The Integrated Business Statistics Program is Ten Years Old! A Methodogical Perspective
Conference
64th ISI World Statistics Congress
Format: CPS Paper
Keywords: framework, innovation, standardized
Session: CPS 11 - Finance and business statistics VII
Monday 17 July 4 p.m. - 5:25 p.m. (Canada/Eastern)
Abstract
Statistics Canada introduced in 2010 the Integrated Business Statistics Program (IBSP). The aim of this program is to achieve greater efficiency in processing Statistics Canada’s business surveys while maintaining or improving the quality of data. The IBSP provides a standardized framework for over a hundred and twenty economic surveys. IBSP surveys use Statistics Canada’s Business Register as a common frame and collect data primarily through electronic questionnaires that are based on harmonized concepts and content. Surveys share common sampling, collection and processing methodologies that are driven by metadata and Statistics Canada’s suite of generalized systems. In addition, common tools are in place to edit, correct, and analyze data.
The IBSP was first deployed in 2013, launching the processing of 38 economic surveys in 2014. This article provides a methodological perspective on the ten years of experience at Statistics Canada of integrating and processing economic surveys under this standardized framework. This will cover methodological challenges faced in satisfying the needs of IBSP surveys, in the context of such aspects as the diversity of subject matter fields (for example finance, agriculture, environment, etc.), the desired frequency (biannual, annual, sub-annual), the focus (for example trends vs cross-sectional), the availability of secondary data (for example tax data), and coordination between programs (for example Census of Agriculture vs agriculture surveys).
The article will also reflect on the opportunities for these surveys to take advantage of the many innovations that were implemented in IBSP, including methods to take into account complex unit structures and multiple allocation objectives at the sample design stage, orienting the active collection efforts to optimize the quality of key estimates, measuring the variability due to imputation for the survey estimates, and expanding the usage of calibrating estimates on secondary totals. The authors will discuss how Statistics Canada adjusted the IBSP methodology over the years considering the evolving needs of surveys, the developments in statistical norms and generalized systems, and to position itself with respect to modernization initiatives. Finally, the article will highlight future strategic orientations and challenges.