Survey of Students and Households and Interviews with Key Stakeholders in Oklahoma

The student data for this study were collected via a survey of undergraduate students enrolled in an Introduction to American Government course at Oklahoma State University. A link to the online survey was sent to 1721 students using Qualtrics survey software.To follow Dillman’s (1978) survey protocols over the internet, the initial survey was sent on August 29, 2016, and students received three reminder emails with links to the survey through September 23, 2016. We received 361 completed surveys for an overall response rate of 21%.

The household study used a stratified sampling procedure to identify household samples from the cities of Shawnee (zip codes: 74801, 74802, and 74804) and Sapulpa (zip codes: 74066 and 74067) in Oklahoma. The survey was sent out following (Dillman et al., 2009) procedures. Each household was sent as many as four survey packages (waves 1, 3, 4, and 5) and one reminder postcard (wave 2). The survey was collected from September 12 through October 23, 2016. The mailing list was obtained from Experian Information Solutions Inc. Due to moves and an inaccurate address, 34 addresses were taken out from the original mailing list. The questionnaires were sent out by USPS Mailing Service. Of the total number of survey packages, 1476 were mailed to Sapulpa and 1490 to Shawnee. There were 461 household respondents (436 mailed back and 25 online), 44 households refused, and 273 undeliverable survey packages, resulting in a response rate of 17.40%. Of the respondents, 217 were from Sapulpa and 244 were from Shawnee.

As part of this research effort, we conducted in-depth, semi-structured, interviews with 18 key stakeholders across the state, including local officials at the city and county level (primarily emergency managers and planners), state officials, nongovernmental organisations and representatives from the oil and gas industry. For the privacy of participants, we are not able to share the interview transcripts, but are sharing the interview guide.

本研究的学生数据通过对俄克拉荷马州立大学（Oklahoma State University）《美国政府导论》课程的本科生进行调查收集。使用Qualtrics调查软件向1721名学生发送了在线调查链接。为遵循迪尔曼（Dillman, 1978）的互联网调查协议，初始调查于2016年8月29日发出，学生们在2016年9月23日前收到了三封带有调查链接的提醒邮件。我们共收到361份完整问卷，总回复率为21%。

家庭研究采用分层抽样法（stratified sampling procedure）从俄克拉荷马州的肖尼市（邮政编码：74801、74802、74804）和萨普尔帕市（邮政编码：74066、74067）识别家庭样本。调查遵循迪尔曼等人（Dillman et al., 2009）的流程发放。每个家庭最多收到四套调查包（第1、3、4、5轮）和一张提醒明信片（第2轮）。调查收集时间为2016年9月12日至10月23日。邮寄名单来自益百利信息解决方案公司（Experian Information Solutions Inc.）。由于搬迁和地址不准确，原始邮寄名单中剔除了34个地址。问卷由美国邮政服务公司（USPS Mailing Service）寄出。在所有调查包中，1476份寄往萨普尔帕，1490份寄往肖尼。共有461个家庭受访者（436份邮寄回复，25份在线回复），44个家庭拒绝参与，273份调查包无法投递，最终回复率为17.40%。受访者中，217人来自萨普尔帕，244人来自肖尼。

作为本研究的一部分，我们对全州18名关键利益相关者进行了深度半结构化访谈（in-depth semi-structured interviews），包括市县两级地方官员（主要是应急管理人员和规划人员）、州级官员、非政府组织以及石油和天然气行业代表。为保护参与者隐私，我们无法共享访谈记录（transcripts），但会共享访谈指南。