A Way Forward to Integrating Cascading Risk into Local Resilience Strategies in Central Asia

This research developed a way forward to integrate cascading risk into local resilience strategies in Nur-Sultan, Bishkek, and Dushanbe, the respective capitals of Kazakhstan, Kyrgyzstan, and Tajikistan. To do so, it met three core objectives: 1) To examine stakeholder awareness and perceptions of cascading risk and interconnected failures. 2) To evaluate the extent to which cascading risk and interconnected failures are considered in existing strategies for resilience. 3) To understand local capacities and priorities for actions to integrate cascading risk into local resilience strategies. It was hypothesised that if the integration of cascading risk and interconnected failures in existing resilience strategies was insufficient, then the results could outline: 1) Inadequate awareness (consciousness) and understanding (knowledge) of cascading risk amongst key stakeholders. 2) Acknowledgement of the relevance of cascading risk to resilience but limited local capacities to develop and implement strategies. Research followed a mixed methods approach comprising a questionnaire and semi-structured interviews. This dataset includes only the questionnaire data, to preserve anonymity. Data collection was carried out in July 2021 and targeted mixed stakeholders working in fields relating to disaster risk reduction, resilience, infrastructure, urban development, and urban policy in Central Asia. The questionnaire was designed in SurveyMonkey, comprising 20 questions across six pages including an introduction which stated the target audience and completion time (5–10 minutes). Judgment and snowball sampling identified participants. The questionnaire was made available in English and Russian and completed online on a self-administered basis, accessible via two web-links distributed via email and LinkedIn. Question 1 asked what location participants answers best related to, with an option for general/regional if participant knowledge was not specific to one locality. The questionnaire received 115 responses, of which 55 were considered valid based on a completion rate of at least 50 percent. The high number of responses with low completion was considered a result of the questionnaire design and sampling method. Almost half of participants only answered Q1 (location). Subsequent questions were not visible until ‘next’ was clicked, and whilst the introduction stated the target audience, participants contacted through snowball sampling may have deemed their experience not relevant after reviewing questions in the following sections. Sheet 1 'Raw' is the raw data for the 115 responses collected. Sheet 2 'Accepted' includes the 55 responses that were accepted for this research, based on a completion rate of at least 50 percent. Sheet 3 'Accepted & Translated' includes the accepted responses, with the 15 Russian responses translated to English. Sheet 4 'Numerical' details the accepted responses, converted to numerical data where possible.

本研究为哈萨克斯坦、吉尔吉斯斯坦、塔吉克斯坦三国首都努尔苏丹（Nur-Sultan）、比什凯克（Bishkek）与杜尚别（Dushanbe）的地方韧性战略纳入级联风险（cascading risk）提供了可行路径。为此，本研究设定三项核心目标：1） 梳理利益相关者对级联风险与关联失效（interconnected failures）的认知与感知情况；2） 评估现有韧性战略中纳入级联风险与关联失效的程度；3） 明确将级联风险纳入地方韧性战略的本地能力与行动优先级。本研究提出如下假设：若现有韧性战略对级联风险与关联失效的整合不足，则研究结果将呈现两类情况：1） 关键利益相关者对级联风险的认知与理解存在欠缺；2） 尽管各方认可级联风险与韧性建设的相关性，但本地在制定与落实相关战略方面的能力有限。本研究采用混合研究方法，涵盖问卷调查与半结构化访谈。为保护受访者匿名性，本数据集仅包含问卷调查相关数据。数据采集工作于2021年7月开展，研究对象为中亚地区从事灾害风险降低、韧性建设、基础设施、城市发展与城市政策相关领域的多元利益相关者。本问卷通过SurveyMonkey平台设计制作，共包含20道题目，分布于6个页面，其中开篇说明部分明确了目标受众与作答时长（5-10分钟）。本次研究采用判断抽样与滚雪球抽样相结合的方式招募受访者。问卷提供英文与俄文两个版本，采用线上自主作答形式，通过邮件与领英（LinkedIn）分发的两条链接进行访问。第1题询问受访者最贴合的所在地区，若受访者的专业认知未特指某一具体地区，可选择「通用/区域」选项。本次问卷共回收115份作答，其中以作答完成率不低于50%为筛选标准，最终得到55份有效问卷。本次回收的作答中，完成度较低的样本占比较高，该情况被认为与问卷设计及抽样方法有关。近半数受访者仅完成了第1题（所在地区）。由于后续题目需点击「下一步」按钮才会显示，且尽管开篇已明确说明目标受众，但通过滚雪球抽样招募的受访者在浏览后续题目后，可能认为自身经验与研究主题不符，因此未完成后续作答。本数据集包含四张工作表：工作表1「Raw」为本次回收的115份作答的原始数据；工作表2「Accepted」收录了按照50%及以上完成率标准筛选出的55份有效作答；工作表3「Accepted & Translated」包含上述有效作答，其中15份俄文版本的作答已被译为英文；工作表4「Numerical」对有效作答进行了整理，尽可能将相关内容转换为数值型数据。