Künstliche Intelligenz (KI) kann im Prozess der Leistungsbewertung assistieren und diesen transformieren. Besonders lohnend scheint eine KI-Assistenz bei der Bewertung von komplexem, geschriebenem Text. Jedoch ist der Einsatz von KI im Bewertungsprozess «hochriskant» (EU 2024) und bedarf umfangreicher Analysen. Die vorliegende Studie untersucht, inwiefern ChatGPT-4o die Auswertung und Interpretation von Lerntagebucheinträgen objektiv vornehmen kann. Dafür werden 757 Lerntagebucheinträge aus der geförderten Weiterbildung in Deutschland von Mensch und Maschine bewertet. Sowohl Mensch als auch Maschine erhalten hierzu Kriterien, nach denen die Bewertung vorzunehmen ist; ChatGPT-4o wird diesbezüglich mit einem Prompt unterstützt. Die Übereinstimmung der Bewertungen wird anhand der Masse Sensitivität und Spezifität gemessen. Die Ergebnisse zeigen, dass die Bewertungsvorschläge von ChatGPT-4o eine moderate bis hohe Übereinstimmung mit den menschlichen Bewertungen aufweisen; gleichzeitig neigt ChatGPT-4o jedoch zu einer optimistischen Bewertung der Lerntagebucheinträge. Die Ergebnisse weisen darauf hin, dass eine hybride Intelligenz, also eine Kombination der Stärken von Mensch und Maschine, gewinnbringend für Bewertungsprozesse sein kann. Künftig denkbar sind halbautomatisierte Bewertungsprozesse von Lerntagebucheinträgen, in denen die KI die Bewertung der Lerntagebucheinträge übernimmt und Lehrkräfte bei kritischen Fällen regulierend eingreifen. So könnte die Korrektureffizienz ohne bedeutende Qualitätsverluste gesteigert werden.

Learning diaries are reflective tools, often used as formative assessments in adult education with the aim to promote cognitive and metacognitive learning strategies. As grading of and feedback on learning diaries is effortful for teachers, artificial intelligence (AI) may assist teachers in evaluating learning diaries. A prerequisite is that AI's ratings show high accordance with the teachers' ratings. AI accuracy, measured via absolute accuracy and bias, is the focus of the current study with N = 540 learning diary entries focusing on learning strategies, seven teachers, and ChatGPT-4o. Findings revealed that AI evaluations align closely with teacher assessments, indicated by high overall accuracy and low bias. Interestingly, the accuracy varied based on the types of learning strategies assessed in the diaries. Additionally, individual teacher assessments influenced the alignment between human and AI evaluations, suggesting that teachers applied their profession-specific expertise to the assessment process while AI produced somewhat generic evaluations. Overall, the study results indicate that AI can enhance the efficiency of formative assessments while providing timely feedback to learners.

For successful learning, monitoring one's learning behavior and outcomes is essential. While both the use of monitoring strategies and monitoring judgment accuracy are regarded as indicators for metacognitive monitoring, their joint contribution to metacognitive monitoring and their interrelationship is unclear. The current study investigated how monitoring strategies and monitoring judgments (spontaneous as well as prompted judgments) change micro-genetically and influence each other reciprocally during an exam preparation phase. In this diary study, N = 273 higher education students completed knowledge tests and surveys on five consecutive days. Results indicated interindividual differences in dependence of prior performance and gender for the accuracy of prompted judgments but not for monitoring strategies or spontaneous judgments. Random intercept cross-lagged panel models revealed that higher quality monitoring strategies led to the spontaneous formation of monitoring judgments, while the quality of monitoring strategies and prompted judgment accuracy did not influence each other significantly.

Entrepreneurs benefit significantly from resources within their entrepreneurial ecosystems, but under what conditions do they decide to contribute back (a mechanism called downward causation) and thus revitalize the ecosystem they originated from? Drawing on social exchange theory, we develop a set of drivers of such contribution behaviors and test their influence through a metric-conjoint experiment involving 234 entrepreneurs. Our findings confirm the impact of social exchange theory constructs on entrepreneurs' contribution behaviors and highlight the moderating effects of personal traits—in particular, self-interest and other-orientation—on these dynamics. The key insight of our study is that social exchange structures and entrepreneurs’ relational contexts shape contribution behaviors that underlie the microfoundational dynamics of ecosystem development. This investigation underscores the importance of social structures within entrepreneurial ecosystems and enhances our understanding of the micro-level mechanisms that sustain ecosystem health and development. Furthermore, it offers practical insights that transcend traditional policy approaches, focusing on tailored strategies for cultivating entrepreneur-centered ecosystems.

Context plays a critical role in talent development, yet most national analyses continue to rely on individual-centered talent concepts. This paper highlights the limitations of traditional models for assessing how countries support talent and proposes a resource-oriented, systemic alternative. Building on the Educational and Learning Capital Approach (ELCA), this study argues that national talent development depends on the availability, accessibility, and orchestration of both endogenous and exogenous learning resources across systemic levels. By analyzing the clumping patterns of excellence in STEM, the arts, sports, and innovation, this paper illustrates the unequal global distribution of talent-supportive environments. Seven key principles for effective resource orchestration are outlined, offering a framework for evaluating and strengthening national talent ecosystems. The paper concludes that systematic assessment and strategic enhancement of national resource landscapes are critical for sustainable talent development and for ensuring that human potential can flourish more equitably across countries.

Der Softwaremarkt zählt zu den wachstumsstärksten Sektoren der deutschen Wirtschaft. Unternehmen wie Celonis und Personio zeigen, welches Potenzial in Software-Startups steckt – sowohl im Hinblick auf Umsatz als auch auf die internationale Wettbewerbsfähigkeit. Dennoch ist die Ausfallquote hoch: Viele Startups scheitern frühzeitig, häufig am fehlenden Product-Market-Fit. Gerade Software-Startups stehen vor besonderen Herausforderungen. Sie entwickeln digitale, häufig komplexe Produkte, deren Nutzen sich schwer kommunizieren lässt. Der Irrtum „Build it and they will come“ führt dazu, dass Produkte ohne fundiertes Kundenverständnis entwickelt werden. Um dieses Risiko zu minimieren, gibt es nutzer- und marktzentrierte Methoden. Sie helfen dabei, frühzeitig Rückmeldung aus dem Markt einzuholen, den Produktnutzen klar zu definieren und ein tiefes Verständnis der Zielgruppe zu gewinnen. Wer als Startup transparent kommuniziert und seine Zielgruppe aktiv einbindet, kann nicht nur Produktideen validieren, sondern auch erste loyale Kundengruppen gewinnen. Dieser Beitrag zeigt, dass die Integration nutzer- und marktorientierter Methoden die Erfolgschancen von Software-Startups deutlich erhöhen kann. Eine praxisorientierte Übersicht am Ende des Beitrags zeigt, wie zentrale Methoden je nach Entwicklungsphase gezielt eingesetzt werden können.

School students need to make decisions about their career paths after graduating. In Germany, students can choose between more than 300 vocational training programs, which can be overwhelming. Frequently, the students hesitate to talk with career counselors. The objective of this research is, therefore, to provide a recommendation system for school students to support their decision-making, which is based on their interests and provides recommendations with explanations based on a LLM. This system was developed with a social robot as the user interface to make it easy to use and appeal to the young target group. Based on user observations, preliminary findings indicate that the system is a valuable and engaging approach to support career counseling activities.

Understanding a learner’s resources as a system of interacting components, the success of a learning process is determined by the effectiveness of their interactions. Theoretical assumptions and empirical findings clearly show the importance of resource availability in learning systems but do not sufficiently consider the individuality or the temporal and situational aspects of resource regulation. Therefore, the current study addresses the complex interplay between learning resources (educational and learning capitals) in an individual learner (N = 1) by utilizing multivariate time series data of a 50-day vocabulary learning process with daily assessments of learning resource availability, performance, learning duration, and stress. We draw on methods of psychometric network analysis, modeling all variables in simultaneous interaction and allowing predictions between all variables from measuring point to measuring point (temporal dynamics). Specifically, using a Graphical Vector Autoregressive (graphicalVAR) model, yielding a contemporaneous and a temporal dynamics network model, we identified pivotal resources in regulating the student’s learning processes and outcomes, including resources with strong connections to other variables, intermediary resources, and resources maintaining the system’s homeostasis. This innovative approach has possible applications as a diagnostic tool that lays the foundation for tailored interventions.

This study aims to investigate the influence of two interaction designs on user comfort and intention to use during pre-interaction phase. As part of a field experiment in a retail bank, a proactive and a passive interaction design of a social robot were compared. A standardized questionnaire was used to determine how the interaction design affects the comfort, trust and usage intention of customers and what role trust plays as a mediating factor. The data analysis shows that the proactive design was rated better in terms of psychological comfort and emotional value, but not in terms of trust and intention to use. Comfort with robots positively influenced the intention to use the social robot, with trust serving as a key mediator; in the proactive variant, the effect was only indirect via trust, while in the passive variant, both direct and indirect effects were observed. According to dual processing theory, proactive designs rely on automatic, emotion-driven processes that directly influence comfort, while passive designs encourage reflective decision-making, supporting trust and increasing usage intention despite lower comfort. A balanced integration of both approaches can enhance customer comfort and trust, facilitating the successful adoption of social robots in retail.

AI-based technologies are becoming increasingly significant while transforming human-machine interactions. Yet, many important questions remain unanswered. One example is the research question of the present contribution, regarding the effects of artificial intelligence (AI) on users’ perceptions of dependency on chatbots and smart home systems. The objective is to analyze to what extent users perceive dependencies on these technologies and which factors influence this perception. Based on a survey of 325 users, it was found that dependency perception is currently low but increases with more frequent usage. The results show a substantial association between perception of dependence and frequency of use, with a stronger effect among users of chatbots compared to users of smart home systems. Furthermore, a slight negative effect was observed for attitudes toward AI on dependency perception for both chatbots and smart home systems. To test the hypotheses, linear regression analyses were employed, revealing substantial associations between usage frequency and dependency perception. Despite limitations such as gender imbalance in the sample—even though gender had a negligible effect on PoD –, this study provides valuable insights into the societal impacts of AI and lays the groundwork for future research in this area. Future studies should include larger and more representative samples and develop validated measurement instruments to enhance the generalizability of findings. The results emphasize the necessity of critically evaluating the deployment and integration of AI technologies to identify potential dependencies at an early stage.