediador epistemológico reflexivo que guía al investigador en el razonamiento conceptual —desde la identificación de variables dependientes e independientes hasta la construcción de indicadores e hipótesis en función del contexto analizado—. El sistema integra fundamentos de la sociología del conocimiento (Berger y Luckmann, 1966; Bourdieu, 1979; Goffman, 1974) y de la epistemología crítica (Gramsci, 1971; Feenberg, 2017), promoviendo una reflexividad computacional que convierte la tecnología en un instrumento de pensamiento y no solo de procesamiento. Las validaciones realizadas en España, Cuba y Brasil demuestran su impacto formativo en la consolidación del razonamiento científico, la coherencia teórico-metodológica y la autonomía cognitiva del investigador en formación. 1. Overview SOCIOKAIROS was conceived as a response to the epistemological gap between thinking and analyzing in social research. Whereas CAQDAS tools (NVivo, Atlas.ti, MAXQDA) start from the © Víctor Hugo Pérez Gallo – University of Zaragoza – [email protected] empirical corpus, SOCIOKAIROS begins from the scientific problem itself, working through its conceptual components: problem reformulation; identification of variables (DV/IV); definition of dimensions and indicators; and generation of hypotheses adjusted to context. The process is dynamic and context-sensitive: the assistant detects social domains (education, gender, poverty, community, etc.) and activates indicator modules accordingly. 2. Implementation SOCIOKAIROS is implemented in Python (PyQt5 GUI) and operates cross-platform (macOS, Windows, Linux). The core algorithm uses natural language parsing and heuristic mapping rules to detect: (1) conceptual relations between terms; (2) contextual categories (e.g., barrio, redes, pobreza, violencia de género); and (3) relevant dimensions for operationalization (Resources, Culture/Norms, Networks, Institutions). It integrates Quality Gates that ensure the coherence and statistical validity of proposed indicators. The export module automatically generates a Word document (SOCIOKAIROS_DOC.docx) including variable tables, indicator formulas, units, sources, and methodological notes. A modular dictionary based on classical sociologists (Marx, Durkheim, Weber, Goffman, Gramsci, Bourdieu) ensures theoretical grounding and epistemological traceability. 3. Impact The system was experimentally applied between 2024–2025 in three universities (Spain, Cuba, Brazil), totaling N=126 students. Results showed a +26-point increase in conceptual clarity and an 85% improvement in students’ ability to connect theory and method. Qualitative evaluations highlighted its capacity to foster epistemic reflexivity, dialogical learning, and autonomous problem construction. 4. Conclusion SOCIOKAIROS demonstrates that digital technology can act as an epistemological mediator, not merely an analytical instrument. By making explicit the reasoning structure behind scientific problem formulation, it offers a path toward computational reflexivity—a mode of learning where researchers think with the machine, not through it. Future developments include integration with CAQDAS tools, semantic ontologies, and open educational repositories. References • Berger, P. L., & Luckmann, T. (1966). The Social Construction of Reality. Anchor Books. • Bourdieu, P. (1979). La distinction. Éditions de Minuit. • Feenberg, A. (2017). Technosystem: The Social Life of Reason. Harvard University Press. • Goffman, E. (1974). Frame Analysis. Harvard University Press. • Gramsci, A. (1971). Selections from the Prison Notebooks. International Publishers. • Pérez Gallo, V. H. (2025). De los CAQDAS a los asistentes heurísticos: hacia una epistemología computacional en sociología del conocimiento. Manuscript, University of Zaragoza. © Víctor Hugo Pérez Gallo – University of Zaragoza – [email protected] Appendix A – University of Zaragoza – SOCIOKAIROS Research Suite v15 (2025) Main Interface (Dashboard) © Víctor Hugo Pérez Gallo – University of Zaragoza – [email protected] Reformulation Window © Víctor Hugo Pérez Gallo – University of Zaragoza – [email protected] Operationalization and Output Panel © Víctor Hugo Pérez Gallo – University of Zaragoza – [email protected] Sociological Dictionary Module © Víctor Hugo Pérez Gallo – University of Zaragoza – [email protected] Appendix B – Brief Technical Manual Workflow: Step 1 — Reformulate the problem; Step 2 — Identify DV/IV; Step 3 — Define dimensions and indicators; Step 4 — Generate hypotheses; Step 5 — Export report. Interface Modules: Dashboard: input and context; Operationalization Module: suggests variables and indicators; Export Panel: produces a Word report with variables, indicators, formulas, and sources. Usage Tips: Use concise problem statements; specify scope and population; review suggested indicators and sources; keep ORCID and metadata consistent. License: This document and the software description are released under CC BY 4.0.