Texas Department of Health organizational structure changes: A reflection of public health policy and agency expansion, September 1, 1946--June 30, 1994

The premise of this study is that changes in the agency's organizational structure reflect changes in government public health policy. Based on this premise, this study tracks the changes in the organizational structure and the overall expansion of the Texas Department of Health to understand the evolution of changing public health priorities in state policy from September 1, 1946 through June 30, 1994, a period of growth and new responsibilities. It includes thirty-seven observations of organizational structure as depicted by organizational charts of the agency and/or adapted from public documents. ^ The major questions answered are, what are the changes in the organizational structure, why did they occur and, what are the policy priorities reflected in these changes in and across the various time periods. ^ The analysis of the study included a thorough review of the organizational structure of the agency for the time-span of the study, the formulation of the criteria to be used in ascertaining the changes, the delineation of the changes in the organizational structure and comparison of the observations sequentially to characterize the change, the discovery of reasons for the structural changes (financial, statutory - federal and state, social and political factors), and the determination of policy priorities for each time period and their relation to the expansion and evolution of the agency. ^ The premise that the organizational structure of the agency and the changes over time reflect government public health policy and agency expansion was found to be true. ^

A Bayesian approach to estimating the regression coefficients of a multinomial logit model

A Bayesian approach to estimation of the regression coefficients of a multinominal logit model with ordinal scale response categories is presented. A Monte Carlo method is used to construct the posterior distribution of the link function. The link function is treated as an arbitrary scalar function. Then the Gauss-Markov theorem is used to determine a function of the link which produces a random vector of coefficients. The posterior distribution of the random vector of coefficients is used to estimate the regression coefficients. The method described is referred to as a Bayesian generalized least square (BGLS) analysis. Two cases involving multinominal logit models are described. Case I involves a cumulative logit model and Case II involves a proportional-odds model. All inferences about the coefficients for both cases are described in terms of the posterior distribution of the regression coefficients. The results from the BGLS method are compared to maximum likelihood estimates of the regression coefficients. The BGLS method avoids the nonlinear problems encountered when estimating the regression coefficients of a generalized linear model. The method is not complex or computationally intensive. The BGLS method offers several advantages over Bayesian approaches. ^