Innovating in Digital Government in the American States

Objective. The purpose of this research is to examine why some states have embraced digital government more extensively than others. Methods. Multivariate regression analysis is used to empirically test explanations for state innovation in e–government. The primary dependent variable is the percent of state–level government websites offering online services to citizens. Results. Republican–controlled legislatures are more likely to embrace e–government, implying that efficiency concerns may drive reliance on digital government. Innovators in e–government were states with fewer households with Internet access and less use of the initiative process, indicating that citizen demand was not a factor. More extensive use of e–government is also associated with legislative professionalization and professional networks—factors that may be useful for explaining the diffusion of other administrative reforms, and technical issues lacking political salience.
Conclusions. These data suggest e–government implementation is driven by legislative professionalism and, to a lesser extent, state professional networks, rather than citizen demand. These indicators fit Lowi's (1972) conception of "constituent policy" as a top–down process, confined to administrative or legislative circles, compared to distributive, regulatory, and redistributive policy. We hypothesize that other administrative reforms, particularly those lacking political salience, may exhibit similar relationships with legislative professionalization and professional networks.     

From Freud to Feminism: Gendered Constructions of Blame Across Theories of Child Sexual Abuse

Most theories of child sexual abuse are, to some degree, gendered, with nonoffending mothers bearing the burden of blame, ideologically and legally, for the transgressions of predominantly male offenders. This article explores the social construction of blame for child sexual abuse via critical analyses of evolving theoretical perspectives on maternal culpability for the inception and maintenance of abuse dynamics. Drawing on selected conceptual and research knowledge that supports and refutes anecdotal claims, this synthesis of the literature culminates in the proposal of an evidence-informed, feminist-grounded, multitheoretical child sexual abuse framework that disrupts dominant mother-blaming discourse and guides socially just and ethically responsive policy, practice, and research.     

Indigenous perspectives for strengthening social responses to global environmental changes: A response to the social work grand challenge on environmental change

ABSTRACT

The “Grand Challenges for Social Work,” is a call to action for innovative responses to society’s most pressing social problems. In this article, we respond to the “Grand Challenge” of Creating Social Responses to a Changing Environment from our perspective as Indigenous scholars. Over the last several decades, diminishing natural resources, pollution, over-consumption, and the exploitation of the natural environment have led to climate change events that disproportionately affect Indigenous peoples. We present how environmental changes impact Indigenous peoples and suggest culturally relevant responses for working with Indigenous communities. We propose a decolonizing cyclical, iterative process grounded in Indigenous Ways of Knowing.     

A bayesian predictive approach to the selection of variables in multiple regression

The squared error loss function applied to Bayesian predictive distributions is investigated as a variable selection criterion in linear regression equations. It is illustrated that “cost-free” variables may be eliminated if they are poor predictors. Regression models where the predictors are fixed and where they are stochastic are both considered. An empirical examination of the criterion and a comparison with other techniques are presented.     

A bayesian analysis of imperfect inspection models

An examination of sampling inspection models is provided for the case where inspection is not perfect and classification errors can be made. The conjugate family of distributions is obtained under t h e assumption that defective items are generated according to a Bernoulli process. To simplify analysis, the use of a single beta prior distribution is considered. Relevant predictive distributions are obtained.     

Against the odds: the impact of woman abuse on maternal response to disclosure of child sexual abuse

Although the co-occurrence of woman abuse and child sexual abuse is high little research exists exploring the impact of woman abuse on maternal response to child sexual abuse (CSA). Findings from two qualitative studies indicate the form of woman abuse to have differential impact on maternal response. Mothers who were abused in non-physical ways, psychological, or emotional, displayed more ambivalent, less supportive responses. Mothers who were physically abused by their partners more often acted supportively to their children and displayed less ambivalence about separating from the perpetrator. Some survivors reported non-disclosure as children because of their fear of the perpetrator due to the woman abuse they were exposed to while being sexually abused. Practice and research implications are discussed.