Array
(
    [0] => stdClass Object
        (
            [journal] => stdClass Object
                (
                    [id_jnl] => 87
                )

        )

    [1] => stdClass Object
        (
            [section] => stdClass Object
                (
                    [section] => 281
                )

        )

    [2] => stdClass Object
        (
            [title] => Array
                (
                    [0] => Measuring the micro-level impact of generative AI on science journalists’ professional routines: A methodological model for science communication contexts@en
                    [1] => Impacto de la IA generativa en las rutinas profesionales del periodista: modelo metodológico a nivel micro@es
                )

        )

    [3] => stdClass Object
        (
            [abstract] => Array
                (
                    [0] => Objective. This study proposed a methodological framework to measure the micro-level impact of generative artificial intelligence (AI) on science journalists’ professional routines, while remaining adaptable to broader journalistic contexts, taking the journalist as the primary unit of analysis. The study defined “impact” as the degree of transformation attributable to the integration of generative AI into everyday journalistic practices. This integration encompassed operational, cognitive, ethical, and professional dimensions.
Design/Methodology/Approach. This study proposed a mixed-methods measurement design structured as a sequential protocol. First, it conceptualized the construct and identified six analytical dimensions: use patterns, perceived productivity, autonomy and editorial decision-making, ethical risk perception, audience orientation and algorithmic mediation, and well-being and cognitive load. Second, it proposed an operationalization strategy through observable indicators and Likert-type items, including examples for each dimension. Third, it delineated a validation pathway that integrated content validity procedures, pilot testing, and dimensionality assessment through exploratory and confirmatory factor analysis. The incorporation of qualitative components, such as semi-structured interviews and usage diaries, was a methodological approach employed to support item development and interpretative triangulation.
Conclusions. The contribution lay in the specification of an operational measurement architecture that connected theoretical constructs with empirical indicators. The proposed model elucidated the process of transitioning from abstract dimensions to measurable variables and furnished a structured approach for constructing a composite impact index or a set of sub-indices that reflected diverse facets of AI integration. Additionally, it shed light on significant methodological challenges, including the interpretation of mixed positive and negative effects and the role of qualitative data in refining quantitative measures. The study proposes a replicable and adaptable framework for measuring the micro-level impact of generative AI in journalism. Explicitly defining constructs, dimensions, indicators, and validation procedures, it offers a robust basis for future empirical applications, pilot testing, and comparative research across media contexts.
Originality/Value. The study presents a theoretically grounded and methodologically explicit proposal for developing instruments in the study of AI, science journalism, and science communication. This proposal bridges the gap between conceptual debate and empirical measurement.@en
                    [1] => Objetivo. Este estudio desarrolla el nivel micro de un modelo metodológico para medir el impacto real de la inteligencia artificial generativa en el periodismo, tomando al periodista como unidad de análisis y atendiendo a rutinas, percepciones y toma de decisiones.Diseño/Metodología/Enfoque. Se propone un diseño mixto que integra (a) una encuesta estructurada a periodistas en ejercicio que emplean IA generativa, con indicadores tipo Likert en seis dimensiones (patrones de uso, productividad percibida, autonomía y decisiones, percepción de riesgos éticos, competencias digitales y bienestar o carga cognitiva); (b) entrevistas semiestructuradas a perfiles diversos en redacciones híbridas (redactores, responsables de audiencias, editores de datos, responsables de innovación y verificadores); y (c) un índice compuesto de impacto mediante análisis de componentes principales, ajustado con análisis temático y triangulación.Resultados/Discusión. El modelo entiende el impacto más allá de la automatización de tareas e incorpora cambios en autonomía, responsabilidad y criterio editorial, junto con tensiones entre eficiencia y riesgos como alucinaciones, opacidad y sesgos. Además, incluye la dimensión de distribución y relación con audiencias mediada por sistemas de recomendación, permitiendo perfilar tipos de periodistas según intensidad y modalidad de integración de la IA.Conclusiones. El marco micro propuesto es replicable en distintos tipos de redacciones y facilita investigación comparada. Aporta una base rigurosa para evaluar una adopción responsable al vincular indicadores medibles con experiencias profesionales situadas.Originalidad/Valor. La propuesta ofrece una arquitectura de medición aplicable y teóricamente fundamentada que sintetiza exposición objetiva, percepciones subjetivas y contexto organizativo en un enfoque evaluativo único.@es
                )

        )

    [4] => stdClass Object
        (
            [author] => Array
                (
                    [0] => Francisco J. Cristófol
                )

        )

    [5] => stdClass Object
        (
            [subject] => Array
                (
                    [0] => Generative ai@en
                    [1] => Science journalism@en
                    [2] => Science communication@en
                    [3] => Measurement model@en
                    [4] => Professional routines@en
                    [5] => Mixed methods@en
                    [6] => Ia generativa@es
                    [7] => Periodismo@es
                    [8] => Rutinas profesionales@es
                    [9] => Metodologías mixtas@es
                    [10] => Autonomía@es
                    [11] => Índice de impacto@es
                )

        )

    [6] => stdClass Object
        (
            [source] => stdClass Object
                (
                    [vol] => 6
                    [nr] => 
                    [year] => 2026
                    [theme] => 
                )

        )

    [7] => stdClass Object
        (
            [datePub] => Array
                (
                    [0] => 2026-06-09
                )

        )

    [8] => stdClass Object
        (
            [DOI] => Array
                (
                    [0] => stdClass Object
                        (
                            [type] => DOI
                            [value] => Array
                                (
                                    [0] => 10.47909/ijsmc.340
                                )

                        )

                )

        )

    [9] => stdClass Object
        (
            [http] => Array
                (
                    [0] => stdClass Object
                        (
                            [type] => HTTP
                            [value] => Array
                                (
                                    [0] => https://ijsmc.pro-metrics.org/index.php/i/article/view/340
                                )

                        )

                    [1] => stdClass Object
                        (
                            [type] => HTTP
                            [value] => Array
                                (
                                    [0] => https://ijsmc.pro-metrics.org/index.php/i/article/view/340/216
                                )

                        )

                )

        )

    [10] => stdClass Object
        (
            [language] => Array
                (
                    [0] => en
                )

        )

    [11] => stdClass Object
        (
            [license] => Array
                (
                    [0] => Copr
                    [1] => by-nc/4.0
                )

        )

)