Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: In 1996 the state of Georgia significantly altered its civil service system. One aspect of the reform mandated that all employees hired or promoted after July 1 of that year would be included in the state's unclassified service. As such, these employees would have very limited job protections and would serve in effect on an "at will" basis. Employees hired earlier would continue in the traditional classified service unless they were promoted. Classified employees enjoy an array of job protections and are entitled to due process procedures prior to termination. An examination of the attitudes of both groups of employees toward the Georgia reforms reveals generally pessimistic views. However, unclassified employees are found to be significantly more supportive and positive about the reforms than are their classified colleagues.

Publication Type: Conference Paper/Unpublished Manuscript
Abstract: We develop a framework and tools for applying a computer‐assisted context analysis system and find that it achieves levels of accuracy comparable to humans for about 80% less effort when starting from scratch (no labeled examples). The system is presented using a case study of Congressional bill titles as a proxy for the full text of Congressional bills. We also demonstrate that the system can use information learned from previous experiments to reduce the labeling requirements still further to over 90% savings of the current human effort.

This study assumes that social scientists have a need to locate individual documents in a subject area. To support this need, "Topic classification," where documents are coded according to some organizing framework, is used to facilitate search and summarization. Our proposed framework for effectively employing machine learning methods mitigates the high costs of the standard method of topic classification - human labeling. We scientifically evaluate the efficacy and accuracy of the automated approach using a large corpus of 380,000 human-labeled events and a classification system that includes 20 major policy topics, 226 subtopics, and a demonstrably strong level of human inter-coder agreement.

Publication Type: Oral Presentation
Abstract: Improving nitrogen (N) management practices for corn is important for increasing profitability and reducing the environmental pollution. The need for better N management within individual fields is also driven by the possibility to apply variable rates of N. However, there is a lack of reliable information about the magnitude and frequency of yield response to N fertilizer on the scale of individual fields and farms.

The purpose of this study was to characterize and classify spatial variability in yield response to incremental increases in N fertilizer.

Fertilizer N was applied at five rates in replicated strips that entirely covered six fields during two years. Yield responses to N were measured in 22-25 individual trials per acre in grid patterns. Distributions of yield responses to N in individual trials were used to characterize variability in yield response and calculate probability of yield response and profit to fertilizer. The yield responses were classified into response categories with different fertilizer requirements by using soil map units, soil series, apparent soil electrical conductivity, and relative elevation. Analysis showed the classification explained a relatively small percentage variability in yield response to N for a combined sample of two years.


The task of classifying variability in yield responses into different response categories was found to be difficult due to interactions of soil properties and weather. This task, however, is manageable when a group of producers are determined to collect site-specific information of yield responses to N on their farms over several years across different management practices.

Publication Type: Symposium Paper
Abstract: Adolescents know less than adults about police questioning and interrogation. From an intervention perspective, it would be helpful to In this paper we use Latent Class Analysis to understand to what degree families understand, or fail to understand, basic components of rights and process. Three latent classes combined to classify families’ knowledge risk status, which was predicted by youth age, IQ, and gender, as well as interactions between youth age and youth justice system experience, youth race, parental IQ, and household socioeconomic status. Implications for identification and intervention with youthful defendants and their parents are discussed.

Publication Type: Conference Paper/Unpublished Manuscript
Abstract: The criminalization of terrorism has become a divisive issue over the last decade. Events such as the Oklahoma City Bombing on April 19, 1995, and the incidents of September 11, 2001, have spawned the debate over the how to incorporate terrorism into a crime agenda for local municipalities. On the one hand, some scholars believe terrorism should be included in everyday crime fighting, while other scholars believe terrorism should be treated as an independent problem. This paper will discuss how the elements of crime and terrorism parallel and diverge through analyzing definition and theory. Policy implications will also be presented.