Identification of Causal Mechanisms from Randomized Experiments: A Framework for Endogenous Mediation Analysis

2023 | Information Systems Research | Citations: 0

Authors: Peng, Jing

Abstract: Experimental research in the business disciplines often focuses on the overall t ... Expand

Abstract: Experimental research in the business disciplines often focuses on the overall treatment effect and the heterogeneity therein. Whereas this type of research allows us to understand the strength and direction of the treatment effect under different conditions, it does not directly speak to the generative mechanisms, namely, why and how the effect arises. A standard procedure to identify the mechanisms underlying a treatment effect is mediation analysis, but extant mediation analysis frameworks either have no causal interpretation or require the mediators to be unconfounded. Because mediators are posttreatment variables that typically cannot be preassigned beforehand, the endogeneity of mediators remains a serious concern even in randomized experiments. In response to this issue, we present a flexible endogenous mediation analysis framework that still has causal interpretation when the mediator is endogenous. We then discuss the identification conditions for different types of endogenous mediators, including unobserved or partially observed ones, under this framework. We show that endogenous mediation models can be parametrically identified without an instrumental variable when the generating process of the mediator is nonlinear. We further examine how the identification strengths of these models vary with a series of factors, including the level of endogeneity, the goodness of fit of the mediator model, the percentage of observed mediator values, and the misspecification of the error terms. Finally, we provide guidelines on when and how to use endogenous mediation analysis and discuss implications for experimental design and empirical research. We offer an R package that implements the proposed endogenous mediation models.History: Ravi Bapna, Senior Editor; Anuj Kumar, Associate Editor.Supplemental Material: The online appendix is available at https://doi.org/10.1287/isre.2022.1113. Collapse

Topics: logistics management accounting

Methods: simulation mediation analysis sensitivity analysis experiment experimental design

Peer Priming? A Large-Scale Field Experiment Studying the Impact of Popular Rankings on Demand in Mobile Retail

2023 | International Conference on Information Systems | Citations: 0

Authors: Zheng, Shuang; Tong, Jack; Fang, Sihan; Gopal, Anand; Li, Xianneng; Jiang, Qiancheng

Abstract: Consumers on mobile retail apps face significant search costs due to the small ... Expand

Abstract: Consumers on mobile retail apps face significant search costs due to the small screen size of devices. One of the search aid features to improve the search convenience is to show consumers a small set of frequently used searches conducted by peer consumers on the platform as a prime cue. We refer to this feature as the popular ranking search aid (PRSA). Collaborating with Meituan, a leading services mobile app in China, we implement a large-scale field experiment to explore how PRSA affects consumer search activities and purchases. Our analyses generate three key findings. First, PRSA leads to an increase of 18.6% in page views and a 6.4% increase in purchases. Second, the change in shopping behavior emerges through a change in search behavior with more nondirected searches and fewer directed searches. Third, our mediation analysis supports that search behavior mediates the business outcomes. We offer theoretical and managerial implications. Collapse

Topics: mobile application recommender system crowdsourcing mobile business systems design

Methods: experimental group experiment field experiment mediation analysis BERT

Theoretical and practical applications of blockchain in healthcare information management

2022 | Information & Management | Citations: 0

Authors: Sadeghi R., J. Kiarash; Prybutok, Victor R.; Sauser, Brian

Abstract: A primary objective of blockchain technology is to address information security ... Expand

Abstract: A primary objective of blockchain technology is to address information security and efficiency issues related to existing information sharing systems. For the sharing of health records, little is known about the application of blockchain in management information systems. There are strict regulations for sharing health information due to insecure systems and privacy concerns. To significantly and effectively improve medical diagnosis, it is beneficial to have efficient, reliable, and accurate accessibility to a patient’s full medical history. Due to concerns with the security of health systems and privacy concerns, medical histories are not always accessible to healthcare providers. To help increase accessibility options, this research proposes a blockchain-based model that facilitates sharing medical records in a manner that is beneficial to both healthcare providers and patients. Social exchange theory provides the theoretical support for the conceptual model presented. Experimental findings based on 151 participants revealed that the blockchain technology can provide a secure information system and increase patient motivation to share medical records. To show the applicability of the proposed use of blockchain from a practical managerial perspective, we show feasibility by developing an Android software application. Collapse

Topics: blockchain electronic health record healthcare data health information system Android

Methods: experiment structural equation modeling data transformation factor analysis experimental group

Theories: social exchange theory

The Piggy in the Middle: The Role of Mediators in PLS-SEM-based Prediction: A Research Note

2021 | ACM SIGMIS Database | Citations: 0

Authors: Danks, Nicholas P.

Abstract: Researchers are becoming cognizant of the value of conducting predictive analysi ... Expand

Abstract: Researchers are becoming cognizant of the value of conducting predictive analysis using partial least squares structural equation modeling (PLS-SEM) for both the evaluation of overfit and to illustrate the practical value of models. Mediators are a popular mechanism for adding nuance and greater explanatory power to causal models. However, mediators pose a special challenge to generating predictions as they serve a dual role of antecedent and outcome. Solutions for generating predictions from mediated PLS-SEM models have not been suitably explored or documented, nor has there been exploration of whether the added model complexity of such mediators is justified in the light of predictive performance. We address that gap by evaluating methods for generating predictions from mediated models, and propose a simple metric that quantifies the predictive contribution of the mediator (PCM). We conduct Monte Carlo simulations and then apply the methods in an empirical demonstration. We find that there is no simple best solution, but that all three approaches have strengths and weaknesses. Further, the PCM metric performs well to quantify the predictive qualities of the mediator over-and-above the non-mediated alternative. We present guidelines on selecting the most appropriate method and applying PCM for additional evidence to support research conclusions. Collapse

Topics: evaluation criteria

Methods: simulation Monte Carlo simulation machine learning descriptive statistic partial least squares path modeling

Theories: unified theory of acceptance and use of technology

A Note on the Impact of Daily Deals on Local Retailers’ Online Reputation: Mediation Effects of the Consumer Experience

2020 | Information Systems Research | Citations: 0

Authors: Bai, Xue; Marsden, James R.; Ross, William T.; Wang, Gang

Abstract: This study investigates the impact of daily deals on local retailers’ (restauran ... Expand

Abstract: This study investigates the impact of daily deals on local retailers’ (restaurants’) online ratings. We collected and utilized a comprehensive panel data set that combines information on restaurants’ deal offerings (Groupon or LivingSocial) with their Yelp review details. Although demonstrating a negative main effect of daily deals on a restaurant’s monthly average ratings, we worked to uncover the underlying mechanisms by focusing on the mediating role of consumers’ postconsumption perception. Our mediation analyses show that daily deals are associated with the reduction of both consumers’ perceived food quality and perceived service quality as revealed in review texts, which leads to subsequent declines in a restaurant’s online ratings. We further noted and studied two types of reviews that existed during the deal-redemption period: (1) reviews that mentioned daily deals (DD Reviews) and (2) reviews that did not mention daily deals (NDD Reviews). NDD Reviews are usually from regular customers, whereas DD Reviews are likely from a different base of customers, deal users. Our separate analyses demonstrate differential mediation processes between DD Reviews and NDD Reviews. For DD Reviews, both perceived food and service quality had mediation roles, suggesting a mismatch effect. That is, daily deals attract discount-focused consumers, including new customers, who are less likely to appreciate the food or service and, therefore, leave a negative review. In contrast, for NDD Reviews, only perceived service quality had a mediation role, suggesting a negative externality effect. That is, the large volume of deal redemptions by deal users may induce a longer waiting time (to be seated and/or served) and, thus, reduced perceived service quality for regular customers. These results deepen our understanding of the impact of daily deals on business online reputation and provide important practical guidance to both local retailers and daily deal platforms. Collapse

Topics: service quality online review online word of mouth social influence Facebook

Methods: mediation analysis natural language processing Yelp review sentiment analysis support vector machine

Getting to Know You: Motivating Cross-Understanding for Improved Team and Individual Performance

2020 | Organization Science | Citations: 0

Authors: Janardhanan, Niranjan S.; Lewis, Kyle; Reger, Rhonda K.; Stevens, Cynthia K.

Abstract: Many contemporary organizations depend on team-based organizing to achieve high ... Expand

Abstract: Many contemporary organizations depend on team-based organizing to achieve high performance, innovate services and products, and adapt to environmental turbulence. Significant research focuses on understanding how teams develop, assimilate, and apply diverse information; yet, organizational practices have evolved in new ways that are not fully explored in the teams literature. Individuals with diverse motivations, knowledge, and perspectives are often assigned to teams, creating burdens for members to develop effective ways to work together, learn from each other, and achieve goals amid the complexity of today’s organizational contexts. In this paper, we examine a multilevel model of how team goal orientation affects cross-understanding—the extent to which team members understand the other members’ mental models—which in turn, affects team and individual performance. We examine these effects using 160 teams of 859 participants who completed a semester-long business simulation. Findings show that the more team members are motivated by learning goals, the greater a team’s cross-understanding and subsequent team and individual performance. These effects are dampened when members are motivated by performance goals—to avoid mistakes or prove competence. This study expands the cross-understanding literature, revealing motivational antecedents that explain why some teams develop higher cross-understanding than others. We also contribute to the goal orientation literature by demonstrating that team goal orientation influences members’ learning about other members and in so doing, also affects team and individual performance. Because team motivation can be influenced by organizational practices, our findings also contribute practical insights for organizational leaders. Collapse

Topics: team productivity information exchange balanced scorecard

Methods: statistical hypothesis test simulation hierarchical linear modeling longitudinal research survey

Beyond Overall Treatment Effects: Leveraging Covariates in Randomized Experiments Guided by Causal Structure

2020 | Information Systems Research | Citations: 1

Authors: Tafti, Ali; Shmueli, Galit

Abstract: Researchers using randomized controlled trials (RCTs) often subgroup or conditio ... Expand

Abstract: Researchers using randomized controlled trials (RCTs) often subgroup or condition on auxiliary variables that are not the randomized treatment variable. There are many good reasons to condition on auxiliary variables—also referred to as control variables or covariates—in randomized experiments. In particular, designing and conducting RCTs is costly to researchers and subjects, and therefore it is important to derive greater value from RCT data; measuring not just the average treatment effect (ATE), but also finding more nuanced insights about the underlying theoretical mechanisms and generalizing the inferences. Unfortunately, there are many confusing and even contradictory guidelines on the use of subgroups or auxiliary variables in RCTs. We show how researchers can leverage covariates without biasing their causal inferences, by applying a few simple rules based on Judea Pearl’s causal diagramming framework. We demonstrate how to create a causal schema, through careful and deliberate operationalization of auxiliary covariates, in order to analyze the intermediate effects along a causal chain from the treatment to outcome; and we discuss some other ways to leverage covariates for theory development and generalization of findings from RCTs. We present a criterion for distinguishing pretreatment and posttreatment variables that is based on directed acyclic graphs (DAGs). We provide a succinct set of guidelines to help readers begin to employ some essential techniques of DAG-based causal analysis. Finally, we provide a series of short tutorials (with accompanying simulated data and R scripts) to help readers explore the connections between RCT and observational contexts in causal diagramming. This commentary aims to raise awareness of the DAG methodology, explain its usefulness to experimental research, and encourage adoption in the IS community for studies using RCTs as well as observational data. Collapse

Topics: anonymity online dating digital platform open source

Methods: experiment simulation quantitative observational study theory development regression analysis method

CONSUMER STUDIES ON DIGITAL PLATFORMS ADOPTION AND CONTINUANCE: A STRUCTURED LITERATURE REVIEW

2019 | European Conference On Information Systems | Citations: 0

Authors: Faber, Roel

Abstract: As digital platforms have become the locus of competition and innovation in the ... Expand

Abstract: As digital platforms have become the locus of competition and innovation in the digital economy, it is imperative to understand consumer decisions on adoption and usage. Digital platforms have unique traits, as they are extensible and often mediate between users. These traits may require understandings to move beyond generic (post-)adoption theories. This paper reviews the rapidly growing body of empirical consumer studies on digital platforms in order to identify emerging trends in this field. After a structured literature review, we conduct a network analysis using the hypotheses that are supported in the papers. We find a wide variety of over 130 concepts is considered. Factors from generic (post-) adoption theories are often studied, alongside factors related to the concept of network effects. We find differences between the importance of direct, indirect, local and global network effects. We find that studies pay relatively little attention to characteristics that arise from the extensibility of digital platforms, such as openness, control, security and privacy. We posit that IS researchers have the important task of opening up this black box of digital platform characteristics and theorize how notions of extensibility and generativity affect consumer decisions, going beyond mainstream theories on adoption and continuance. Collapse

Topics: digital platform network effect database system privacy mobile application

Methods: network analysis structured literature research literature sample statistical hypothesis test design artifact

Theories: technology acceptance theory technology acceptance model

Information and Communications Technologies as an Enabler of Supply Chain Integration

2018 | Americas Conference on Information Systems | Citations: 1

Authors: Oguz, Abdullah; Xie, Wei; Palvia, Prashant; Amoako-Gyampah, Kwasi

Abstract: This study is a meta-research analysis that examined the role of information and ... Expand

Abstract: This study is a meta-research analysis that examined the role of information and communication technologies (ICTs) as an enabler of supply chain integration. Our analysis covered 55 papers in 34 journals published between 2007 and 2017. RFID systems were the most investigated ICTs in the supply chain integration literature. Besides its impact as an independent variable, ICT’s mediating and moderating roles are also noteworthy. Our study has demonstrated the underutilization of theories in general, dominance of surveys and single case studies, and future research opportunities for the underutilized IS types in SCI. Based on the findings, we proposed a research framework. Our framework illustrated two types of moderating factors between the use of ICT and level of SCI: (1) ICT-related factors as an enabler of SCI, and (2) other factors as an enabler of SCI. ICT-related factors are suitability, criticality, and maturity of ICT; ICT capabilities of, and integration within, an organization; ICT integration with SC partners; and technological differences among available ICT applications. Other factors are availability of industry standards; open information-sharing environment; institutional pressures and organizational culture; and involvement of top management. Collapse

Topics: supply chain management radio frequency identification interorganizational system electronic data interchange enterprise resource planning

Methods: literature sample meta analysis literature study case study operations research

Modelling Online Community Members' Motivation: A Computational Model Based on Social Exchange Theory

2018 | European Conference On Information Systems | Citations: 0

Authors: Van den Broek, Tijs; Litvak, Nelly; Visser, Mike; Need, Ariana

Abstract: Recent studies show a significant effect of online communication on campaign don ... Expand

Abstract: Recent studies show a significant effect of online communication on campaign donations. Social exchange theory has been used in the literature to explain fundraisers motivatio. A limitation, however, is that motivation and the relation between online and offline actions are often derived from questionnaires that may suffer from selection and self-report biases, are often cross-sectional, do not scale-up and are often too costly to conduct. In an attempt to decrease such limitation, our study models the motivation of participants in an online campaign with social media analysis. We present a preliminary Satiation-Deprivation model, based on social exchange theory, that estimates fundraisers’ motivation levels over time based on a Twitter mention network and a donation dataset obtained from the Movember 2014 campaign. Despite our limited sample size, we find that motivation levels may serve as predictors for likelihood of donations and may improve baseline predictors such as co-occurrences of mentions and donations. After further validation, our preliminary model may inform strategic mentioning policies by campaign organizers, in order to maintain high motivation levels in their on-line community. Our findings suggest that a quantitative approach may serve as an accessible and valuable tool for predicting effectiveness of donation campaigns in the future. Collapse

Topics: Twitter online community application programming interface knowledge sharing social media

Methods: experiment simulation

Theories: social exchange theory