Doctor of Philosophy / Department of Agricultural Economics / Jason S. Bergtold / Timothy J. Dalton / Adoption of technological improvements are crucial to increase agricultural productiviy to help reduce poverty by obtaining higher farm incomes due to higher productivity and lower production costs. However, the introduction of new agricultural technologies has not always been successful or had diffuse adoption. Factors that determine farmers’ adoption decisions are: 1) farm and farmers' characteristics; 2) technology attributes, and 3) the farming objective. Understanding these factors and how they affect adoption of new technologies on the farm is crucial to assure higher levels of adoption. The over all purpose of this thesis is to explore the adoption process of new technologies and practices by farmers. This is accomplished through three essays to meet the objectives of the thesis. The purpose of the first essay was to evaluate whether or not farmers in the western U.S. are willing to grow specialized oilseed crops that could be used for certified hydrotreated renewable jet (HRJ) fuel production and incorporate them into existing wheat-based production systems under contract. Results indicate that providing oilseeds crops and contracts with desired attributes and features would positively affect farmers' decisions to incorporate oilseed crops into their rotation system. Preferred seed and contract attributes that may affect a farmer’ adoption decision differ across different geographic regions of the U.S. The second essay focused on identifying factors that impact participation and farmers' decision to adopt soil conservation and fertilization management practices for cassava producers in Thailand and Vietnam. Results indicate that asset ownership and cassava yield positively influence participation. Adoption of new practices was positively linked to farmers’ participation in training activities, use of fish ponds (as a measure of alternative agricultural practices), presence of a nearby starch factory, and slope of the land. Finally, the purpose of the third essay was to examine extension educators' characteristics that affect educators' selection decision of outreach methods in the U.S. This essay examines the diffusion process that impacts adoption of best management practices by farmers. The decision extension educators make for selecting a teaching method is affected by the relationship between the objectives of the learning process and the characteristics of the teaching method.

The eCommerce industry introduced new business principles, as well as new strategies for achieving these principles, and as a result some traditional measures of success are no longer valid. We classified and ranked the performance of twenty business-to-consumer eCommerce companies by developing critical benchmarks using the Balanced scorecard methodology. We applied a Latent class model, a statistical model along the Bayesian framework, to facilitate the determination of the best and worst performing companies. An eCommerce site's greatest asset is its customers, which is why some of the most valued and sophisticated metrics used today evolve around customer behavior. The results from our classification and ranking procedure showed that companies that ranked high overall also ranked comparatively well in the customer analysis ranking, For example, Amazon.com, one of the highest rated eCommerce companies with a large customer base ranked second in the critical benchmark developed towards measuring customer analysis. The results from our simulation also showed that the Latent class model is a good fit for the classification procedure, and it has a high classification rate for the worst and best performing companies. The resulting work offers a practical tool with the ability to identify profitable investment opportunities for financial managers and analysts.

Choice-Based Conjoint (CBC) models are often used for pricing decisions, especially when scanner data models cannot be applied. Up to date, it is unclear how Choice-Based Conjoint (CBC) models perform in terms of forecasting real-world shop data. In this contribution, we measure the performance of a Latent Class CBC model not by means of an experimental hold-out sample but via aggregate scanner data. We find that the CBC model does not accurately predict real-world market shares, thus leading to wrong pricing decisions. In order to improve its forecasting performance, we propose a correction scheme based on scanner data. Our empirical analysis shows that the hybrid method improves the performance measures considerably. (author's abstract) / Series: Report Series SFB "Adaptive Information Systems and Modelling in Economics and Management Science"

Developing sound or reliable statistical models for analyzing vehicle crashes is very important in highway safety studies. A difficulty arises when crash data exhibit overdispersion. Over-dispersion caused by unobserved heterogeneity is a serious problem and has been addressed in a variety ways within the negative binomial (NB) modeling framework. However, the true factors that affect heterogeneity are often unknown to researchers, and failure to accommodate such heterogeneity in the model can undermine the validity of the empirical results. Given the limitations of the NB regression model for addressing over-dispersion of crash data due to heterogeneity, this research examined an alternative model formulation that could be used for capturing heterogeneity through the use of finite mixture regression models. A Finite mixture of Poisson or NB regression models is especially useful when the count data were generated from a heterogeneous population. To evaluate these models, Poisson and NB mixture models were estimated using both simulated and empirical crash datasets, and the results were compared to those from a single NB regression model. For model parameter estimation, a Bayesian approach was adopted, since it provides much richer inference than the maximum likelihood approach. Using simulated datasets, it was shown that the single NB model is biased if the underlying cause of heterogeneity is due to the existence of multiple counting processes. The implications could be poor prediction performance and poor interpretation. Using two empirical datasets, the results demonstrated that a two-component finite mixture of NB regression models (FMNB-2) was quite enough to characterize the uncertainty about the crash occurrence, and it provided more opportunities for interpretation of the dataset which are not available from the standard NB model. Based on the models from the empirical dataset (i.e., FMNB-2 and NB models), their relative performances were also examined in terms of hotspot identification and accident modification factors. Finally, using a simulation study, bias properties of the posterior summary statistics for dispersion parameters in FMNB-2 model were characterized, and the guidelines on the choice of priors and the summary statistics to use were presented for different sample sizes and sample-mean values.

No description available.

The Virginia Guidelines for Uniform Performance Standards and Evaluation Criteria for Teachers represented a significant overhaul of conventional teacher evaluation criteria in Virginia. The policy outlined seven performance standards by which all Virginia teachers would be evaluated. This study explored the application of cognitive diagnostic modeling to measure teachers' perceptions of intra-organizational mechanisms available to support educational professionals in implementing this policy. It was found that a coarse-grained, four-attribute compensatory, re-parameterized unified model (C-RUM) fit teacher perception data better and had lower standard errors than the competing finer-grained models. The Q-matrix accounted for the complex loadings of items to the four theoretically and empirically driven mechanisms of implementation support including characteristics of the policy, teachers, leadership, and the organization. The mechanisms were positively, significantly, and moderately correlated which suggested that each mechanism captured a different, yet related, component of policy implementation support. The diagnostic profile estimates indicated that the majority of teachers perceived support on items relating to "characteristics of teachers." Moreover, almost 60% of teachers were estimated to belong to profiles with perceived support on "characteristics of the policy." Finally, multiple group multinomial log-linear models (Xu and Von Davier, 2008) were used to analyze the data across subjects, grade levels, and career status. There was lower perceived support by STEM teachers than non-STEM teachers who have the same profile, suggesting that STEM teachers required differential support than non-STEM teachers. The precise diagnostic feedback on the implementation process provided by this application of diagnostic models will be beneficial to policy makers and educational leaders. Specifically, they will be better prepared to identify strengths and weaknesses and target resources for a more efficient, and potentially more effective, policy implementation process. It is assumed that when equipped with more precise diagnostic feedback, policy makers and school leaders may be able to more confidently engage in empirical decision making, especially in regards to targeting resources for short-term and long-term organizational goals subsumed within the policy implementation initiative. / Ph. D.

No description available.

Mapping of mortality rates has been a valuable public health tool. We describe novel Bayesian methods for constructing maps which do not depend on a post stratification of the estimated rates. We also construct posterior modal maps rather than posterior mean maps. Our methods are illustrated using mortality data from chronic obstructive pulmonary diseases (COPD) in the continental United States. Poisson regression models have attracted much attention in the scientific community for their superiority in modeling rare events (including mortality counts from COPD). Christiansen and Morris (JASA 1997) described a hierarchical Bayesian model for heterogeneous Poisson counts under the exchangeability assumption. We extend this model to include latent classes (groups of similar Poisson rates unknown to an investigator). Also, it is standard practice to construct maps using quantiles (e.g., quintiles) of the estimated mortality rates. For example, based on quintiles, the mortality rates are cut into 5 equal size groups, each containing $20\%$ of the data, and a different color is applied to each of them on the map. A potential problem is that, this method assumes an equal number of data in each group, but this is often not the case. The latent class model produces a method to construct maps without using quantiles, providing a more natural representation of the colors. Typically, for rare events, the posterior densities of the rates are skewed, making the posterior mean map inappropriate and inaccurate. Thus, although it is standard practice to present the posterior mean maps, we also develop a method to provide the joint posterior modal map (i.e., the map with the highest posterior probability over the ensemble). For the COPD data, collected 1988-1992 over 798 health service areas, we use Markov chain Monte Carlo methods to fit the model, and an output analysis is used to construct the new maps.

No description available.

Rift Valley fever (RVF) is a mosquito-borne zoonotic disease caused by a virus of the family Bunyaviridae, genus Phlebovirus. It is responsible for extensive outbreaks of disease in livestock in Africa with significant mortality and economic impact. Virus neutralization is considered the gold standard for confirming Rift Valley fever virus (RVFV) infection but the procedure is time consuming and expensive. Real-time reverse transcription-polymerase chain reaction (rRT-PCR), histopathology, and immunohistochemistry (IHC) are the diagnostic methods most often used in South Africa to confirm or exclude a diagnosis of RVF in necropsied animals. Validated estimates of diagnostic accuracy of these tests, in naturally infected livestock, however, have not been published. The objective of this study was to estimate the diagnostic sensitivity and specificity of rRT-PCR, histopathology, and IHC using Bayesian latent class methods in the absence of a gold standard. A secondary objective was to estimate stratum-specific values based on species, age, degree of specimen autolysis, and the presence/absence of tissue pigments. The Sensitivity (Se) and Specificity (Sp) of qRT-PCR were 97.4% (95% credibility interval (CI): 95.2% - 98.8%) and 71.7% (95% CI: 65% - 77.9%) respectively. The extraordinary analytical sensitivity of PCR makes this test very susceptible to false positive reactions, and thus reduced specificity. This is more likely during large-scale epidemics due to crosscontamination of specimens at necropsy facilities or testing laboratories. The Se and Sp of histopathology were 94.6% (95% CI: 91% - 97.2%) and 92.3% (95% CI: 87.6% - 95.8%) respectively. Single cases of RVF could be confused with acute poisoning with plants, bacterial septicaemias, and viral diseases such as infectious bovine rhinotracheitis and Wesselsbron disease. Most of these conditions, however, can be excluded using histological examination of the liver, special stains, bacterial culture, and toxicological or serological investigations. The Se and Sp of IHC were 97.6% (95% CI: 93.9% - 99.8%) and 99.4% (95% CI: 96.9% - 100%) respectively. Immunohistochemistry is highly specific because characteristic positive immunolabelling of the cytoplasm of hepatocytes can be correlated with the presence of hepatocellular injury typical for RVFV infection. False negative results are sometimes obtained with IHC because of reader error or loss of the antigenic epitopes due to advanced autolysis. Scant positive immunolabelling might be missed or viral proteins might be absent from sections of liver with advanced hepatocellular damage. The stratified analysis suggested differences in test accuracy in foetuses and severely autolysed specimens. The Sp of histopathology in foetuses (83.0%) was 9.3% lower than the value obtained for the sample population (92.3%). Lesions in some foetuses are more subtle and the typical eosinophilic intranuclear inclusions are often difficult to detect. In severely autolysed specimens, the Se of IHC decreased by 16.1% and the Sp of rRT-PCR by 17.4%. There is no plausible biological explanation for this decrease in the Sp of rRTPCR since the RNA of RVFV is resistant to degradation in autolysed tissues. Conversely, the antibody used to detect RVFV using IHC detects epitopes raised against nucleoproteins of the virus and it is possible that viral proteins become too widely dispersed and/or degraded in autolysed tissues to detect by light microscopy. It is possible that the marked decrease in Se of histopathology and IHC in severely autolysed specimens caused an apparent decrease in Sp of rRT-PCR, due to the latent class method. In conclusion, the high estimated Sp (99.4%) of IHC and the low Sp of rRT-PCR (71.3%) suggests that the definitive diagnosis or exclusion of RVF should not rely on a single PCR test and that IHC would be an effective confirmatory test for rRT-PCR positive field cases necropsied during an epidemic. Immunohistochemistry results from severely autolysed specimens, however, should be interpreted with caution and aborted foetuses in areas endemic for RVF should be screened using a variety of tests. The diagnostic Se and Sp of histopathology was much higher than expected confirming the value of routine post mortem examinations and histopathology of liver specimens. The most feasible RVF testing option in areas that do not have suitably equipped PCR laboratories, and where disease is often not detected in livestock until after human cases have been diagnosed, would be routine histopathology screening with IHC confirmation. Key Words: Rift Valley fever; Rift Valley fever virus; Bayesian; latent-class model; real-time reverse transcription-polymerase chain reaction; immunohistochemistry; histopathology; diagnosis; sensitivity; specificity. / Dissertation (MSc)--University of Pretoria, 2014. / gm2014 / Paraclinical Sciences / unrestricted