Survival Analysis of Organizational Networks -- An Exploratory Study

TL;DR

This study analyzes how various organizational factors influence the survival of organizational networks using survival analysis techniques, including Kaplan-Meier estimates and accelerated failure time models.

Contribution

It applies advanced statistical survival analysis methods to organizational network data, addressing the impact of multiple prognostic factors on network longevity.

Findings

Certain organizational and collaboration strategies significantly affect network survival.

Accelerated failure time models provide insights when proportional hazards assumptions are violated.

The study offers a methodological framework for analyzing organizational survival data.

Abstract

Organizations interact with the environment and with other organizations, and these interactions constitute an important way of learning and evolution. To overcome the problems that they face during their existence, organizations must certainly adopt survival strategies, both individually and in group. The aim of this study is to evaluate the effect of a set of prognostic factors (organizational, size, collaborate strategies, etc.) in the survival of organizational networks. Statistical methods for time to event data were used to analyze the data. We have used the Kaplan-Meier product-limit method to compute and plot estimates of survival, while hypothesis tests were used to compare survival times across several groups. Regression models were used to study the effect of continuous predictors as well as to test multiple predictors at once. Since violations of the proportional hazards were found for several predictors, accelerated failure time models were used to study the effect of explanatory variables on network survival.