Main navigation | Main content
Recommender systems are becoming a salient part of many social media and e-commerce websites. Much research has focused on advancing recommendation technologies to improve accuracy of predictions, while behavioral aspects of using recommender systems are often overlooked. In this study, we conduct three experiments to explore how consumers' preferences at the time of consumption are affected by their interactions with recommender systems. The results provide strong evidence that the rating presented by a recommender system serves as an anchor for the consumer's constructed preference. The main finding of the study is that the consumers' preference ratings are malleable and can be significantly influenced by the recommendation received.
Demonstrating compelling causal evidence of the existence and strength of peer to peer influence in online social networks has become the holy grail of the modern research in online social networks. Among the most frequently cited explanations for this effect are peer influence and homophily, but separating these out from observational data is virtually impossible. In this paper, we present a novel randomized experiment that tests the existence of causal peer influence in the general population of a particular large-scale online social network. We conduct a randomized control trial on a user-base of 4 million listeners of Last.fm, a music listening social network, and show that, for a median social network user, the odds of adopting the paid subscription increase by 116% due to peer influence when her friend adopts it.
Online social networks are credited with inspiring political action, driving viral and word-of-mouth driven spread of products and services and playing a significant role in the diffusion of ideas. A primitive underlying this conjectured "influence" is a notion that people trust the opinions, intentions or actions of individuals they are connected to. We design a novel Facebook application to play a non-anonymous investment game that allows us to study the linkage between strength of social ties and trust. We estimate the marginal effects of three different strength of ties measures computed within Facebook. We find that each wall-post made on a friend's wall results in a 21% increase in trust. Similarly, we find that each photo two friends jointly appear in, a signal of social affinity and physical world ties, is associated with a 5.1% increase in trust. Interestingly, the unconditional correlation between the classic "number of common friends" metric – a widely used strength of ties measure - and trust is not significant.
People experience a lot of desires in everyday life, be they about basic needs such as food, drink, sex, and sleep, or about more 'modern cravings' such as with regard to smartphones, coffee, or other drugs. Oftentimes, our desires are quite unproblematic, but they can turn into temptations to the extent to which they conflict with important other goals in our lives. The field study (called the Everyday Temptation Study) we conducted was a first investigation into how desire and its control show up in people's everyday environments.
We used experience sampling to measure desires and desire regulation in everyday life. Over the course of a week, 205 adults furnished a total of 7,827 desire reports. Results revealed substantial domain-related differences in desire frequency, strength, and degree of conflict with other goals, as well as in the likelihood and success with which desires are resisted. Desires for sleep and sex were experienced most intensively, whereas desires for tobacco and alcohol had the lowest average desire strength despite being thought of as addictive. Desires for leisure and sleep conflicted the most with other goals, and desires for media and work brought about the most self-control failure.