Dataset associated with Framing Climate Uncertainty: Frame Choices Reveal and Influence Climate Change Beliefs.

Dataset description. We investigated in two pre-registered studies (1) how 'communicators' verbally frame a confidence interval regarding projected change in winter precipitation due to climate change (N = 512, speakers2.RData). We assessed this by asking study participants to choose one of two different verbal frames for communicating the above projection (a ‘concerned’ compared to an ‘unconcerned’ frame). Using a five-item scale, we measured their perceived severity of consequences of this projection (including health, financial and property losses and one general severity item) as main predictor, and environmental values (New Ecological Paradigm-scale by Dunlap et al., 2000), political affiliation (7-point scale ranging from ‘strong democrat’ to ‘strong republican’), levels of numeracy (using the adaptive version of the Berlin Numeracy Test, BNT, Cokely et al., 2012), age, gender, and levels of education. We studied further (2) how ‘listeners’ interpret the two different verbal frames (N = 385, listener2.RData). Listeners were randomly allocated to two different experimental groups. As main dependent variable, we measured their perceived severity of consequences of precipitation change, and as covariates environmental values, levels of numeracy, and demographic variables, all as described for study 1. Samples for both studies were recruited via Amazon Mechanical Turk in spring 2017. Answers of participants were collected using the survey software Unipark. Both studies were preregistered at the Open Science Framework, and all materials and measures used can be found at osf.io/3tr4h.
The R-Code used for analysing the study results is stored in the .R-files models_speakers_replicationApril2017.R and models_listeners_replicationApril2017.R.

数据集描述。我们通过两项预注册研究开展了调查：(1)‘传播者’如何对气候变化导致的冬季降水量预测变化相关的置信区间（confidence interval）进行语言框架化（N=512，数据文件：speakers2.RData）。我们通过要求研究参与者为传达上述预测选择两种不同语言框架（‘担忧型’与‘非担忧型’）之一来开展评估。我们使用五题量表测量了他们对该预测后果的感知严重性（包括健康、经济与财产损失及一项总体严重性题项），并将其作为主要预测变量；同时测量了环境价值观（采用Dunlap等人2000年提出的新生态范式量表）、政治立场（七点量表，范围从‘坚定民主党人’到‘坚定共和党人’）、数字素养水平（采用Cokely等人2012年开发的柏林数字素养测试（Berlin Numeracy Test，BNT）自适应版本）、年龄、性别及教育水平。(2)‘听众’如何解读这两种不同的语言框架（N=385，数据文件：listener2.RData）。听众被随机分配至两个不同实验组。我们将其对降水量变化后果的感知严重性作为主要因变量进行测量，并将环境价值观、数字素养水平及人口统计学变量作为协变量（均与研究1所述一致）。两项研究的样本均于2017年春季通过Amazon Mechanical Turk招募。参与者的回答通过调查软件Unipark收集。两项研究均在Open Science Framework平台预注册，所有使用的材料与测量工具可通过osf.io/3tr4h获取。用于分析研究结果的R代码存储于以下.R文件中：models_speakers_replicationApril2017.R及models_listeners_replicationApril2017.R。