26013 research outputs found
Sort by
Invoking love and disgust: promoting multispecies ecologies of solidarity and care through a praxis of Total Liberation
No full text
Resourcing universities to increase their civic impact: meeting challenges of communication, complexity and commitment
Get PDFThe 21st century has witnessed an increasing focus on universities’ civic role and their impacts on the social and economic wellbeing of their communities. Recent work in the UK has emphasised the multifaceted character of universities' potential impacts in a
context of global and local challenges, but also the persistent difficulties they face in realising such potential at a time when universities’ own financial position is increasingly precarious. Drawing on findings from an action learning program involving 14 university-led partnerships, this article introduces a set of resources developed to support ‘civic’ universities and considers their utility and limitations in the context of the challenges identified by participants. These resources include definitions of civic activity, a framework for assessing impact and a theory of civic change. The article contextualises this material within the wider history of civic universities and draws on insights from institutional theory to reflect on the gaps between aspiration and implementation
Investigating the component structure of the Health of the Nation Outcomes Scales for people with Learning Disabilities (HoNOS-LD)
Get PDFBackground:
Outcome measurement is increasingly recognised as a vital element of high-quality service provision, but practice remains variable in the field of intellectual disabilities. The Health of the National Outcome Scales for people with Learning Disabilities (HoNOS-LD) is a widely used Clinician Reported Outcome Measure in the UK and beyond. Over its 20-year lifespan, its psychometric properties have been frequently investigated. Multiple dimensionality reduction analyses have been published, each proposing a different latent structure.
Aim:
To analyse a set of HoNOS-LD ratings to test its internal consistency, to identify the optimal number of latent variables, and to propose the items that group together in each domain.
Methods:
A Principal Component Analysis of 169 HoNOS-LD ratings was performed to produce an initial model. The component loadings for each HoNOS-LD item were then examined, allowing the model to be adjusted to ensure the optimal balance of statistical robustness and clinical face-validity.
Results:
HoNOS-LD’s internal consistency (18 items) was ‘acceptable’ (Cronbach’s alpha = 0.797). On excluding three items that had no bivariate correlations with the other 15 items internal consistency rose to ‘good’ (Cronbach’s alpha = 0.828). The final, four-component solution, using the 15 items possessed good internal reliability.
Conclusion:
HONOS-LD statistical properties compared favourably to the other published latent structures and adheres to the tool’s rating guidance. The four-component solution offers an acceptable balance of statistical robustness and clinical face validity. It provides advantages over other models in terms of internal consistency and/or viability for use at a national level in the UK
University ambidexterity: assessing the nature of interdependence between knowledge exchange and knowledge creation in UK universities
Get PDFThe concept of university ambidexterity has been advanced to capture the simultaneous pursuit of exploration (research) and exploitation (knowledge exchange). While ambidexterity can enhance organisational performance, tensions and barriers have been identified suggesting it is far from straightforward to achieve. In light of this, it has been proposed that universities follow a ‘twisting learning path’ that alternates between exploration and exploitation. However, this violates the idea that innovation activities are persistent in nature. In order to assess the nature of the ambidexterity of UK universities, we use data from the UK Higher Education Business & Community Interaction survey to examine the temporal dynamics of the relationship between research and KE. Through estimating a suite of Panel Variance Auto Regressive models, our results suggest that university ambidexterity has three key characteristics: (1) a determinant temporal ‘path dependent’ effect, whereby research and KE activities exhibit a significant autoregressive component; (2) an inter-temporal bi-directional relationship between research and KE activities; and (3) short time lags between the implementation of research and KE coupled with the dissipation of the relationship over time that is indicative of a persistent relationship between the two. Consequently, we propose an alternative model of university ambidexterity by highlighting the continuous interdependency of research and knowledge exchange within UK universities and its persistent nature
Improving National and International Surveillance of Movement Behaviours in Childhood and Adolescence: An International Modified Delphi Study
Get PDFBackground
The actions required to achieve higher-quality and harmonised global surveillance of child and adolescent movement behaviours (physical activity, sedentary behaviour including screen time, sleep) are unclear.
Objective
To identify how to improve surveillance of movement behaviours, from the perspective of experts.
Methods
This Delphi Study involved 62 experts from the SUNRISE International Study of Movement Behaviours in the Early Years and Active Healthy Kids Global Alliance (AHKGA). Two survey rounds were used, with items categorised under: (1) funding, (2) capacity building, (3) methods, and (4) other issues (e.g., policymaker awareness of relevant WHO Guidelines and Strategies). Expert participants ranked 40 items on a five-point Likert scale from ‘extremely’ to ‘not at all’ important. Consensus was defined as > 70% rating of ‘extremely’ or ‘very’ important.
Results
We received 62 responses to round 1 of the survey and 59 to round 2. There was consensus for most items. The two highest rated round 2 items in each category were the following; for funding (1) it was greater funding for surveillance and public funding of surveillance; for capacity building (2) it was increased human capacity for surveillance (e.g. knowledge, skills) and regional or global partnerships to support national surveillance; for methods (3) it was standard protocols for surveillance measures and improved measurement method for screen time; and for other issues (4) it was greater awareness of physical activity guidelines and strategies from WHO and greater awareness of the importance of surveillance for NCD prevention. We generally found no significant differences in priorities between low-middle-income (n = 29) and high-income countries (n = 30) or between SUNRISE (n = 20), AHKGA (n = 26) or both (n = 13) initiatives. There was a lack of agreement on using private funding for surveillance or surveillance research.
Conclusions
This study provides a prioritised and international consensus list of actions required to improve surveillance of movement behaviours in children and adolescents globally
MI5, the Security State and Communist Political Refugees from Nazism in Second World War-Era Britain: The Case of Gustav Beuer, 1938–1946
No full textThis article examines the British Security Service’s monitoring of Gustav Beuer, a Sudeten German communist and former member of the Czechoslovak parliament who arrived in London as a refugee in 1938 and set up home there until 1946. Although from a friendly country, Beuer was placed under continuous surveillance and was interned, together with five of his compatriots, in 1940–41. The article uses this case to challenge previous interpretations of MI5’s role in spying on communist refugees from Central Europe. It does so by taking a model developed by German political scientist Matthias Lemke for analysing the widely varying levels of threat to democracy posed by exceptional security measures taken by liberal states in closely defined time-frames and applying this to Second World War Britain. Far from pursuing a linear strategy based on anti-communist prejudice alone, MI5 was obliged, inadvertently, to muddle along in its policy towards Beuer and other communist refugees, assimilating unexpected triggers, shifting legal-bureaucratic frameworks, changes in wartime alliances and public opinion, and instances of ministerial action and inaction. The paradoxical result of this was to inject a certain amount of (unforeseen) pragmatism into MI5’s handling of ‘suspect’ refugees, while at the same time undermining its belief that covert fact-finding alone was enough to establish who was and who was not a threat. The article concludes that the self-doubts and unease within the British security state, which are often attributed to intelligence failings and spy scandals in the early Cold War period, had deeper roots in the Second World War era
Gender bias detection on hate speech classification: an analysis at feature-level
Get PDFHate speech is a growing problem on social media due to the larger volume of content being shared. Recent works demonstrated the usefulness of distinct machine learning algorithms combined with natural language processing techniques to detect hateful content. However, when not constructed with the necessary care, learning models can magnify discriminatory behaviour and lead the model to incorrectly associate comments with specific identity terms (e.g., woman, black, and gay) with a particular class, such as hate speech. Moreover, some specific characteristics should be considered in the test set when evaluating the presence of bias, considering that the test set can follow the same biased distribution of the training set and compromise the results obtained by the bias metrics. This work argues that considering the potential bias in hate speech detection is needed and focuses on developing an intelligent system to address these limitations. Firstly, we proposed a comprehensive, unbiased dataset to unintended gender bias evaluation. Secondly, we propose a framework to help analyse bias from feature extraction techniques. Then, we evaluate several state-of-the-art feature extraction techniques, specifically focusing on the bias towards identity terms. We consider six feature extraction techniques, including TF, TF-IDF, FastText, GloVe, BERT, and RoBERTa, and six classifiers, LR, DT, SVM, XGB, MLP, and RF. The experimental study across hate speech datasets and a range of classification and unintended bias metrics demonstrates that the choice of the feature extraction technique can impact the bias on predictions, and its effectiveness can depend on the dataset analysed. For instance, combining TF and TF-IDF with DT and MLP resulted in higher bias, while BERT and RoBERTa showed lower bias with the same classifier for the HE and WH datasets. The proposed dataset and source code will be publicly available when the paper is published