COLLABORATIVE AND DIGITAL ANALYSIS OF BIG QUAL DATA IN TIME SENSITIVE CONTEXTS - LISTEN

在时间敏感的背景下对大质量数据进行协作和数字分析 - 听

• 批准号: MR/W029766/1
• 负责人: Cecilia Vindrola
• 金额: $ 57.79万
• 依托单位: University College London
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2022
• 资助国家: 英国
• 起止时间: 2022 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=MR%2FW029766%2F1
• 关键词: COLLABORATIVE; DIGITAL; ANALYSIS; BIG; QUAL

Research needs to be efficient and rigorous to inform evidence-based emergency response. Qualitative data provides public health authorities with key insight into behaviours, beliefs, and contextual factors that shape complex and, often, unprecedented situations. However, qualitative research is frequently excluded from emergency response research plans due to misconceptions around the length of time it requires, its biases, and the level of training required to carry it out properly. Valuable qualitative data is therefore not used to its full potential in decision-making processes, putting at risk the applicability of findings for each context. This project will refine, consolidate and disseminate the first method for the rapid analysis of large qualitative datasets, so big qualitative data can be routinely used to inform response efforts in the context of emergencies.Our multidisciplinary team based at University College London and Oxford University is strategically well placed to carry out this work due to our state of the art knowledge about research methods, our large international network of researchers, and popular training courses in English and Spanish. The project will consist of four steps, each resulting in concrete outputs that address the sub-aims of the study: (1) We will review the literature and learn from other methods designed for the rapid analysis of qualitative data. We will use this information to design LISTEN, an analysis method that combines digital tools to manage and visualise large volumes of data, and collaborative approaches to involve stakeholders in the interpretation of findings. (2) We will test LISTEN on three existing datasets (a large interview study capturing the experiences healthcare workers delivering care during the COVID-19 pandemic in the UK, social media data on Long Covid, and open-ended survey responses about the impact of environmental disasters on mental health in Peru). (3) We will evaluate the quality of LISTEN by studying the successful involvement of people with lived experience, the reliability of findings across multiple researchers and teams, and the financial and infrastructure resources required to conduct LISTEN analysis. (4) We will then host an online international symposium to present LISTEN to a global network and we will develop an accessible toolkit and teaching materials to promote the future use of LISTEN by other teams. Our project is informed by co-production principles, and our team is committed to open science. We will convene a Consultation Group of experts including researchers, clinicians, policymakers, emergency response experts and lived experience researchers, to advise across all stages of the project. Our primary outputs will include a dynamic website and a project page on the Open Science Framework (OSF) to share openly available and accessible study results. Establishing the LISTEN method will shift the focus of emergency response from quantitative indicators, to rich indicators developed from the bottom-up. This will have a deep impact on the sensitivity of emergency response to local needs, as well as local capacities, increasing the potential for speedy societal and economic community recovery.

研究必须高效和严格，以便为基于证据的应急反应提供信息。定性数据为公共卫生当局提供了对行为，信仰和背景因素的关键见解，这些因素形成了复杂的，而且往往是前所未有的情况。然而，定性研究经常被排除在应急响应研究计划之外，这是因为人们对定性研究所需的时间长度、偏见以及正确执行定性研究所需的培训水平存在误解。因此，在决策过程中没有充分利用有价值的定性数据，使调查结果对每种情况的适用性面临风险。该项目将完善、巩固和推广快速分析大型定性数据集的第一种方法，以便在紧急情况下可以常规使用大型定性数据来为响应工作提供信息。我们位于伦敦大学学院和牛津大学的多学科团队具有战略优势，能够开展这项工作，因为我们在研究方法方面拥有最先进的知识，我们庞大的国际研究人员网络，以及英语和西班牙语的热门培训课程。该项目将由四个步骤组成，每个步骤都将产生解决研究子目标的具体成果：（1）我们将回顾文献，并学习其他用于快速分析定性数据的方法。我们将使用这些信息来设计LISTEN，这是一种分析方法，它结合了数字化工具来管理和可视化大量数据，并采用协作方法让利益相关者参与对结果的解释。(2)我们将在三个现有的数据集上测试LISTEN（一项大型访谈研究，记录了英国新冠肺炎大流行期间医护人员提供护理的经验，关于长期新冠肺炎的社交媒体数据，以及关于秘鲁环境灾害对心理健康影响的开放式调查回应）。(3)我们将通过研究具有生活经验的人的成功参与，多个研究人员和团队的调查结果的可靠性以及进行LISTEN分析所需的财务和基础设施资源来评估LISTEN的质量。(4)然后，我们将举办一个在线国际研讨会，向全球网络介绍LISTEN，我们将开发一个可访问的工具包和教材，以促进其他团队未来使用LISTEN。我们的项目遵循联合生产原则，我们的团队致力于开放科学。我们将召集一个专家咨询小组，包括研究人员、临床医生、政策制定者、应急专家和现场经验研究人员，为项目的各个阶段提供建议。我们的主要成果将包括一个动态网站和一个开放科学框架（OSF）上的项目页面，以分享公开可用和可访问的研究结果。建立LISTEN方法将把应急反应的重点从量化指标转移到自下而上制定的丰富指标。这将对紧急反应对当地需要的敏感性以及当地能力产生深刻影响，增加社会和经济社区迅速恢复的潜力。