No description available.

No description available.

No description available.

No description available.

No description available.

Bakgrund: Könsdiskriminering påverkar kvinnors liv, ekonomi och hälsa, över hela världen. Hälso- och sjukvården har en etisk obligation att likvärdigt behandla patienter och framförallt sjuksköterskan via dess koder och professionalitet. Syfte: Syftet med studien är att undersöka förekomsten av könsskillnader inom hälso- och sjukvård för kvinnor. Metod: Litteraturöversikt med induktiv ansats som utgår från ett feministiskt Standpoint Theory ramverk. Kvalitativa såsom kvantitativa artiklar används. Resultat: Olika nivåer och instanser av sjukvården präglas av könsskillnader som negativt påverkar kvinnor. Områden såsom smärta, medicinering, psykiatri och akademi har någon form av könsdiskriminering i sig. Konklusion: Orsaken till könsskillnaderna anses vara normativt tankesätt, misogyni och ett androcentriskt system på vården. Sjuksköterskor har en etisk och laglig plikt att förhindra negativa könsskillnader och bias för att bredda sitt helhetsvårdande perspektiv. / Background: Discrimination on the base of gender affects the economy, lives and health of women, all over the world. Health care as an institue possesses a legal and ethical obligation to deliver indiscriminate care and especially registered nurses. Aim: The aim of the study is to examine the presence of gender differences in health care of women. Method: A literature review—with the approach of induction—that works within the feminist Standpoint Theory’s framework. The study utilize both qualitative and quantiative data from other academic literature. Findings: Gender differences that negatively affect women was found to be present in a variety of settings and levels of health care. Examples of these areas were pain, mediciation, psychiatry and academia. Conclusion: The cause behind the results could be ascribed to normative thinking, mysoginy and systemic androcentrism. Registered nurses have an ethical duty to prevent gendered differences and biases that negatively impact women in health care. It would also broaden their holistic caring perspective.

The purpose of this study was to examine how race and gender bias influence music educators' perceptions of musical style and evaluations of brief jazz and classical piano performances. Previous research has shown that race and gender bias and stereotype activation influence our judgment of others. These factors could result in biased evaluations of musical performances, including ensemble auditions and college level juries. I constructed an instrument designed to test these biases by experimentally manipulating race and gender variables of jazz and classical performances. Videos of a black male, white male, white female, and black female pianist were synchronized with identical audio recordings to control for performer ability. The first experiment measured how stereotypes influence participants' proper identification of jazz and classical styles in a series of 2-second video clips. The second experiment measured how race and gender bias influence participants' evaluations of jazz and classical performances in a series of 10-second video clips. The participants in this study were a national sample of applied music faculty (n=315). Participants were randomly assigned to four test conditions in a 2x2 (performer race X performer gender) between subjects blind experimental design. The dependent variables were classical jury grade predictions, jazz jury grade predictions, and accuracy of style identification. Results of a 2x2 ANOVA revealed significant differences in style identification by gender and interaction of race and gender. Participants were more likely to associate female performers with classical music and the black male performer with jazz. There were also significant differences in classical jury grade predictions by race, and jazz jury grade predictions by the interaction of race and gender. The black male performer received the lowest average jury grade predictions in both jazz and classical performances, scoring between 0.5 and 1 letter grade lower than the other performers. Results suggest that a negative association of females and jazz music still exists, as well as a stereotype of a black male jazz performer. While females did not receive significantly lower jazz jury grade predictions than the male performers, they may still feel marginalized in college jazz programs and ensembles. The results also suggest that black males may be at a significant disadvantage in college music admissions, auditions, and juries. These results have serious implications for music educators at every level. We must strive for fair and equitable audition processes and ensure that every child, regardless of race or gender, has an equal opportunity to participate in ensembles and music programs. / Music Education

When this PhD started, the translation of speech into text in a different language was mainly tackled with a cascade of automatic speech recognition (ASR) and machine translation (MT) models, as the emerging direct speech translation (ST) models were not yet competitive. To close this gap, part of the PhD has been devoted to improving the quality of direct models, both in the simplified condition of test sets where the audio is split into well-formed sentences, and in the realistic condition in which the audio is automatically segmented. First, we investigated how to transfer knowledge from MT models trained on large corpora. Then, we defined encoder architectures that give different weights to the vectors in the input sequence, reflecting the variability of the amount of information over time in speech. Finally, we reduced the adverse effects caused by the suboptimal automatic audio segmentation in two ways: on one side, we created models robust to this condition; on the other, we enhanced the audio segmentation itself. The good results achieved in terms of overall translation quality allowed us to investigate specific behaviors of direct ST systems, which are crucial to satisfy real users’ needs. On one side, driven by the ethical goal of inclusive systems, we disclosed that established technical choices geared toward high general performance (statistical word segmentation of the target text, knowledge distillation from MT) cause an exacerbation of the gender representational disparities in the training data. Along this line of work, we proposed mitigation techniques that reduce the gender bias of ST models, and showed how gender-specific systems can be used to control the translation of gendered words related to the speakers, regardless of their vocal traits. On the other side, motivated by the practical needs of interpreters and translators, we evaluated the potential of direct ST systems in the “augmented translation” scenario, focusing on the translation and recognition of named entities (NEs). Along this line of work, we proposed solutions to cope with the major weakness of ST models (handling person names), and introduced direct models that jointly perform ST and NE recognition showing their superiority over a pipeline of dedicated tools for the two tasks. Overall, we believe that this thesis moves a step forward toward adopting direct ST systems in real applications, increasing the awareness of their strengths and weaknesses compared to the traditional cascade paradigm.

Females continue to be underrepresented in leadership despite research demonstrating that leadership effectiveness does not vary by leader gender (Paustian-Underdahl et al., 2014). The current study examines the gender bias in leadership through the lens of leadership perceptions and evaluations; in particular, how perceivers' ratings of a leader would change as a function of the leader's gender. Leadership judgments are based on the leader prototype activated in the perceiver and how consistent/inconsistent the leader is perceived to be with the activated prototype (Lord et al., 2001). Due to the mismatch between the communal-oriented female gender stereotype and agentic-oriented expectations of a successful leader (Eagly and Karau, 2002), it was expected that the female leader would be rated more negatively than the male leader. Furthermore, the perceiver's gender and prior engagement in emotion suppression are investigated as two additional factors that could bias information processing when evaluating leaders. Male perceivers, who tend to hold a stronger masculine understanding of leadership (Koenig et al., 2011), were expected to evaluate the female leader more harshly than the male leader. Additionally, those depleted of their finite self-regulatory resources due to prior emotion suppression (i.e. being in a state of ego depletion; Baumeister et al., 1998) were predicted to rely more heavily on their stereotypes when making subsequent judgments; hence, ego-depleted individuals would demonstrate more bias in their ratings of the female leader relative to the male leader. In the current study, participants were randomly assigned to an emotion suppression or no suppression condition as they watched funny clips from the comedy series "The Office''. Then, they watched four business videos featuring a leader and three business managers. Participants were also randomly assigned to one of the two versions of the business videos portraying either a male or female leader. Leadership perception and leader effectiveness ratings were collected after each of the four business videos, and leader competence and leader warmth ratings were measured once after all four videos. Additionally, behavior recognition accuracy of agentic and communal leadership behaviors that were displayed in the four business videos was assessed. Contrary to expectations, the study findings demonstrate a dominant female leader effect; the female leader was evaluated more favorably than the male leader on all four leader judgments. This was observed both within the repeated measures and overall leadership ratings. An ego depletion effect was also observed; ego-depleted individuals showed lower accuracy in behavior recognition ratings and more leniency in leader warmth ratings. Furthermore, ego-depleted individuals showed less discernment by giving higher leader effectiveness ratings over time compared to non-ego-depleted individuals. Perceiver gender did not meaningfully affect leadership judgments. The unexpected pattern of bias in favor of the female leader instead of against her suggests that the nature of gender and leader stereotypes may be changing; the incongruence between the female stereotype and leader expectations may be decreasing, leading to more favorable evaluations of the female leader by both male and female perceivers. Moreover, the ability to provide fair and accurate judgments of leader effectiveness is reduced when depleted. Limitations and future research directions are discussed. / Doctor of Philosophy / The gender gap persists in leadership; although leader effectiveness has not been found to vary by the leader's gender, female leaders tend to be perceived and evaluated more negatively than male leaders. One reason for this is the mismatch between societal expectations for how women are ideally expected to behave and the expectations associated with a successful leader. In this study, gender bias in leader judgments and behavior recognition accuracy is examined by a leader's gender. Additionally, the perceiver's gender and prior engagement in emotion suppression are studied as two additional factors that can influence bias in leader ratings. Study findings demonstrate an unexpected but dominant female leader effect, where the female leader was perceived as more leader-like and rated more effective, more competent, and warmer than the male leader by both male and female perceivers. The amount of self-regulatory resources available also affected subsequent processing capabilities; those who suppressed their emotions and were depleted of their self-regulatory resources were less accurate in their behavior recognition ratings and were more lenient in their leader warmth ratings. Future research should explore if and how the nature of gender and leader stereotypes are changing, as evaluations of female leaders may not be as negatively-biased as it was previously.

Mitochondrial DNA (mtDNA) mutations have been widely observed to cause a variety of human diseases, especially late-onset neurodegenerative disorders. The prevalence of mitochondrial diseases caused by mtDNA mutation is approximately 1 in 5,000 of the population. There is no effective way to treat patients carrying pathogenic mtDNA mutation; therefore preventing transmission of mutant mtDNA became an important strategy. However, transmission of human mtDNA mutation is complicated by a large intergenerational random shift in heteroplasmy level causing uncertainty for genetic counseling. The aim of this dissertation is to gain insight into how human mtDNA heteroplasmy is inherited. By working closely with our experimental collaborators, the computational simulation of mouse embryogenesis has been developed in our lab using their measurements of mouse mtDNA copy number. This experimental-computational interplay shows that the variation of offspring heteroplasmy level has been largely generated by random partition of mtDNA molecules during pre- and early postimplantation development. By adapting a set of probability functions developed to describe the segregation of allele frequencies under a pure random drift process, we now can model mtDNA heteroplasmy distribution using parameters estimated from experimental data. The absence of an estimate of sampling error of mtDNA heteroplasmy variance may largely affect the biological interpretation drawn from this high-order statistic, thereby we have developed three different methods to estimate sampling error values for mtDNA heteroplasmy variance. Applying this error estimation to the comparison of mouse to human mtDNA heteroplasmy variance reveals the difference of the mitochondrial genetic bottleneck between these organisms. In humans, the mothers who carry a high proportion of m.3243A>G mutation tend to have fewer daughters than sons. This offspring gender bias has been revealed by applying basic statistical tests on the human clinical pedigrees carrying this mtDNA mutation. This gender bias may partially determine the mtDNA mutation level among female family members. In conclusion, the application of population genetic theory, statistical analysis, and computational simulation help us gain understanding of human mtDNA heteroplasmy inheritance. The results of these studies would be of benefit to both scientific research and clinical application. / Ph. D.