How scholars can support government analytics: Combining employee surveys with more administrative data sources towards a better understanding of how government functions
本文提出一个概念框架，说明行政数据可诊断和改善公共生产函数各环节，并回顾689篇定量文章发现员工调查使用多而行政微观数据利用不足，建议学者拓展数据源以支持政府分析。
Abstract With the digitization of administrative systems, governments have gained access to rich data about their administrative operations. How governments leverage such data to improve their administration—what we call government analytics—will shape government effectiveness. This article summarizes a conceptual framework which showcases that data can help diagnose and improve all components of a public administration production function—from inputs such as personnel and goods, to processes and management practices, to outputs and outcomes. We then assess to what extent public administration scholarship analyses these data sources and can thus inform government analytics. A review of 689 quantitative articles in two public administration journals in 2013–2023 finds that 50% draw on surveys of public employees and 25% on surveys of citizens or firms. By contrast, administrative micro data (14% of articles) are underexploited. Practitioners and scholars would thus do well to expand the data sources used to inform better government.