The Evolution of Organizations: Suggestions from Complexity Theory About the Interplay Between Natural Selection and Adaptation

There has been much debate in the managementliterature between neo-Darwinists (who believe in thenatural selection of populations of organizations) andadaptationists (who contend that changes in organization structure and behavior occur in response to theenvironment). The general thesis of neo-Darwinism isthat species are blindly selected for survival by theenvironment. The latest empirical support for the dominant neo-Darwinism perspective adopted bymost biologists is based primarily on the experimentsconducted by Salvador Luria who claims to haveconclusively demonstrated that genes mutate randomly.Recently, however, biologists have re-examined Luria sresearch methods and, after replications of hisexperiments, now question some aspects of the validityof his results. Moreover, there is now new researchwhich provides support for the earlier adaptationistposition, namely, the existence of evolutionary driversand directors existing within self-organizing systems.Of particular importance to the present study is the experimental indication thatself-organizing systems play a conscious role in theirown evolution. We propose that similar mechanisms orprocesses operate in organizational adaptation, thuspointing toward a theoretical modification ofneo-Darwinism that embraces both adaptation and naturalselection in a general, unified theory.     

Reflections on facilitating a participatory community self-evaluation

This paper is a reflective essay about our experience in conducting a participatory, community self-evaluation of a neighborhood revitalization effort in Milwaukee, Wisconsin. We took a controversial approach to evaluation by focusing on learning community dialogues as a vehicle for program evaluation. As we implemented our strategy, two sets of contradictions surfaced. The first points to the fundamental differences between conventional and constructivist research paradigms. The second centers around our focus on dialogue as a strategy for evaluation. In response to these contradictions we performed three complementary roles: program evaluator, process consultant, and learning facilitator. We are not implying that we performed an outstanding job in these roles. Rather, this article is our way of taking stock of what we learned about our practice as evaluators and how we may expand our repertoire of skills in the future.     

Improvement of Highway Safety I: Identification of Causal Factors Through Fault-Tree Modeling1

As the first article of a two-part series, the purpose of this paper is to examine the functional factors that contribute to automobile accident occurrence and to model the causation structure in the form of a fault-tree. The fault-tree model provides an intuitive framework for qualitatively decomposing possible pathways to accident occurrence. Fault-tree analysis also provides a statistical representation of how interacting driver, vehicle, and environmental factors contribute to the likelihood of automobile accident occurrence. The application of this model facilitates pinpointing those factors that most contribute to accident causation and subsequently enables the identification and comparison of potential crash avoidance technologies.     

BAYESIAN HIDDEN MARKOV MODELS FOR LONGITUDINAL COUNTS

This paper describes a Bayesian approach to modelling carcinogenity in animal studies where the data consist of counts of the number of tumours present over time. It compares two autoregressive hidden Markov models. One of them models the transitions between three latent states: an inactive transient state, a multiplying state for increasing counts and a reducing state for decreasing counts. The second model introduces a fourth tied state to describe non‐zero observations that are neither increasing nor decreasing. Both these models can model the length of stay upon entry of a state. A discrete constant hazards waiting time distribution is used to model the time to onset of tumour growth. Our models describe between‐animal‐variability by a single hierarchy of random effects and the within‐animal variation by first‐order serial dependence. They can be extended to higher‐order serial dependence and multi‐level hierarchies. Analysis of data from animal experiments comparing the influence of two genes leads to conclusions that differ from those of Dunson (2000). The observed data likelihood defines an information criterion to assess the predictive properties of the three‐ and four‐state models. The deviance information criterion is appropriately defined for discrete parameters.