Evaluating the Representativeness in the Geographic Distribution of Twitter User Population
Evaluating the Representativeness in the Geographic Distribution of Twitter User Population
复制标题
评估 Twitter 用户群体地理分布的代表性
DOI：
10.1145/3281354.3281360
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
Van Hook, Jennifer
中科院分区：
文献类型：
--
作者：
Yin, Junjun;Chi, Guangqing;Van Hook, Jennifer
Twitter数据正在成为大数据流，并吸引了多学科的兴趣，以研究传统调查无法很好衡量的人口特征和社会问题。然而，由于我们对用户的人口特征知之甚少，出于对人口代表性的担忧，推特数据的使用遭到了强烈抵制。评估Twitter用户在多大程度上代表了不同人口群体的群体是至关重要的。这项研究评估了推特用户群体的代表性，并检查了其地理分布及其与真实人群的对应关系。通过估计2014年毗邻美国的推特用户人口统计，初步结果显示，相对于县级真实人口，某些人口群体的代表性过高或过低。代表性指数被用来评估推特样本在地理上的代表性，这可能有助于进一步的研究，以确定偏差的决定因素。
Twitter data are becoming a Big Data stream and have drawn multidisciplinary interests to study population characteristics and social problems that cannot be measured well by traditional surveys. However, the use of Twitter data has been strongly resisted because of concerns about the representativeness of the population as we know little about the demographic characters of the users. It is critical to evaluate the extent to which Twitter users represent the population across different demographic groups. This study evaluates the representativeness and examines the geographic distributions of Twitter user population and its correspondence to the real population. By estimating Twitter user demographics for the contiguous U.S. in 2014, the preliminary results revealed both over- and under-representation of certain demographic groups against the real population at county-level. A representation index is used to assess the representativeness of Twitter samples geographically, which may help further studies to identify the determinants of biases.