No description available.

No description available.

<p dir="ltr">The quality of crowdsource annotations has always been a challenge due to the variability in annotators backgrounds, task complexity, the subjective nature of many labeling tasks, and various other reasons. Hence, it is crucial to evaluate these annotations to ensure their reliability. Traditionally, human experts evaluate the quality of crowdsourced annotations, but this approach has its own challenges. Hence, this paper proposes to leverage large language models like ChatGPT-4 to evaluate one of the existing crowdsourced MAVEN dataset and explore its potential as an alternative solution. However, due to stochastic nature of LLMs, it is important to discern when to trust and question LLM responses. To address this, we introduce a novel approach that applies Rubin's framework for identifying and using linguistic cues within LLM responses as indicators of LLMs certainty levels. Our findings reveal that ChatGPT-4 successfully identified 63% of the incorrect labels, highlighting the potential for improving data label quality through human-AI collaboration on these identified inaccuracies. This study underscores the promising role of LLMs in evaluating crowdsourced data annotations offering a way to enhance accuracy and fairness of crowdsource annotations while saving time and costs.</p><p dir="ltr"><br></p>

This thesis presents an investigation into secondary school English teachers’ and students’ conceptualisations of good writing, and how they might use their understandings of quality in writing for the purpose of improving writing. By focusing on the views and classroom practices of twelve-year-old students and their teachers, the research aims to advance understanding of teachers’ and students’ conceptual thinking about writing quality, and the underlying constructs. The research utilises data from an ESRC-funded project titled Grammar for Writing?: The Impact of Contextualised Grammar Teaching on Pupils’ Writing and Pupils’ Metalinguistic Understanding (grant number RES-062-23-0775). This data was gathered from thirty-one teachers and their Year 8 students over three terms. Lesson observations took place once each term, and were followed by interviews with each project teacher and one teacher-chosen student from each class. Interview questions relating to beliefs about good writing were included in the project schedules and were inductively analysed to discern themes in participants’ responses. Interviews with students took the form of ‘writing conversations’ during which students commented on samples of their own and their peers’ writing. A small-scale follow-up study with three Year 8 classes in one secondary school was used to confirm initial findings and to provide additional data on students’ beliefs about good writing. The research found that teachers’ conceptualisations of writing quality were internally consistent but that variation between teachers was marked. Teachers not only valued different qualities in writing but experienced different degrees of conflict and ambiguity when relating their personal construct of quality to the official, public construct, as embodied in national assessment criteria. The findings support earlier views of teacher judgement as richly textured and complex, drawing on different available indexes, including idiosyncratic conceptualisations of writing quality. Whilst students’ criteria for good writing echoed their teachers’ criteria to some extent, there was also evidence of students drawing on their own conceptualisations of quality, especially in relation to the intended impact of writing on the reader. Many students expressed a strong awareness of writing for an audience and clearly valued writing as a social practice. They especially valued peer judgement of their writing. However, students’ strategies for improving writing were often difficult to articulate, formulaic and generalised, or circumscribed by limited linguistic subject knowledge. The study is significant in offering an insight into teachers’ and students’ conceptualisations of writing quality and how these might be brought into play in the writing classroom. The findings may have particular resonance since they are reported at a time of radical change to assessment policy and practice in secondary schools in England.

The purpose of this study was to develop and test a model which addresses special education program evaluation needs. As such, the focus was on development. Often development and research are seen as one (e.g., Department of R & D; Director of R & D). They are, however distinctively different in process and product. The model developed provides general and special education leaders responsible for providing special education services with high quality data and procedures for decision making related to special education. Providing special education services is a complex responsibility. Not only are critical lifelong decisions related to students and their parents made, but there are stringent federal laws, complex state agency policies, detailed financial and programmatic reporting requirements, and often linkages to a variety of outside professionals and service provider agencies and organizations. There is a need for an effective program evaluation model useful to the uniqueness of special education. A variety of program evaluation models have been used in education and other organizational environments (e.g., Mabry, 2002; Patton, 2002; Posavac & Carey, 2003; Renger & Titcomb, 2002; Warburton, 2003). But, their application to special education has been limited and often ineffective or inefficient to address the evaluation needs of special education. This evaluation model development utilized the best of knowledge and procedures of existent evaluation models and adapted them to the uniqueness of special education. The special education evaluation model developed named Program Effectiveness in Special Education (PEiSE) identified espoused and in-use actions in a school district. This information with analysis, discussions, and data provided powerful special education information. To form the structure of the model, PEiSE utilized aspects of the CIPP Evaluation Model developed by Stufflebeam (2002), Logic Framework Model (Suchman, 1967), and the Utilization-Focused Evaluation Model (Patton, 1978). The process brought a number of within the district (Brunner & Guzman, 1989) and outside the district stakeholders into the development process which provided an expertise enhancing model effectiveness (Eisner, 1983). Information gathered from all stakeholders came in various forms and contained data acquired with little or no bias in the instruments or process used (Scriven, 1974; Provus, 1971; Cronbach, 1981; Stake, 1973). These processes not only had potential to improve the special education programs but also to improve the evaluation process itself (Eraut, 1984). The model also considered the limitations of resources of special education services (Stufflebeam, 1971; Tripodi, Pellin, & Epstein, 1971; Gold, 1988). Finally the process proved instrumental in bringing the primary discipline of general education and the complementary discipline of special education physically, philosophically and practically together for the benefit and improvement of services to all students. In conceptualizing the process, a flowchart of events was developed utilizing the form and philosophy of existing best practices in evaluation models and the foundational theory of organizational and program improvement and effectiveness (Argyris & Schon, 1974) PEiSE required the development of plans to reduce or eliminate discrepancies between what practices are espoused and what are actually in-use by practitioners. The PEiSE process included twelve phases: Point of Contact; Scope of Evaluation; Identify Formal Decision Makers; Structured Interviews with Formal Decision Makers; Compose List of Best Practices with Definitions; Formal Decision Makers Meeting/Approval of Best Practices List; Compose Espoused/In-Use Questionnaire; Collect/Analyze Questionnaire/Supportive Data; Recommendations for Action; Generate Action Plans Designed to Reduce or Eliminate Discrepancies; Execute Action Plans; and Measure Progress. An emphasis throughout PEiSE was that change is a necessary and welcomed part of organizational effectiveness as well as an integral part of organizational learning (Argyris & Schon, 1974). PEiSE guided administrators through the process of clearly articulating the change needed with development and implementation of action plans for change. PEiSE facilitated bringing together general and special education in a mutually beneficial manner to improve the quality and success of services to students with special needs. Specific differences in community and school district approaches to responding to compliance and intent of local, state, and federal regulations and initiatives are managed in the model developed. PEiSE was tested in a large, suburban school district. The testing indicated the model’s potential to: 1) advance evaluation of special education; 2) suggest new collaborative models for general and special education; 3) identify needed areas of research on evaluation, organization, issues of responsibilities, and professional expertise; 4) identify needed areas of pre-service and continuing professional preparation and development; 5) promote researched based programs related to student success. It was recommended that PEiSE include an additional phase of practitioner input on concerns and complaints of existing espoused best practices with suggestions or recommendations for different practices the district should consider. / text

INTRODUÇÃO: O sucesso da terapia com ventilação não invasiva com pressão positiva (VNIPP) está associada com a escolha adequada da interface. A máscara Total face® (TF) é considerada mais confortável, porém possui grande espaço morto (875 ml) e vazamento constante elevado. Os ventiladores próprios para ventilação mecânica invasiva (convencionais) têm sido utilizados, habitualmente, para ventilação não invasiva em ambiente de UTI. OBJETIVOS: Avaliar o desempenho de nove ventiladores convencionais com uso da máscara TF e compará-los com um ventilador próprio para VNIPP (Respironics BiPAP Vision). MÉTODOS: Utilizou-se um modelo com dois simuladores mecânicos do sistema respiratório conectados a uma cabeça de manequim onde foi adaptada a máscara TF que foi conectada aos ventiladores testados. O esforço inspiratório foi simulado utilizando-se o modo pressão controlada. Os ventiladores foram testados na modalidade espontânea ventilação com pressão suporte sendo ajustados dois valores de PEEP (5 e 10cmH2O) e 3 valores de pressão suporte (5, 10 e 15 cmH2O). Foi testado se os ventiladores funcionavam com a máscara TF e seu desempenho em relação à compensação de vazamento, pressurização, pico de fluxo atingido, atrasos inspiratório e expiratório. RESULTADOS: O ventilador Vision funcionou em todas as situações. Quatro ventiladores convencionais funcionaram (Horus, Vela, E500 e Servo i). O principal problema com os ventiladores que não funcionaram foi o autodisparo e o desligamento do fluxo inspiratório. O pico de vazamento medido foi maior que 1L/s, em média, e o pico de fluxo gerado, muitas vezes, atingia a capacidade máxima em alguns ventiladores. A capacidade de compensar vazamento foi variável entre os ventiladores, mas aqueles com maior dificuldade (E500 e Horus) foram os que mantiveram os menores valores de PEEP e, também, maiores atrasos no disparo, os demais ventiladores, apresentaram atrasos iniciais menores que 100ms. A ciclagem ocorreu por critérios de segurança nos ventiladores Horus, Vela e E500 em algumas medidas. A capacidade de pressurização foi avaliada pelo cálculo do PTP com 500ms e com 1 segundo. A área de pressurização com 1 segundo ficou abaixo de 50% da área esperada para todos os ventiladores, inclusive para o Vision, específico para VNIPP, sendo que o pior desempenho foi do ventilador Horus. CONCLUSÕES: Entre nove ventiladores convencionais testados, apenas quatro funcionaram com a TF. O desempenho entre os ventiladores foi variável, sendo que, alguns deles não se mostraram adequado para uso com VNIPP usando a máscara TF. A maior dificuldade para o funcionamento dos ventiladores convencionais foi lidar com o grande vazamento, com ocorrência de autodisparos ou desligamento do fluxo de ar do ventilador, acusando desconexão. O vazamento de ar pelos orifícios da máscara é elevado. Os ventiladores Horus e E500 tiveram atrasos maiores que 100ms no disparo; e a ciclagem ocorreu por critérios de segurança em todos ventiladores convencionais, em algumas medidas, exceto o Servo i / BACKGROUND: The success of therapy with noninvasive ventilation with positive pressure (VNIPP) is associated with interface choice. The Total face® mask (TFM) is an interface considered more comfortable than other, but it has a large dead space (875 ml) and constant high leakage. However, intensive care ventilators have been usually used for noninvasive ventilation in the ICU environment, their ability to operate with high air leakage is not known. OBJECTIVES: To evaluate the performance of nine ICU ventilators using TFM and compare them with a VNIPP mode only ventilator (Respironics BiPAP Vision). METHODS: a mechanical respiratory system simulator with two compartments was adapted to TFM what was connected to tested ventilators. The inspiratory effort was simulated using pressure control mode in Newport E500 ventilator. The ventilators were tested in spontaneous mode being adjusted at two values of PEEP (5 and 10cmH2O) and 3 values of pressure support (5, 10 and 15 cmH2O). It was tested if ventilators worked properly with TFM and its performance to compensation for leakage, its pressurization, the capability to reach peak flow target, and trigger and cycling delays. RESULTS: The Vision ventilator worked properly in all situations. Four conventional ventilators (Horus, Vela, E500 and Servo) worked. The main problem with failed ventilator was auto triggering and inspiratory flow turning off. Among worked ventilators, peak inspiratory leakage average was greater than 1L / s , generated peak flow reached maximum capacity in some settings with NIV mode only ventilator. The ability to compensate for leak was variable between ventilators, but those with greater difficulty (E500 and Horus) maintained the lowest values of PEEP and also had great trigger delays, the other ventilators showed trigger delays smaller than 100ms. The cycling occurred by security criteria on Horus, Vela and E500 ventilators in some settings. The ability of pressurization was evaluated by calculating the PTP with 500ms and 1 second. The area of pressurization with 1 second remained below 50% of target area for all ventilators, including for Vision, specifically for VNIPP. Horus ventilator has the worst pressurization performance. CONCLUSIONS: Among nine conventional tested ventilators, only four worked with the TFM. The performance among the ventilators was variable; as a result some of them were not suitable for use with NIV using TFM. The greatest difficulty for conventional ventilator operation was dealing with the large leakage, occurring auto triggering or inspiratory flow turning off, alarming disconnection. The air leakage through the mask holes (exhalation port) was high. Horus and E500 ventilators had trigger delays greater than 100ms and cycling occurred by security criteria for all conventional ventilators, except the Servo, in some setting

Existing research shows the importance of teacher professional development and that decisions regarding professional development in schools lie with administrators. However, while studies have been conducted on the need for administrators fostering professional development in schools in the USA, there appears to be limited research on administrator views of professional development in international school environments. The purpose of this study was to consider views of administrators in international schools regarding professional development activities. Using a mixed methods sequential explanatory design, a questionnaire followed by in-depth interviews, data was collected from a convenience and purposive sample of administrators from international schools in Europe, Asia, Africa, Latin America and the Middle East so as to explore and provide answers to the main research question: “What are the views of administrators on how professional development is planned, implemented, evaluated and resourced in international schools?” Findings from quantitative data indicated strong agreement that administrators should work collaboratively with teachers to determine the professional development needs of the school and that teachers should be involved in assessing professional development effectiveness. Analysis of qualitative data indicated the following themes: teachers are sent to conferences/workshops for professional development or content experts are brought to the school; decisions about professional development should align with school goals; professional development needs should be determined by teachers and administrators collaboratively; there are no significant professional development evaluation processes in place; 2% of the budget is standard allocation in schools for professional development and school boards approve the budget while administrators decide on allocation. By addressing the study’s purpose, this research seeks to contribute to the larger conversation on how administrator views on professional development in international schools can add knowledge to the limited research on effective avenues to professional development in the international school context.

The aim of this diploma thesis is to introduce the issues of evaluating public tenders generally and in a specific organization, to orientate in methods of multicriteria assessment analysis and by their means, to suggest suitable method of evaluating public tenders for the chosen organization. The theoretical part of this thesis deals with the description of the process of evaluating public tenders in the Czech Republic, as well as the explanation of multicriteria decision making concept and the description of methods of multicriteria assessment analysis according to the types of information required. The practical part provides the characteristic of the chosen organization, more precisely its divisions. It also provides the regulation that is followed by the organization when realizing competitive tendering. The analysis itself involves the overview of the tenders realized in the scope of activity of the chosen division in the period of five years, as well as the information on the used evaluating criteria. Out of this overview, the sample of the typical tenders is defined, on which the chosen methods of multicriteria assessment analysis are applied. The aim of this process is to assess the order of the variations after the application of the methods of multicriteria assessment analysis and the newly determined evaluating criterion, as well as compare the results with the originally given basic evaluating criterion. Point method, WSA method and TOPSIS method are used for the assessment. In the final part of this thesis, the recommendation on more frequent utilization of multicriteria assessment is suggested to the contract owner, more precisely using such criteria that would be more effective when using public funds.

The goal of diploma work is to analyze process maps in a company as well as to review level of managing particular processes in relation to their efectivity, performance and effectivness at meeting organization targets. The theoretical part clarifies basis, relevance and principles of quality management. The practical part is focused on mapping quality management system in selected company, especially in the field of process management. First part is oriented to quality system implementation with link to management responsibility, communication and process management support. Second part analysis current system in the company with focus to review of level of particular process management. Third part describes gradual improvement of quality system including evaluation of benefit for the company. Last chapter summarizes outcomin suggestions and recommendations to the company.

INTRODUÇÃO: O sucesso da terapia com ventilação não invasiva com pressão positiva (VNIPP) está associada com a escolha adequada da interface. A máscara Total face® (TF) é considerada mais confortável, porém possui grande espaço morto (875 ml) e vazamento constante elevado. Os ventiladores próprios para ventilação mecânica invasiva (convencionais) têm sido utilizados, habitualmente, para ventilação não invasiva em ambiente de UTI. OBJETIVOS: Avaliar o desempenho de nove ventiladores convencionais com uso da máscara TF e compará-los com um ventilador próprio para VNIPP (Respironics BiPAP Vision). MÉTODOS: Utilizou-se um modelo com dois simuladores mecânicos do sistema respiratório conectados a uma cabeça de manequim onde foi adaptada a máscara TF que foi conectada aos ventiladores testados. O esforço inspiratório foi simulado utilizando-se o modo pressão controlada. Os ventiladores foram testados na modalidade espontânea ventilação com pressão suporte sendo ajustados dois valores de PEEP (5 e 10cmH2O) e 3 valores de pressão suporte (5, 10 e 15 cmH2O). Foi testado se os ventiladores funcionavam com a máscara TF e seu desempenho em relação à compensação de vazamento, pressurização, pico de fluxo atingido, atrasos inspiratório e expiratório. RESULTADOS: O ventilador Vision funcionou em todas as situações. Quatro ventiladores convencionais funcionaram (Horus, Vela, E500 e Servo i). O principal problema com os ventiladores que não funcionaram foi o autodisparo e o desligamento do fluxo inspiratório. O pico de vazamento medido foi maior que 1L/s, em média, e o pico de fluxo gerado, muitas vezes, atingia a capacidade máxima em alguns ventiladores. A capacidade de compensar vazamento foi variável entre os ventiladores, mas aqueles com maior dificuldade (E500 e Horus) foram os que mantiveram os menores valores de PEEP e, também, maiores atrasos no disparo, os demais ventiladores, apresentaram atrasos iniciais menores que 100ms. A ciclagem ocorreu por critérios de segurança nos ventiladores Horus, Vela e E500 em algumas medidas. A capacidade de pressurização foi avaliada pelo cálculo do PTP com 500ms e com 1 segundo. A área de pressurização com 1 segundo ficou abaixo de 50% da área esperada para todos os ventiladores, inclusive para o Vision, específico para VNIPP, sendo que o pior desempenho foi do ventilador Horus. CONCLUSÕES: Entre nove ventiladores convencionais testados, apenas quatro funcionaram com a TF. O desempenho entre os ventiladores foi variável, sendo que, alguns deles não se mostraram adequado para uso com VNIPP usando a máscara TF. A maior dificuldade para o funcionamento dos ventiladores convencionais foi lidar com o grande vazamento, com ocorrência de autodisparos ou desligamento do fluxo de ar do ventilador, acusando desconexão. O vazamento de ar pelos orifícios da máscara é elevado. Os ventiladores Horus e E500 tiveram atrasos maiores que 100ms no disparo; e a ciclagem ocorreu por critérios de segurança em todos ventiladores convencionais, em algumas medidas, exceto o Servo i / BACKGROUND: The success of therapy with noninvasive ventilation with positive pressure (VNIPP) is associated with interface choice. The Total face® mask (TFM) is an interface considered more comfortable than other, but it has a large dead space (875 ml) and constant high leakage. However, intensive care ventilators have been usually used for noninvasive ventilation in the ICU environment, their ability to operate with high air leakage is not known. OBJECTIVES: To evaluate the performance of nine ICU ventilators using TFM and compare them with a VNIPP mode only ventilator (Respironics BiPAP Vision). METHODS: a mechanical respiratory system simulator with two compartments was adapted to TFM what was connected to tested ventilators. The inspiratory effort was simulated using pressure control mode in Newport E500 ventilator. The ventilators were tested in spontaneous mode being adjusted at two values of PEEP (5 and 10cmH2O) and 3 values of pressure support (5, 10 and 15 cmH2O). It was tested if ventilators worked properly with TFM and its performance to compensation for leakage, its pressurization, the capability to reach peak flow target, and trigger and cycling delays. RESULTS: The Vision ventilator worked properly in all situations. Four conventional ventilators (Horus, Vela, E500 and Servo) worked. The main problem with failed ventilator was auto triggering and inspiratory flow turning off. Among worked ventilators, peak inspiratory leakage average was greater than 1L / s , generated peak flow reached maximum capacity in some settings with NIV mode only ventilator. The ability to compensate for leak was variable between ventilators, but those with greater difficulty (E500 and Horus) maintained the lowest values of PEEP and also had great trigger delays, the other ventilators showed trigger delays smaller than 100ms. The cycling occurred by security criteria on Horus, Vela and E500 ventilators in some settings. The ability of pressurization was evaluated by calculating the PTP with 500ms and 1 second. The area of pressurization with 1 second remained below 50% of target area for all ventilators, including for Vision, specifically for VNIPP. Horus ventilator has the worst pressurization performance. CONCLUSIONS: Among nine conventional tested ventilators, only four worked with the TFM. The performance among the ventilators was variable; as a result some of them were not suitable for use with NIV using TFM. The greatest difficulty for conventional ventilator operation was dealing with the large leakage, occurring auto triggering or inspiratory flow turning off, alarming disconnection. The air leakage through the mask holes (exhalation port) was high. Horus and E500 ventilators had trigger delays greater than 100ms and cycling occurred by security criteria for all conventional ventilators, except the Servo, in some setting