Merlo, G., Taibi, D., & Becchimanzi, M. (2025, November 27–28). Using a Telegram Bot for longitudinal educational research: An analysis of engagement and its longitudinal decay [Paper presentation]. CKBG Conference 2025: Educazione e Creazione di Valore nell'Era Digitale, Florence, Italy.

In recent years, the integration of technology into educational processes has become a wellestablished phenomenon, and research interest in the field is progressively shifting from a focus on didactic efficiency (Chauhan, 2017; Hillmayr et al., 2020) to a deeper need related to digital humanism (Werthner & Nida-Rümelin, 2022). This paradigm believes technology's ultimate goal must be to promote student well-being and personal growth, rather than to optimize instructional delivery. From this perspective, technology becomes crucial for understanding the complex dynamics between educational institutions and students, viewed as whole individuals. The pervasive adoption of smartphones facilitates the application of an effective methodological solution: Ecological Momentary Assessment (EMA) (Vries et al., 2021). This approach allows for the collection of “in-the-moment” data directly within real-life contexts, offering the advantage of reducing biases associated with traditional questionnaire-based measurements. It is particularly effective for capturing fluctuations in thoughts, emotions, and behaviors, and it finds applications in fields such as psychology, education, and clinical research. However, the implementation of EMA often faces significant obstacles: participants are frequently required to install custom-developed applications. This approach introduces several practical barriers, such as resistance to installing new apps and the need to learn new interfaces, which can jeopardize student participation in data collection (Wrzus & Neubauer, 2022). To address these limitations, the present study explores the use of a Telegram bot for EMA data collection, based on the rationale that this method may better integrate into adolescents’ communication workflows. Although Telegram bots are already employed in educational research contexts (Ong et al., 2021; Safitri et al., 2023) and an initial example of an EMA application has been proposed (Vogel, 2025), a knowledge gap is evident in the existing literature. Indeed, while the literature shows that general compliance with EMA tends to be high (Runyan et al., 2013; Wen et al., 2017; Wrzus & Neubauer, 2021), it remains unclear how participation might vary differentially based on students’ involvement in specific contextual activities. Consequently, the present study aims to analyze patterns of participation in an EMA data collection conducted via a Telegram bot, by comparing a group of students involved in a psychocorporal workshop with a control group. The research is guided by the following questions: RQ1: In the ecological monitoring of well-being via a Telegram bot, do the experimental and control groups show an equivalent level of initial engagement, or is there an immediate difference in participation? RQ2: What is the influence of time on student engagement in the ecological monitoring of wellbeing via a Telegram bot? Does the intervention modulate the slope of engagement trajectories, generating a differential attrition rate between the groups? RQ3: Does the effect of the intervention manifest as a direct increase in student engagement with the monitoring, or rather as a protective factor that moderates its longitudinal decay?