110 likes | 302 Views
General aspects of the data validation process. ICP-Africa Regional Workshop Pretoria, South Africa 20 - 24 June 2011. General aspects of the data validation process. Outline. Introduction to data validation. General validation process. Validation stages. Why D ata Validation?.
E N D
General aspects of the data validation process • ICP-Africa Regional Workshop • Pretoria, South Africa • 20 - 24 June 2011
General aspects of the data validation process Outline Introduction to data validation General validation process Validation stages
Why Data Validation? These prices are used to calculate purchasing power parities (PPPs) which are further used to derive measures of price and volumerelatives NCAs of the countries participating in the ICP provide the RCA with a set of purchasers’ prices for a selection of items The regional PPPs are then linked to form a global set of PPPs and measures of price and volume relatives The measures are subsequently published by the RCA and the World Bank thereby reaching a variety of users including policy makers, economic analysts, researchers, politicians, journalists and the general public It is therefore essential that the prices on which the PPPs are based on are rigorously checked and corrected i.e. validated
Objectives of theData validation 1st objective Clean data from pricing errors A price error occurs when a price or related metadata is recorded incorrectly or error is introduced afterwards in the process A product error occurs when price collectors price products that do not match the product specification 2nd objective Ensure the comparability of prices The aim of the comparison is to compare like with like; loose product specifications or difficult survey areas may introduce product errors Cannot be attributed to the data of any particular country as they are not errors per se at the national level, given that the data is coherent at sub-national level
Process of theData Validation Validation stages
Interaction during theData Validation Validation Queries on Validation Queries on • Quarterly Prices and Metadata • Annual Prices and Metadata • Validation Tables • PPPs by the ICP Classification • Quarterly Prices and Metadata • Annual Prices and Metadata 1 2 National Level Regional Level Global Level Original or Edited Original or Edited • Quarterly Prices and Metadata • Annual Prices and Metadata • Quarterly Prices and Metadata • Annual Prices and Metadata • Validation Tables • PPPs by the ICP Classification
IntroductiontoCountry level validation Intra-country validation is a stage carried out by the NCAs to establish that 1 Reported prices are correct 2 Price collectors within a country have priced products that match the productspecifications 3 Survey frame worked as anticipated Validation stage for items on regional item lists Quarterly Data National and sub-national Annual Data
IntroductiontoRegional levelvalidation Inter-country validation is a stage carried out by the NCAs and RCAs to establish that 1 Countries within the region have correct i.e. error free data 2 Countries within the region have priced comparableproducts Validation stage for items on regional item list Quarterly Data National and sub-national Annual Data
IntroductiontoGlobal level validation Global validation is a stage carried out by the NCAs, RCAs and the Global Office to establish that 1 Reported prices are correct 2 Price data between the regions is comparable 3 Global results are plausible Validation stage for Global Core List items and Regional PPPs Quarterly Data National averages Annual Data
Iterativenessduringinter-country and global validation • Inter-country and global validation are iterative stages - cycle continues until the data is considered to be final • RCAs or the GO leading the process • NCAs to be pro-active • Different speeds withinthe region may need to be followed • Normally 4 rounds issufficient for inter-country • Different speeds, overlapping price collection and validation needs careful planning