Varieties of Egalitarianism. Mapping the Politics of Inequality with Online Crowdcoding

尽管经济不平等具有无可争议的政治和社会相关性，但我们对政党如何看待和谈论（不）平等以及这如何影响不平等一无所知。考虑到政党是代议制民主的关键行为体，并考虑到选民对（不）平等的态度受到大量研究和数据集的影响，这是值得注意的。为了解决这一差距，该项目创新性地利用了“群众的智慧”，绘制了自1980年以来10个经合组织国家的政党如何看待平等，以及这如何影响政策选择和不平等模式。该项目的第一个目标是收集关于各国和各党派之间平等概念差异的数据。平等概念的定义是，在何种程度上平等被视为结果平等、机会平等和社会流动性平等、权利平等和不歧视，在何种程度上平等被视为其他具体（如生态）术语，或仅被提及而不具体说明。这些数据是通过政治文本的在线众包收集的。多个独立的判断被组合用于文本片段的每个编码-遵循群体智慧的想法，根据该想法，外行的独立判断的聚合可以导致与专家的结果相当的结果。我最近的工作表明，在什么条件下，即使是在迄今为止被认为过于复杂的编码任务中，人群也可以与专家相匹配。基于IP地址的使用和5000个测试问题，将由研究小组校准，类别方案将扩大到100万个编码器判断。这种方法将专家的上下文敏感性与人群编码无与伦比的可重复性，速度和成本效益相结合。根据由此产生的数据，项目组将追踪平等概念的变化，并研究各方在多大程度上同意这些概念。除了总的趋势和趋同与分歧的问题外，还应考虑具体问题，例如，有争议的观点，即各政党不再强调经济平等，将受到考验。这一项目阶段最明显的成果之一将是一个公开的“平等主义的品种”数据库。 第二个目标是找出平等概念如何影响政策和不平等。我们将评估平等概念对政策概况和广泛的不平等结果的影响。我们还将研究平等概念和政策选择如何在塑造不平等方面相互作用。为了理解平等概念、政策和不平等之间的这些联系，将考虑各种经济和体制条件的影响以及我们所研究的时间范围的相关性。分析将基于回归分析，使用10个国家约100个政府的数据，但将结合对政治制度和市场经济类型截然不同的3个经合组织国家（丹麦、德国、美国）的案例研究。

Despite the undisputed political and societal relevance of economic inequality, we know nothing about how parties conceive of and speak about (in-)equality and how that affects inequality. This is remarkable considering that parties are pivotal actors in representative democracy and given that voters’ attitudes towards (in-)equality are subject to a host of studies and datasets. To address this gap, the project makes innovative use of the ‘wisdom of the crowd’ to map how parties in 10 OECD countries since 1980 conceive of equality and how that affects policy choices and inequality patterns. The project’s first aim is to gather data on how concepts of equality differ across countries and parties. The concepts of equality are defined by the extent to which equality is conceived of as equality of outcome, equal chances and social mobility, equal rights and non-discrimination, in other specific (e.g. ecological) terms, or only mentioned without specification. The data is gathered through Online Crowdcoding of political texts. Multiple independent judgments are combined for each coding of a text fragment – following the idea of the wisdom of the crowd, according to which aggregation of independent judgments of laypersons can lead to results on par with the results of experts. My recent work shows under which conditions the crowd can match experts even for coding tasks thus far considered too complex. Based on the use of IP addresses and 5000 test questions that will be calibrated by the research group, the category scheme will be scaled up to 1,000,000 coder judgments. This approach combines the context sensitivity of experts with the unmatched reproducibility, speed, and cost-efficiency of crowd coding. Based on the resulting data, the project group will trace changes in the concepts of equality and look at the extent to which parties agree on them. In addition to general trends and questions of convergence and divergence, specific questions, e.g., the contested idea that parties have de-emphasized economic equality, will be tested. One of the most visible deliverables of this project phase will be a publicly available ‘Varieties of Egalitarianism’ database. The second aim is to find out how the concepts of equality shape policies and inequality. We will assess the effects of equality concepts on policy profiles and on a broad set of inequality outcomes. We will also examine how equality concepts and policy choices interact in shaping inequality. To understand these linkages between equality concepts, policies, and inequality, the impact of various economic and institutional conditions and the relevance of the time horizon we look at will be taken into account. The analysis will be based on regression analysis that uses data on approximately 100 governments in the 10 countries, but will be combined with case studies of three OECD countries with vastly different political systems and types of market economies (Denmark, Germany, the US).