Potential Uses of Scanner Data - A Case Study Using Coffee Data - ARCHIVED

Statistical agencies such as Statistics Canada are investigating the use of scanner data for their own purposes. Interest has grown in the potential uses of this data to improve the quality of price indexes. This paper reports on initial research done in Prices Division. The paper looks at scanner data and the feasibility of its use to produce CPI estimates; evaluates current CPI methodology and procedures; and the impact that use of scanner data would have on the CPI commodity indexes. The main focus of the study, however, is to explore the impact that scanner data would have on the CPI basic commodity indexes covered by scannable items. Since the CPI criterion relates to a limited selection of scanner data, an examination will be made of the impact of gradually relaxing the criteria to include more products and outlets from the scanner data. The initial subset was derived by applying the CPI criteria of volume selling brands and outlets. Each of these changes in criteria yielded a different subset of scanner data. Calculations were performed using these various subsets of scanner data and their results compared to the CPI. An analysis of the results will be used in determining the strengths and limitations of CPI data, detect any deficiencies and provide information for revision of pricing selection.