Big Data Meets Survey Science

19Artificial Intelligence and Machine Learning Derived Efficiencies for Large‐Scale Survey Estimation Efforts

Steven B. Cohenand Jamie Shorey

RTI International, Research Triangle Park, NC, USA

19.1 Introduction

“End‐product” analytic resources used to inform policy, and action must have rigorous statistical integrity. To achieve this goal, statistical and analytic staff devote substantial time and effort to implement estimation and associated imputation tasks, which are essential components of the end‐product analytic databases derived from national or subnational surveys and related data collections. These efforts require a substantial commitment of project funds to achieve, and significant lag times often exist from the time data collection is completed to the time the final analytic data file is released. This chapter focuses on the development and implementation of artificial intelligence (AI)‐ and machine learning (ML)‐enhanced applications to imputation for national surveys that achieve cost and time efficiencies while satisfying well‐defined levels of accuracy that ensure data integrity. We emphasize enhanced processes as an alternative to manual, repetitive, or time‐intensive tasks; operationalize decisions based on predefined outcome preferences and on access to input data that sufficiently inform the decisions; and facilitate real‐time interpretation and interactions for accessing and acting on the AI‐derived decisions so users can focus on higher‐order thinking and ...