12Methodology for the Use of Administrative Data in Business Statistics

Arnout van Delden1 and Danni Lewis2

1Statistics Netherlands, The Hague, The Netherlands

2Office for National Statistics, Newport, United Kingdom

12.1 Introduction

Traditionally, official business statistics have been produced as single‐source statistics based on surveys in which a coherent set of variables is observed. The advantage of this approach is that units, population, variables, and timing can be defined by the National Statistical Institute (NSI). Nowadays, a wide variety of sources other than surveys are available, such as public administrative data, data from business administrations, and big data sources. These new data sources offer the possibility to increase the information richness of the outputs, reduce data collection costs and response burden, increase output granularity, and achieve more up to date and more frequent outputs. Since these new sources were generally designed for purposes other than official statistics, their units, populations, variables, and frequencies often differ from those of the intended target population. The secondary use of these alternative data sources implies that specific methodology is needed that is aimed to yield “best,” i.e. unbiased and accurate, population estimates.

Our focus is on administrative data, which we define as data collected by an organization external to the statistical office for administrative purposes, thus not targeted for use in official ...

Get Advances in Business Statistics, Methods and Data Collection now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.