International Journal of Information Management最新文献

GenAI-driven technologies such as ChatGPT influence activities in all areas of life and are used in private and work contexts. This study uses an individual-centered perspective to explain what motivates users to use ChatGPT continuously. We propose that four motivational factors and two technology characteristics together lead to continuance intention among individual ChatGPT users. Therefore, we use a mixed-methods design to combine findings from a quantitative survey study and a qualitative interview study. In Study 1, we follow a configurational approach to analyze multi-wave data from 279 participants with fsQCA. We identify five configurations that lead to high continuance intention and show that perceived ease of use and perceived novelty are necessary for this outcome. Interestingly, the observed factors together cannot explain low continuance intention. In Study 2, we complement these findings with insights based on 15 semi-structured interviews. We illustrate the configurations by identifying 27 individual use cases in the private and work contexts as well as additional factors that facilitate and hinder individual ChatGPT continuance intention. We draw meta-inferences by combining findings of both studies to develop five propositions. Based on that, we contribute a motivational, individual perspective on GenAI continuance intention, present practical implications as well as valuable future research opportunities.

As one of the essential components of daily life, artificial intelligence (AI) assistants gradually form a relationship with users as either servant or partner. This study employed a mixed-methods approach integrating both quantitative and qualitative methodologies to examine the influence of various user-AI assistant relationship types on subjective well-being. The results indicate that partnerships are more conducive to fostering subjective well-being in users than master-servant relationships. This influence is mediated through social support, which encompasses both informational support and emotional support, and is moderated by the presence of feelings of social loneliness. These findings contribute to the available literature on user-AI relationships and subjective well-being and can inform the market launch of AI assistants.

Organizations use information and communication technologies (ICT) to enable remote work, yet the performance of groups using ICT varies widely. Individual intelligence is a strong predictor of individual job performance, yet research shows that when groups use ICT to support decision making tasks, group performance is not influenced by the combined intelligence of group members. Thus, group performance may be unpredictable because ICT inhibits the ability of intelligent team members to influence team outcomes. We completed two studies that examine whether an ICT-based technique for structuring collaboration to improve information integration and organization can help groups leverage the intelligence of group members. Results show that the performance of groups that used traditional commercial ICT was not related to the average intelligence of group members. In contrast, the performance of groups using our proposed technique was related to the average intelligence of group members. Thus, the more structured approach to ICT use enabled groups to perform at the level of their intelligence.

This research is based on the stimulus-organism-response framework and adopts a temporal perspective to examine a relatively overlooked social media behavior, namely, restricted use. We specifically investigate how changes in social overload and information overload over time act as stimuli, resulting in dissatisfaction as a negative personal state, which in turn leads to the restricted use of social media. We employ a mixed-methods approach to examine this phenomenon. First, we conduct a quantitative study using a three-wave survey of 664 domestic Chinese social media users. Both social overload change and information overload change contribute to user dissatisfaction, consequently influencing restricted use behavior. Moreover, prominence negatively moderates the relationship between social overload change and dissatisfaction but does not affect the relationship between information overload change and dissatisfaction. Second, we triangulate these findings through in-depth interviews with 26 experienced social media users. The qualitative inputs from Study 2 help explain both the supported and unsupported hypotheses in Study 1. Our study offers new insights into the dynamic processes underlying restricted usage behavior within the context of social media.

Large language models (LLMs) have proven capable of assisting with many aspects of organizational decision making, such as helping to collect information from databases and helping to brainstorm possible courses of action ahead of making a choice. We propose that broad adoption of these technologies introduces new questions in the study of decision support systems, which assist people with complex and open-ended choices in business. Where traditional study of decision support has focused on bespoke tools to solve narrow problems in specific domains, LLMs offer a general-purpose decision support technology which can be applied in many contexts. To organize the wealth of new questions which result from this shift, we turn to a classic framework from Herbert Simon, which proposes that decision making requires collecting evidence, considering alternatives, and finally making a choice. Working from Simon’s framework, we describe how LLMs introduce new questions at each stage of this decision-making process. We then group new questions into three overarching themes for future research, centered on how LLMs will change individual decision making, how LLMs will change organizational decision making, and how to design new decision support technologies which make use of the new capabilities of LLMs.

Traditional adoption models explain the intention to use information technology (IT). These models draw on theories that relate perceptions of IT to its actual use. To advance the IT adoption literature, we direct attention away from individual perceptions of IT towards understanding the drivers of individuals' decisions as they use IT. This approach may offer richer explanations of individual IT-enabled performance. Using five decisions about IT acceptance and the theoretical lens of automaticity as proposed in previous work, we develop the construct of Information Technology Acceptance as being comprised of the five decisions that users make (i.e., to receive, to grasp, to assess, to be given, and to submit) and validate an instrument with data collected from 524 technology users in three organizations. The results support the model, suggesting that the five acceptance decisions are constant across organizational and technological contexts. We conclude with propositions to shape adoption research further using this new construct.

In this study, we examine how individuals react to an ostracism experience in the metaverse. Based on the temporal need-threat model and virtual identity discrepancy model, two experimental studies and a critical incident study were conducted to investigate how metaverse ostracism affects individuals’ need for belonging, mood, avatar–self discrepancy, and behavioral intentions, as well as how avatar customization moderates this relationship. The results of Study 1 revealed that ostracized participants perceived a higher threat to the need for belonging, a less positive mood, and a higher avatar–self discrepancy than non-ostracized participants. The results of Study 2 revealed that experiencing ostracism led to a greater perception of avatar–self discrepancy only when participants interacted with others using assigned avatars. The results of a moderated mediation analysis showed that the ostracism experience with assigned avatars increased avatar–self discrepancy, thereby reducing intention to continue using the avatars and revisit the metaverse. The mediation effect was not significant when participants customized their avatars. Study 3 replicated the results of Studies 1 and 2 with a larger sample and provided further insight through a qualitative analysis. The theoretical and practical implications are highlighted.

This paper explores the dynamics between value-added services, intermediary brands, and consumer privacy concerns in shaping attitudes toward blockchain-enabled consumer services. Grounded in the Antecedents-Privacy Concerns-Outcomes (APCO) framework, we develop a theoretical model that we test in three experimental studies with a total of 1613 participants, utilizing verbal scenarios featuring blockchain applications for international money transfer and hotel booking. Our research reveals that complete disintermediation via pure peer-to-peer blockchain transactions is unlikely. Consumers prefer blockchain applications offering supplementary services like call centers, password assistance, and cancellation options. As consumers become familiar with blockchain technology, privacy concerns intensify due to its distributed and immutable storage. The fears of data breaches are more pronounced when blockchain applications are offered by unknown startups as opposed to well-known Big Tech companies. However, privacy-conscious consumers also value the prospect of distancing themselves from big-data ecosystems by embracing blockchain solutions from startups. Our research extends the APCO framework by clarifying how privacy concerns, brand-based heuristics, and technological attributes interact. For managerial implications, blockchain applications necessitate re-intermediation to meet consumer preferences. Potential intermediaries, including Big Tech firms, startups, and industry incumbents, face unique challenges in developing and marketing blockchain-enabled consumer services.

Omitting relevant variables in research models is a significant challenge in academic research involving non-experimental research methods. Omitted variables may bias the empirical findings and lead to erroneous conclusions about relationships between factors underlying information systems phenomena. Using data coded from 128 samples reported in 105 prior studies, this study applies meta-regression methods to quantify the extent to which omitted variables bias the reported effect sizes between behavioral intention and system use. Since a direct examination of omitted variables is not possible, this study quantifies four measures: independent variables common to both behavioral intention and system use, moderators for the relationship between behavioral intention and system use, moderators for relationships between other independent variables and system use, and control variables for system use. Meta-regression results show that the effect size for the relationship between behavioral intention and system use decreases when independent variables common to behavioral intention and system use or moderators for the relationship between behavioral intention and system use are included in research models. This implies that the non-inclusion of relevant variables distorts the effect size for the relationship between behavioral intention and system use resulting in a biased understanding of the relationships. It is crucial for research models to include relevant variables such that the omitted variables bias can be effectively handled. Several mitigation strategies and research implications are described.

The emergence of data as a critical asset and the prevalence of technologies such as the Artificial Intelligence of Things (AIoT) on the one hand, and the importance of collaborations for value creation on the other hand have given rise to a new breed of ecosystems known as data ecosystems. While data ecosystems provide new business opportunities, proposing and capturing value in those ecosystems is challenging, and the extant literature provides little guidance in this regard. Our research encompasses two studies that address this limitation and establish a framework for business-model archetypes in the context of AIoT data ecosystems. In the first study, exploratory qualitative research on 28 leading AIoT data ecosystem actors leads to the identification of value propositions and value-capture mechanisms in these ecosystems. We identify eight possible value propositions and eight possible value-capture mechanisms. The second, qualitative study centers on 19 expert interviews. Our analysis leads to the identification of two dimensions – control and customization – that guide the conceptualization and formation of business-model archetypes. Using these dimensions, we develop a framework for business-model archetypes in AIoT data ecosystems. Our findings contribute to the discourse on data ecosystems and offer new perspectives valuable for both researchers and industry practitioners.