David Hachen

The Competing Risks Model A Method for Analyzing Processes with Multiple Types of Events

Social scientists are often interested in modeling processes with multiple types of events. One frequently used model—the competing risk model—provides a method for analyzing the rates of different event types. This article addresses two important issues concerning the competing risks model. First, under what conditions is this model appropriate? Second, how should one interpret estimated parameters? Guidelines for answering these questions are provided by (1) contrasting the competing risks model to an alternative model in which one analyzes separately the overall hazard rate and conditional probabilities and (2) showing how different interpretations of estimated parameters are related to the issue of dependencies among the competing risks. An empirical analysis of the rates of within and between employer job shifts illustrates the issues involved in employing the competing risks model.

Three Models of Job Mobility in Labor Markets

This study examines job mobility patterns in the United States through the analysis of job mobility rates. Utilizing event history analysis and retrospective job history data, a variety of job mobility rates are modeled. Besides the overall rate of job exiting, this study also examines rates of voluntary and involuntary exits, within- and between-employer moves, and upward job shifts. Informed by three theoretical models of job mobility—the reward-resource model, the limited opportunity model, and the vacancy competition model—hypotheses concerning the effects of education, occupational status, supervision, time in the labor force, gender, race, and economic sector on various job shift rates are formulated and tested. Among the major conclusions are (a) education increases job mobility, but does not uniquely determine the paths; (b) gender and racial differences pertain more to differences in mobility channels than to differences in the overall rate of job moving; and (c) dualistic accounts of labor market structures are problematic in that they tend to conflate several basic dimensions of job mobility patterns.

INDUSTRIAL CHARACTERISTICS AND JOB MOBILITY RATES

Allfirms operate in at least two markets. the product market and the labor market. I investigate the relationship between characteristics of these two markets by examining the effects of product market (industrial) characteristics on job mobility in labor markets. Informed by a theoretical model in which employer personnel strategies mediate the relationship between industrial characteristics and job mobility rates, hypotheses are formulated concerning the effects of six industrial characteristics concentration levels, conglomerate domination, labor intensity, growth rates, wage levels, and average establishment size on four types of job mobility involuntary exits, quits, intra-firm moves, and upward authority moves. Among the findings are: (1) quit rates, intra-firm mobility rates, and upward authority mobility rates are low in high wage industries; (2) labor intensive industries have high involuntary exit rates and low interand intra-firm mobility rates; and (3) the effects of industry growth on job mobility vary depending on whether growth occurs through the emergence of new firms or the increasing size of existing firms.

Predictors of short-term decay of cell phone contacts in a large scale communication network☆☆☆

Under what conditions is an edge present in a social network at time t likely to decay or persist by some future time t + Delta(t)? Previous research addressing this issue suggests that the network range of the people involved in the edge, the extent to which the edge is embedded in a surrounding structure, and the age of the edge all play a role in edge decay. This paper uses weighted data from a large-scale social network built from cell-phone calls in an 8-week period to determine the importance of edge weight for the decay/persistence process. In particular, we study the relative predictive power of directed weight, embeddedness, newness, and range (measured as outdegree) with respect to edge decay and assess the effectiveness with which a simple decision tree and logistic regression classifier can accurately predict whether an edge that was active in one time period continues to be so in a future time period. We find that directed edge weight, weighted reciprocity and time-dependent measures of edge longevity are highly predictive of whether we classify an edge as persistent or decayed, relative to the other types of factors at the dyad and neighborhood level.

Enhanced Situational Awareness: Application of DDDAS Concepts to Emergency and Disaster Management

We describe a prototype emergency and disaster information system designed and implemented using DDDAS concepts. The system is designed to use real-time cell phone calling data from a geographical region, including calling activity --- who calls whom, call duration, services in use, and cell phone location information --- to provide enhanced situational awareness for managers in emergency operations centers (EOCs) during disaster events. Powered-on cell phones maintain contact with one or more within-range cell towers so as to receive incoming calls. Thus, location data about all phones in an area are available, either directly from GPS equipped phones, or by cell tower, cell sector, distance from tower and triangulation methods. This permits the cell phones of a geographical region to serve as an ad hoc mobile sensor net, measuring the movement and calling patterns of the population. A prototype system, WIPER, serves as a test bed to research open DDDAS design issues, including dynamic validation of simulations, algorithms to interpret high volume data streams, ensembles of simulations, runtime execution, middleware services, and experimentation frameworks [1].

Lessons learned from the netsense smartphone study

Over the past few years, smartphones have emerged as one of the most popular mechanisms for accessing content across the Internet driving considerable research to improve wireless performance. A key foundation for such research efforts is the proper understanding of user behavior. However, the gathering of live smartphone data at scale is often difficult and expensive. The focus of this paper is to explore the lessons learned from a two year study of two hundred smart phone users at the University of Notre Dame. In this paper, we offer commentary with regards to the entire process of the study covering aspects including funding considerations, technical architecture design, lessons learned, and recommendations for future efforts gathering live user data.

Attachments, Grievances, Resources, and Efficacy: The Determinants of Tenant Association Participation Among Public Housing Tenants

ABSTRACT: This study uses data from the Multi-City Study of Urban Inequality to examine variation in tenant association participation among public housing tenants in Boston and Los Angeles. Using logistic regression models we estimate the net effects of four sets of factors on the likelihood that a tenant has attended tenant association meetings: neighborhood attachments, grievances, resources and constraints, and feelings of efficacy. Results show that net of other factors, participation is greater among attached tenants who have resided in public housing longer and who have social ties to other people. Grievances also increase participation, but they do so indirectly by increasing people’s tendency to be more involved in their communities. With the exception of education’s positive effect, resources and constraints are not important determinants of participation. Education and efficacy act like enablers increasing people’s ability to be involved in their communities. The implications of the findings for research and community organizing are explored by examining how three mechanisms account for the findings.

Gender differences in job mobility rates in the united states

Abstract This research examines gender differences in job mobility rates, especially rates of movement into supervisory and managerial positions. Informed by human capital and segmentation theory, hypotheses concerning the effects of labor force withdrawals and gender-based occupational segregation on mobility are investigated. Using job history data and event history analysis a series of models of both voluntary moves and upward authority moves are specified and estimated. Results indicate that while labor force withdrawals and the percentage female in an occupation are related to these mobility rates, net of these factors, women continue to have lower rates of upward authority moves.

A dyadic reciprocity index for repeated interaction networks

A wide variety of networked systems in human societies are composed of repeated communications between actors. A dyadic relationship made up of repeated interactions may be reciprocal (both actors have the same probability of directing a communication attempt to the other) or non-reciprocal (one actor has a higher probability of initiating a communication attempt than the other). In this paper we propose a theoretically motivated index of reciprocity appropriate for networks formed from repeated interactions based on these probabilities. We go on to examine the distribution of reciprocity in a large-scale social network built from trace-logs of over a billion cell-phone communication events across millions of actors in a large industrialized country. We find that while most relationships tend toward reciprocity, a substantial minority of relationships exhibit large levels of non-reciprocity. This is puzzling because behavioral theories in social science predict that persons will selectively terminate non-reciprocal relationships, keeping only those that approach reciprocity. We point to two structural features of human communication behavior and relationship formation—the division of contacts into strong and weak ties and degree-based assortativity—that either help or hinder the ability of persons to obtain communicative balance in their relationships. We examine the extent to which deviations from reciprocity in the observed network are partially traceable to the operation of these countervailing tendencies.

Experiences measuring sleep and physical activity patterns across a large college cohort with fitbits

In the past few years, a wide variety of highly capable and inexpensive wearable health sensors have emerged. One of the interesting aspects of such sensors is the capability for researchers to longitudinally and automatically quantify important health behaviors, such as physical activity and sleep, with little intervention required by the participant. While the accuracy of these devices has been evaluated in laboratory settings, there exists little public data with respect to user compliance and the consistency of the resulting measurements at a large scale. The focus of this paper is to share our experience in distributing five hundred Fitbit Charge HR devices across a group of college freshmen and to introduce the resulting dataset from our study, the NetHealth Study. We find that when users are compliant, they tend to be exceptionally so, having an average compliance of 86%. User non-compliance does play a role, however, reducing the overall average compliance rate to 67%. We discuss various reasons for non-compliance and also briefly highlight preliminary monitored characteristics of physical activity and sleep in our student population.