The Power of Trust: Designing Trustworthy Machine Learning Systems in Healthcare

2023 | International Conference on Information Systems | Citations: 0

Authors: Fecho, Mariska; Zöll, Anne

Abstract: Machine Learning (ML) systems have an enormous potential to improve medical car ... Expand

Abstract: Machine Learning (ML) systems have an enormous potential to improve medical care, but skepticism about their use persists. Their inscrutability is a major concern which can lead to negative attitudes reducing end users trust and resulting in rejection. Consequently, many ML systems in healthcare suffer from a lack of user-centricity. To overcome these challenges, we designed a user-centered, trustworthy ML system by applying design science research. The design includes meta-requirements and design principles instantiated by mockups. The design is grounded on our kernel theory, the Trustworthy Artificial Intelligence principles. In three design cycles, we refined the design through focus group discussions (N1=8), evaluation of existing applications, and an online survey (N2=40). Finally, an effectiveness test was conducted with end users (N3=80) to assess the perceived trustworthiness of our design. The results demonstrated that the end users did indeed perceive our design as more trustworthy. Collapse

Topics: health information system artificial intelligence information privacy concern privacy collaborative work

Methods: survey design science kernel theory experimental group design artifact

Theories: trust theory

Avoiding the Diffusion of Responsibility in Social Networking Groups: A Field Experiment on Responses to Online Help-Request Referrals

2023 | Journal of the Association for Information Systems | Citations: 0

Authors: Feng, Nan; Hu, Zhongtao; Wu, Yi; Choi, Ben

Abstract: Help-request referral (HRR) campaigns are being increasingly adopted by online s ... Expand

Abstract: Help-request referral (HRR) campaigns are being increasingly adopted by online service providers to attract online traffic and engage new users. In these campaigns, participants are required to collect a specific amount of help from their social network in exchange for attractive rewards. Participants often face challenges in obtaining positive feedback from their social networking groups due to the diffusion of responsibility. Based on the triangle model of responsibility, this study proposes and empirically tests a research model that examines two key determinants (i.e., request personalization and relational closeness) of recipients’ perceived responsibility that further shape their responses to HRR. Our field experiment empirically supports the research hypotheses. Overall, this study enriches the literature on online social referrals by focusing on HRR, where social referrals are broadcasted and senders reap referral benefits. The study also provides practical insights for HRR designers and social networking service providers. Collapse

Topics: personalization usability WeChat marketing management online word of mouth

Methods: survey experiment field experiment logistic regression analysis of covariance

Bridging the Gap: Evidence from the Return Migration of African Scientists

2023 | Organization Science | Citations: 0

Authors: Fry, Caroline Viola

Abstract: Although prior research has highlighted the benefits that accrue to brokers in i ... Expand

Abstract: Although prior research has highlighted the benefits that accrue to brokers in innovation networks, much less attention is devoted to understanding who benefits from associating with a broker. This study focuses on the impact of associating with a specific kind of broker—a core/periphery bridge—that is, one that spans central and peripheral actors. I argue that actors associated with a core/periphery bridge benefit more when they have no central connection, or are outsiders in the network, due to their greater need for sponsorship from the broker. I explore this idea in the context of the return migration of American-trained scientists to African institutions, who span the core and the periphery of the global scientific network. I evaluate the impact of their return on the publication outcomes of nonmigrant scientists based in African institutions and find that following the arrival of a returnee in their institution, the nonmigrants who are not already connected to scientists in top global institutions have a greater publication output through improved access to central knowledge and connections. The findings contribute to a better understanding of how brokers can influence innovation systems more broadly. Supplemental Material: The online appendices available at https://doi.org/10.1287/orsc.2022.1580. Collapse

Topics: IT career social capital accounting anonymity interorganizational relationship

Methods: longitudinal research robustness check descriptive statistic experimental group regression analysis method

The Power of Identity Cues in Text-Based Customer Service: Evidence from Twitter

2023 | Management Information Systems Quarterly | Citations: 0

Authors: Gao, Yang; Rui, Huaxia; Sun, Shujing

Abstract: Text-based customer service is emerging as an important channel through which c ... Expand

Abstract: Text-based customer service is emerging as an important channel through which companies can assist customers. However, the use of few identity cues may cause customers to feel limited social presence and even suspect the human identity of agents, especially in the current age of advanced algorithms. Does such a lack of social presence affect service interactions? We studied this timely question by evaluating the impact of customers' perceived social presence on service outcomes and customers' attitudes toward agents. Our identification strategy hinged on Southwest Airlines' sudden requirement to include a first name in response to service requests on Twitter, which enhanced customers' perceived level of social presence. This change led customers to become more willing to engage and more likely to reach a resolution upon engagement. We further conducted a randomized experiment to understand the underlying mechanisms. We found that the effects were mainly driven by customers who were ex ante uncertain or suspicious about the human identity of agents, and the presence of identity cues improved service outcomes by enhancing customers' perceived levels of trust and empathy. Additionally, we found no evidence of elevated verbal aggression from customers toward agents with identity cues, although a mechanism test revealed the moderating role of customers' emotional states. Our study highlights the importance of social presence in text-based customer service and suggests a readily available and almost costless strategy for firms: signal humanization through identity cues. Collapse

Topics: customer service Twitter social presence social media customer satisfaction

Methods: experiment computational algorithm synthetic control method survey support vector machine

Theories: social presence theory big five model

Exploring Personality-based Heterogeneity in Metaknowledge and Human-AI Collaboration

2023 | International Conference on Information Systems | Citations: 0

Authors: Gass, Daniel F

Abstract: While artificial intelligence (AI) has great potential to transform how we work ... Expand

Abstract: While artificial intelligence (AI) has great potential to transform how we work, it's clear that in many contexts, humans and AI will work side by side. According to recent research, humans' ability to capture complementarities while working collaboratively with AI is critically constrained by our limited ability to assess what we know, i.e., metaknowledge. However, extant research has largely overlooked a critical difference between human and AI: Human behavior is shaped by personality. I explore how differences arising from personality traits impact metaknowledge and delegation behavior towards AI. Using an image classification experiment, I find that higher levels of conscientiousness improve, while higher levels extraversion harm humans' ability to work collaboratively with AI. Differences in metaknowledge help to explain this relationship. These findings motivate further research on personality in human-AI collaboration and the design of collaborative systems which foster metaknowledge and improved performance. Collapse

Topics: artificial intelligence personality computer supported cooperative work database system human computer interaction

Methods: experimental group experiment survey linear regression analysis descriptive statistic

Theories: big five model

Fun Shopping: A Randomized Field Experiment on Gamification

2023 | Information Systems Research | Citations: 0

Authors: Ho, Yi-Jen (Ian); Liu, Siyuan; Wang, Lei

Abstract: Gamification utilizes game-like features to engage participants, widely implemen ... Expand

Abstract: Gamification utilizes game-like features to engage participants, widely implemented in a variety of contexts. Such an IT-enabled engagement strategy serves as a marketing device to boost sales and customer loyalty. This study focuses on two significant game elements (i.e., badges and leaderboards) that promote consumer motivations and social comparisons. To qualify the impacts, we conduct a randomized field experiment at one of the largest shopping malls in Asia. In the experiment, we contrast the two elements against coupons regarding various shopping outcomes. A two-period design (consisting of the treatment and posttreatment periods) identifies the long-term behavior changes after the treatment removals. The main results suggest that badging and leaderboarding promote sales by 21.5% and 22.5% in the treatment period, respectively, whereas couponing delivers a more potent effect of 31.7%. In the posttreatment period, the gamification impacts remain significant compared with the baseline, but the influence of couponing fades out. Besides, the additional analyses document the salient heterogeneous treatment effects across demographics. We further discover the substantial differences in the within-group heterogeneity across the treatments. Specifically, badging is a balanced tool for attracting the general public, whereas leaderboarding is a double-edged sword that could encourage self-reinforcing or self-banishing. Finally, gamification brings more explorations that lead to additional sales and engagements. Overall, the robust results can be translated into actionable strategies to utilize gamification proactively.History: Ravi Bapna, Senior Editor; Yili (Kevin) Hong, Associate Editor. Collapse

Topics: gamification wireless LAN consumer behavior technology adoption user interface

Methods: experiment experimental group survey randomized field experiment parametric test

Theories: motivation theory self determination theory

The Effect of Community Managers on Online Idea Crowdsourcing Activities

2023 | Journal of the Association for Information Systems | Citations: 0

Authors: Hornuf, Lars; Jeworrek, Sabrina

Abstract: In this study, we investigate whether and to what extent community managers in o ... Expand

Abstract: In this study, we investigate whether and to what extent community managers in online collaborative communities can stimulate community activities through their engagement. Using a novel data set of 22 large online idea crowdsourcing campaigns, we find that moderate but steady manager activities are adequate to enhance community participation. Moreover, we show that appreciation, motivation, and intellectual stimulation by community managers are positively associated with community participation but that the effectiveness of these communication strategies depends on the form of participation managers wish to encourage. Finally, the data reveal that community manager activities requiring more effort, such as media file uploads vs. simple written comments, have a stronger effect on community participation. Collapse

Topics: crowdsourcing innovation management online community missing data privacy

Methods: propensity score method statistical power analysis Poisson regression regression analysis method econometric modeling

Theories: transformational leadership theory

Take It or Not? Impact of Taking Investments from Tech Giants on IT Startups’ Future Funding

2023 | Americas Conference on Information Systems | Citations: 0

Authors: Huang, Linmei; Wang, Shuting (Ada); Lang, Karl

Abstract: While tech giants have become important external equity financing sources, it r ... Expand

Abstract: While tech giants have become important external equity financing sources, it remains unclear whether ITstartups should take investments from tech giants. In this study we empirically examine how taking investments from tech giants influences future funding of IT-startups. Our results indicate that taking investments from tech giants significantly reduces the amount of funding IT-startups could obtain in the following rounds. Additionally, we find that social ties between IT-startups' regions and locations with substantial capital will attenuate such negative impact of taking investments from tech giants. We further explore the mechanisms. Results suggest that when tech giants make investments in IT-startups, they predominate the new technologies by applying for more patents in the relevant fields and crowd out new investors. We contribute to the literature in IS by examining how investments from tech giants may influence the growth of IT-startups and how social ties may play a role in this interaction. Collapse

Topics: financial performance systems development artificial intelligence innovation management competitive advantage

Methods: propensity score matching experimental group theory development robustness check

Contract Choice, Moral Hazard, and Performance Evaluation: Evidence from Online Labor Markets

2023 | HICSS | Citations: 0

Authors: Huang, Peng; Wang, Yifei

Abstract: Due to the spatial and temporal separations between clients and freelancers, onl ... Expand

Abstract: Due to the spatial and temporal separations between clients and freelancers, online labor markets (OLMs) are particularly susceptible to issues related to information asymmetry. Based on the economics of information, we hypothesize that the choice of contract type—i.e., between the fixed-priced (FP) contract and the time-and-materials (TM) contract—has important implications for curbing moral hazard during contract execution, and therefore will influence the client’s perceived contractual performance upon project completion. We test the predictions by assembling a dataset of data analytics projects completed by freelancers on Upwork, the largest online freelancing platform. We find that, consistent with our hypothesis, freelancers under a TM contract receive significantly lower performance ratings by their clients on average compared to those under an FP contract. Interestingly, we also find that the level of expertise required for a project moderates the effect of contract choice on client satisfaction; the negative impact of a TM contract is smaller (i.e., less negative) when a project requires intermediate-level or expert-level skills. Our study offers useful insights into an important institutional determinant of contractual performance evaluation, which has profound implications for freelancers’ reputations in OLMs. Collapse

Topics: moral hazard Upwork analytical information system programming language Elance

Methods: regression analysis method propensity score matching experimental group theory development longitudinal research

Theories: lemon market theory agency theory

Does Social Influence Change with Other Information Sources? A Large-Scale Randomized Experiment in Medical Crowdfunding

2023 | Information Systems Research | Citations: 0

Authors: Hur, Yun Young; Jin, Fujie; Li, Xitong; Cheng, Yuan; Hu, Yu Jeffrey

Abstract: We examine how social influence interacts with other information sources to affe ... Expand

Abstract: We examine how social influence interacts with other information sources to affect user behaviors in the context of medical crowdfunding. We conduct a large-scale randomized field experiment on a leading medical crowdfunding platform, showing friends’ donation information to donors in the treatment group and not showing such information in the control group, and examine how the likelihood to donate differs. In addition, we conduct a survey on Amazon Mechanical Turk to evaluate the informational value of different case attributes in conveying the patients’ need for help to donors. We find that for cases containing attributes with high informational value (e.g., minor patient, severe conditions), social influence is insignificant. In contrast, for cases lacking attributes with high informational value, social influence significantly increases donors’ likelihood to donate. Overall, our results show that the impact of social influence depends on the informational value of other information sources, suggesting that the social influence in our context is primarily informational. Our findings indicate that rather than generating an entrenchment effect, where cases with attributes of high informational value attract disproportionate benefits, social influence can increase donation likelihood to cases that lack such attributes, promoting more equal access to resources overall. History: Olivia Liu Sheng, Senior Editor; Yuliang (Oliver) Yao, Associate Editor. Supplemental Material: The online appendices are available at https://doi.org/10.1287/isre.2022.1189. Collapse

Topics: social influence crowdsourcing crowdfunding social network user behavior

Methods: experimental group experiment survey randomized field experiment robustness check

Theories: lemon market theory