Penn State Altoona is a commonwealth campus of The Pennsylvania State University and it is located in Logan Township, Pennsylvania. It is one of four full-fledged four year institutions in the Commonwealth Campus network. The full-time student count was 3,645 in 2015.
Although prior research has identified factors associated with bystander willingness to intervene in intimate partner violence (IPV), whether responses differ across heterosexual and same-sex relationships remains underexplored. Using a vignette design with 182 college students, this study examined whether perpetrator gender and relationship sexual orientation influenced awareness of violence, post-situation informal intervention, and willingness to report the incident to police. A significant perpetrator gender & times; relationship sexual orientation interaction emerged for police reporting. The interaction showed a crossover pattern: descriptively, reporting willingness was greater for male than female perpetrators in heterosexual vignettes, whereas the pattern reversed in same-sex vignettes. However, neither within-context perpetrator-gender comparison reached statistical significance in follow-up simple-effects analyses, indicating that the interaction should be interpreted cautiously.
Examining concern regarding opioids is highly salient given the risk such drugs present to the public. Relatedly, there is a need to understand the experiences of those who have family members struggling with substance use disorder (SUD), particularly opioid use disorder (OUD). Drawing on Kaiser Family Foundation (KFF) survey data, this study explores four research questions. First, what proportion of respondents believe that the U.S. is experiencing an opioid crisis? Second, what factors are associated with this concern? Third, what percent of respondents have family members who are dependent on substances, including prescription pain pills, heroin, and fentanyl? Fourth, what percentage of respondents have experienced worsening mental health, financial situations, and relationships due to addiction, and what variables are associated with these conditions? Most of the sample (53%) expressed concern about the opioid problem. Greater concern was evident across women, members of the LGBT community, lower income respondents, and those who have had familial experiences with SUD/OUD. Approximately 30% of the public had a loved one with SUD/OUD and about 5% reported a personal SUD/OUD history. A majority of participants with a history of self and/or familial drug addiction disclosed that this experience has had an adverse impact on their own mental health (70.7%), their family's financial situation (58.1%), and their relationship with their family (74.9%). The extent of these harms varies by specific factors, which may derive from stigma variation, the impact of differential criminalization, and inequities across service access, as well as socio-demographic characteristics. Implications are discussed.
The generative adversarial network (GAN) is an important model developed for high-dimensional distribution learning in recent years. However, there is a need for a comprehensive understanding of its generalization error and convergence behavior. In this research, we study the generalization error of GANs based on classes of functions encompassing discriminator and generator neural networks with piecewise linear activation functions. Under our assumptions, the resulting function classes are VC type with bounded envelope functions, enabling the application of Talagrand’s inequality. By employing Talagrand’s inequality and the Borel–Cantelli lemma, we establish almost sure convergence rates for the generalization error of GANs under fixed discriminator and generator architectures. We further apply the same empirical-process technique to several existing GAN error formulations and obtain corresponding almost sure convergence bounds. The resulting sample-size dependence is compared with existing results under the same fixed-architecture setting, while network-complexity quantities are treated consistently as fixed constants with respect to the sample sizes. In particular, under suitable range and symmetry conditions on the discriminator class, the error defined by the neural network distance is a special case of our general measuring-function objective.
Guided by traumagenic dynamics and evolutionary-developmental (e.g., life history) frameworks, this study examines whether developmental stage and sexual victimization group status (e.g., childhood, adolesces, adulthood, or multiple victimizations across developmental stages) are associated with different risk profiles (e.g., sexual attitudes and behaviors) using data from 932 women’s online self-report surveys. 344 (37%) women experienced unwanted vaginal penetration during their lives (32 childhood only, 78 adolescence only, 118 adulthood only, and 116 multiple stages). ANOVAs with Dunnet T3 posthoc tests and multinomial logistic regressions predicting victimization group revealed significant between-group differences for age of menarche, age of first consensual intercourse, mental health, sociosexuality (SOI-R), and posttraumatic sexual behavior (PTSBI-W). Results demonstrate the need for developmentally informed prevention and treatment policies.