Most studies on perceptions of social structures in organizations rely on cross-sectional evidence and lack a longitudinal perspective. In order to address this gap, we collected whole network perception data at three time points from a cohort of MBA students. First, we asked whether or not individuals become more accurate in their perception of the network over time. We found no significant increase in accuracy. Second, we examined one’s perception of his or her own direct ties and found a consistent tendency to inflate incoming friendship ties, confirming existing studies. However, we find that individuals were quite capable of recognizing the broader dynamics of social hierarchy (i.e., whether they were becoming more or less popular) even as they became no more accurate in understanding either the overall networks or their own ego-net. Third, we explored possible explanations for the persistence of perception errors and showed that most of the errors at time point two and time point three were due to a failure to update previous perception decisions. Finally, we shifted the analysis from accuracy at a given time point and considered the narrative arc of dyadic relations. Our findings suggest that stable dyads across time are more likely to be accurately perceived whereas other types of dyads are poorly tracked. We conclude by presenting possible research questions for future studies to further our understanding of the temporal aspects of network perception.

Increased consumption of fossil fuels in industrial production has led to a significant elevation in the emission of greenhouse gases and to global warming. The most effective international action against global warming is the Kyoto Protocol, which aims to reduce carbon emissions to desired levels in a certain time span. Carbon trading is one of the mechanisms used to achieve the desired reductions.

In survival analysis or reliability and maintenance studies, hazard rate function plays an important role, as it describes the instantaneous failure probability given that the individual survived or the item has not failed up to that instance. Sudden changes in the hazard function may occur due to overhauls, treatment effects, or maintenance activities. A problem of statistical interest is the estimation of the times of the changes, as well as their sizes. It is also of interest to study the impact of covariates on the hazard rate function. In the article, we review the basic estimation methods for hazard change point models for complete and censored data. We also discuss extensions that allow for multiple change points and provide a brief exposure to Bayesian approach.

No abstract