At an increasing rate, individual investors are taking personal control over their financial destinies by investing their money online. Compared to offline do-it-yourself approaches, evidence suggests that investors exhibit lofty expectations and perform significantly worse after going online. However, little is understood about the mechanisms fueling expectancies, the role technologies play in their formation, or how technologies shape investment decisions. Therefore, this article explores the paradoxical nature of online investing technologies, which can give rise to a heightened state of conviction in one's capability to invest successfully. Drawing on Social Cognitive Theory, the concepts of encapsulation and combination are introduced to develop a research model describing how functional and technical self-efficacy judgments independently and collectively shape and influence outcome expectancies. The results suggest that perceptions about what one can accomplish through online investing technologies can lead investors to exaggerate their capabilities, which, in turn, produces elevated expectancies of financial payoffs and nonmonetary rewards. These findings carry important implications. In tasks requiring both computing and functional skills, the principals of encapsulation and combination highlight the importance of comprehensively capturing self-efficacy * This article is dedicated in memory of the first author's father, Arlen B. Looney, Jr., who recently lost his long and heroic battle with cancer. We thank the College of Business and Economics at Washington State University and the McIntire School of Commerce at the University of Virginia, who provided resources and funding for this research. We sincerely appreciate the insightful comments of the editors and reviewers on earlier versions of this manuscript. † Corresponding author. 206Paradoxes of Online Investing beliefs across skill domain boundaries. Moreover, online investing represents a paradoxical case that challenges the traditional assumption that fostering a robust sense of efficacy represents a purely noble enterprise. In fact, strong self-efficacy beliefs can prove counterproductive, leading to severe, irreversible, and unintended consequences. Going forward, these discoveries provide a solid foundation to enhance systems designs and facilitate a deeper understanding of user psychology.

Assigning effective teachers to introductory IS courses represents one intervention strategy that has been broadly advocated to help reverse the sharp decline in students majoring in Information Systems (IS). Using a survey of 305 students enrolled in a multiple-section introductory IS course, this study empirically confirms that students who are taught by effective teachers are more likely to be attracted to the IS discipline. Moreover, based on a robust theoretical foundation grounded in Social Cognitive Theory, the findings reveal the underlying mechanisms through which teaching effectiveness influences students' aspirations to pursue an IS degree. Specifically, teaching effectiveness bolsters students' confidence in their ability to successfully perform as IS majors (i.e., self-efficacy), raises students' expectations that valued rewards will be received by majoring in IS (i.e., outcome expectations), and helps students develop enduring interest in the IS field. In addition to teaching effectiveness, students with high self-efficacy and robust outcome expectations become more interested in IS. In turn, interest serves as the primary channel through which goals to choose the IS major develop. Implications for theory and practice are discussed.

As firms continue to abandon pensions in favor of employee-managed retirement plans, tremendous demands are being placed on the decision-making proficiency of future retirees. As reflected in the equity premium puzzle, individual investors tend to hold overly conservative portfolios that provide meager payoffs over time. Consequently, there is growing concern that the vast majority of retirement accounts might be insufficiently funded when employees reach retirement. Given that most retirement plans can now be managed online, a potential solution lies in designing a Web-based decision support system (DSS) that helps future retirees make more-profitable portfolio management decisions. This paper reports the results of a study in which 159 retirement plan participants were asked to use an experimental website to manage a portfolio of retirement investments over a simulated 30-year period. Using a psychological approach toward designing the DSS, myopic loss aversion is put forth as a theoretical explanation for the psychological mechanisms that encourage investors to hold overly conservative portfolios. Armed with this knowledge, three design features--information horizon, system restrictiveness, and decisional guidance--are implemented as part of an overarching design strategy targeted at increasing investors' willingness to take calculated risks. The results indicate that investor conservatism diminishes when the DSS presents prospective probabilities and payoffs over long time horizons. In contrast, short-term information horizons constitute a major stumbling block for investors. However, when confronted with short-term information horizons, risk aversion can be successfully counteracted by configuring a DSS to either restrict the frequency of decisions or to suggest a relatively aggressive portfolio allocation. These findings carry important implications for theory and practice.decision support systems, decisional guidance, equity premium puzzle, field experiment, human-computer interaction, information horizon, myopic loss aversion, retirement planning, system restrictiveness

No abstract

A computer-based group idea-generation simulator was used to measure the performance of individuals with high and low cognitive ability who received high-or low-quality stimulation; high-and low-quality stimulation was operationalized by presenting controlled idea submissions from simulated group members throughout the idea-generation session. Cognitive ability and stimuli quality interacted such that the high cognitive ability/high-quality stimuli treatment achieved the highest performance. The results suggest that the performance of individuals with high cognitive ability can be enhanced when given high-quality stimuli or be inhibited when exposed to lowquality stimuli. The performance of individuals with low cognitive ability, in contrast, is consistently meager regardless of stimuli quality. The findings suggest that group composition cannot only significantly influence the overall ideational performance of a group, but also that of individual group members. The implications of these findings for future research, as well as the implications for the design of group idea-generation procedures, are discussed.