| 来源类型 | Working Paper |
| 规范类型 | 报告 |
| DOI | 10.3386/w25657 |
| 来源ID | Working Paper 25657 |
| Administrative Data Linking and Statistical Power Problems in Randomized Experiments | |
| Sarah Tahamont; Zubin Jelveh; Aaron Chalfin; Shi Yan; Benjamin Hansen | |
| 发表日期 | 2019-03-18 |
| 出版年 | 2019 |
| 语种 | 英语 |
| 摘要 | Objective: The increasing availability of large administrative datasets has led to a particularly exciting innovation in criminal justice research, that of the “low-cost” randomized trial in which administrative data are used to measure outcomes in lieu of costly primary data collection. In this paper, we point out that randomized experiments that make use of administrative data have an unfortunate consequence: the destruction of statistical power. Linking data from an experimental intervention to administrative records that track outcomes of interest typically requires matching datasets without a common unique identifier. In order to minimize mistaken linkages, researchers will often use “exact matching” (retaining an individual only if all their demographic variables match exactly in two or more datasets) in order to ensure that speculative matches do not lead to errors in an analytic dataset. Methods: In this paper, we derive an analytic result for the consequences of linking errors on statistical power and show how the problem varies across different combinations of relevant inputs, including the matching error rate, the outcome density and the sample size. Results: We show that this seemingly conservative approach leads to underpowered experiments and potentially to the failure of entire experimental literatures. For marginally powered studies, which are common in empirical social science, exact matching is particularly problematic. Conclusions: We conclude on an optimistic note by showing that simple machine learning-based probabilistic matching algorithms allow criminal justice researchers to recover a considerable share of the statistical power that is lost to errors in data linking. |
| 主题 | Econometrics ; Estimation Methods ; Other ; Law and Economics |
| URL | https://www.nber.org/papers/w25657 |
| 来源智库 | National Bureau of Economic Research (United States) |
| 引用统计 | |
| 资源类型 | 智库出版物 |
| 条目标识符 | http://119.78.100.153/handle/2XGU8XDN/583331 |
| 推荐引用方式 GB/T 7714 | Sarah Tahamont,Zubin Jelveh,Aaron Chalfin,et al. Administrative Data Linking and Statistical Power Problems in Randomized Experiments. 2019. |
| 条目包含的文件 | ||||||
| 文件名称/大小 | 资源类型 | 版本类型 | 开放类型 | 使用许可 | ||
| w25657.pdf（563KB） | 智库出版物 | 限制开放 | CC BY-NC-SA | 浏览 | ||
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。
