Session 30

Back to Schedule

Title of session: Quality of statistical production processes

Chair: Maria João Zilhão

Room: S4A Mariacki

Time: 14:30 - 16:00

Date: 28 June

Session 30 - papers & presentations

Presenting AuthorAbstract
Markus Bönisch
Title: <<< Global standards and national procedures – What can the ESS learn from the Programme for the International Assessment of Adult Competencies (PIAAC) >>>
The Programme for the International Assessment of Adult Competencies (PIAAC) is an international OECD survey that compares key competencies of adults (16-65 years) in 33 countries. In order to obtain high quality datathe international PIAAC Consortium produced an elaborate set of standards for almost all aspects of the national implementation. In Austria, a comprehensive set of procedures was put in place for the PIAAC fieldwork. Some of the international requirements for data collection were not reasonable within the national context and required certain adaptations to accomplish a successful fieldwork. The presentation will talk about PIAAC and its methodological background, describe key fieldwork measures in Austria and discuss how specific measures relate to global data collection standards. Reflecting on this national experience, some of the possibilities and limitations of national compliance to global standards will be discussed. Furthermore the multidimensional assessment of quality in PIAAC (Response Rate, Non-Response-Bias, compliance withstandards) will be discussed and related to national contexts. The conclusions will discuss still open issues regarding data quality in cross-national surveys and the balance between global standards for comparability and the degree of freedom to reconcile national differences. Furthermore some best practices and lessons learned will be presented which have the potential to enhance the quality of European statistics and the ESS.
Monica Scannapieco
Title: <<< Standardization of Business Statistics Processes in Istat >>>
Since the second half of 2014, Istat has launched a modernization program to improve statistics quality. In order to increase the effectiveness and efficiency of the production chain, the revision and the standardization of the business statistics production processes have become of utmost importance. Currently, business surveys use highly customized methods and tools. Some processes are tied to specific persons, and built on their knowledge and skills. This kind of organization produces two major negative effects: duplicated work and limited reuse of tools and competencies. The Generalized Process for Business Statistics (GPBS) aims to identify and implement a general data model and architecture to standardize similar steps of business surveys. The GPBS initiative has the main objective to standardize:
- methods and tools for all the statistical phases of production processes;
- the workflow, meant as the best combination of data and statistical services to implement the statistical process.
The core of GPBS will be a set of shared and generic corporate services based on structured metadata, for processing, storing and analyzing data. Such services will be designed and implemented enhancing the software currently available in the Istat’s generalized software repository. Adopting the conceptual model of the ESS Statistical Production Reference Architecture (SPRA), the main components of the TO-BE integrated architecture, will be services related to Generic Statistical Business Process Model (GSBPM) phases and sub-processes and exploiting enterprise-level building blocks.
Boris Frankovič
Title: <<< Calif – the interactive Shiny web application for calibration of weights of statistical surveys >>>
Since many years the Statistical Office of the Slovak Republic has performed calibration of weights in statistical surveys in order to enhance precision requirements and maintain consistency between surveys. Based on our best practice, it has been proved that this procedure has to be carried out with utmost care. To that end, we prepared the first version of Calif with graphical user interface in 2013. This tool enabled users to take advantage of simple environment full of useful statistics that allowed for swift calibration when the most feasible solution was easily found. However, the old-fashioned GUI has found its limits. Therefore, we come with the new interactive web application made in R package Shiny that offers space not only for the enhancement of the calibration process but also for further development. Apart from new graphical user interface that runs in aweb browser, makes calibration an intuitive and convenient process and offers new features, Calif could be used in the future as amore general statistical tool, which would not be limited by just the calibration but also be open for imputation, estimation and variance calculation.
Nicoletta Pannuzi
Title: <<< Innovation to improve data quality: the case of Italian Household Budget Survey >>>
The recent redesign of the Italian Household Budget Survey (HBS) represents the starting point of this contribution whose aim is to assess the impact of both methodological changes and innovation in data collection on the estimates of household consumption expenditures. In 2014, many innovations were introduced in the HBS, in response to changes in European recommendations and purchasing behaviors (in terms of habits and with respect to the choice of goods, services and places of purchase), and to an increased demand for information in the context of social and economic research. New instruments and techniques have been introduced, together with more accurate methodologies, with the aim of improving the estimation quality of household consumption expenditures and of other derived statistics (as the Italian measures of relative and absolute poverty). Because the old and new HBS were conducted in parallel during the last two quarters of 2012 and during all quarters of 2013, it has been possible to compare the effects of the abovementioned changes on the consumption expenditure estimates. The aim of the paper is not only devoted to the impact of different methodological choices but rather to the implications of differences in data collection and variables definition. Moreover, the impact is evaluated not only in terms of levels, but also taking into account the whole distribution. Also the analysis on which subpopulations are more affected by changes is conducted, by explaining the different effects of each introduced innovation. Finally the paper contains an error profile analysis, with the aim of distinguishing changes due to innovations from possible bias linked to sampling coverage or households' selection. Although our considerations refer to the Italian situation, some evidence and, in particular, the models and analytical procedures may be extended to other realities.
Marek Rojicek
Title: <<< Redesign of the Statistical Information System: Czech experience >>>
In the last years, preparation and implementation of the integration of software tools for collection, processing and dissemination of statistical data has been realized. The main tasks of this project were decrease of the administrative burden laid down on respondents together with the increased quality and accessibility of statistical data for every user of the statistical information system (SIS). The specific targets of this project, realised within the general framework of e-Government implementation and financed mostly by EU funds, were the integration of mathematical-statistical methodologies, procedures, tools and technologies used for evaluation of users' requirements, preparation and processing of statistical tasks, analysing and publication of statistical information. Within this project the development of data warehouse has been completed including the creation of system of secure and protected access to data, as well as the unification of administrative data sources and introduction of mutual electronic contact with respondents. The main part of the project has been realised in the years 2013 – 2014 and in 2015 the project was transformed into routine statistical production process. After implementation of the new SIS the need for higher cooperation of all statistical as well as IT departments significantly increased. Although most of the roles were anchored into the line organisation structure, it was necessary to keep partly project type of the management of the whole process due to its complexity. We expect that the initial investment in the form of financial resources and human capacities should be returned in the following years by decreased administrative burden of respondents and increased quality and accessibility of statistical indicators for users.

Back to Schedule

Font Resize