Census Knowledge Base » Browse » GENERAL CENSUS PLANNING AND MANAGEMENT » 12. Quality assurance

Kosovo Population and Housing Census 2011- Final Results: Quality Report

Margaret Mbogoni — Fri, 12 Oct 2012 12:21:31 GMT

The census quality report has as main objective to provide our users with all information required to assess the reliability of census results. It describes all steps used before, during and after the census enumeration in view of increasing data reliability and consequently its usefulness. It also presents a detail comparison analysis of census outputs with the ones obtained from other statistical surveys or administrative information. Together with the report on census coverage and content errors estimates (PES report), this report provides fully transparent and complete information on the Kosovo census outputs.

Census Quality of French Rolling Census

admin — Thu, 08 Sep 2011 06:43:43 GMT

By the National Institute of Statistics (INSEE), France, 2008.

Since 2004, the French census system has worked to a revised method combining two core principles: a rolling census based on a 5-year cycle, and the deployment of a sample survey scheme led on communes counting either side of 10,000 inhabitants, based on a comprehensive housing inventory called the ‘RIL’. These principles are written into national census law. This revamped census system can output detailed year-in-year-out statistics at each geographic level of analysis, from national figures down to commune or district-level scope (§ 1 to 3).
The census reform was implemented in response to a mounting need for recent demographic data, most importantly at more granular levels. The reforms also make it possible to even out both the HR and budget-related census costs, and implement better operational control (§ 4). Finally, the new census format, and particularly the housing inventory, cuts down on omissions.
The population census, in its French format, meets the five core criteria set by the UN (§ 5) as well as the general quality assurance criteria governing statistical operations (§ 6).
Quality is managed end-to-end throughout the census preparation, collection, processing and publishing operations. Census quality is assured through quality control and metrics protocols plus reviews with input from all the census actors involved: communes, data collection agents, INSEE teams, and subcontractors (§ 7 and 8). Pilot tests and phased investigations are run to enhance a quantitative and qualitative analysis of census coverage (§ 9).

[Available in English and French]

Principles and Recommendations for Population and Housing Censuses, Rev. 2

Jan Beise — Wed, 06 Jan 2010 06:05:21 GMT

By United Nations Statistics Division, 2008.

Since its earliest years, the United Nations has issued a series of international principles and recommendations on population and housing censuses to assist national statistical offices and census officials, throughout the world, in planning and carrying out improved and cost-effective censuses.

The Principles and Recommendations for Population and Housing Censuses, Revision 2 is divided into three parts: Part One - Operational aspects of population and housing censuses – covers mainly issues of census planning and management; Part Two – Topics for population and housing censuses – includes topics as well as their definitions and classification for both population and housing; and Part Three – Census products and data utilizations – is aimed at enhancing the dissemination and utilization of census data. The publication also contains a set of recommended tabulations on population and housing that every country is should produce at least once during the 2010 census round. There are also additional tabulations that countries may produce depending on their needs.

Principles and Recommendations for Population and Housing Censuses, Revision 2 (United Nations publication, Sales No. E.07.XVII.8), 2008.

[available in English, Arabic, Russian]

New method for 2010 population and housing census of Turkey - Considerations about data quality and coverage

Jan Beise — Tue, 06 Oct 2009 05:35:00 GMT

Paper by Turkish Statistical Institute

Until 2000, the population census was carried out in Turkey with a traditional method and the enumeration conducted in one day on a “de facto” basis. Therefore, information on usual residence of the population is not available from past censuses. Another problem was population overcount (imaginary population) due to the fact that census results are used to allocate funds to the municipalities, in proportion to their population size.

In view of the 2010 census, the Turkish Statistical Institute aims at changing the method of population census in order to produce more reliable information particularly on population size and distribution. For this purpose, the population registration system has been improved to cover usual residence address of all people living in the country. This system is considered as a base source for the next census. The main purpose of the paper is to present the method developed to increase the coverage of the population registration system, and the planned methodology for the 2010 census.

Paper presented at the Joint UNECE/Eurostat Meeting on Population and Housing Censuses, Geneva, 13-15 May 2008.

Quality Assessments of the 2008 integrated census – Israel

Jan Beise — Mon, 28 Sep 2009 10:33:00 GMT

By the Central Bureau of Statistics, Israel, 2009.

Census operation and results were evaluated by several quality assessment dimensions. These included Coverage, Accuracy, Consistency, Completeness, Reliability and Validity. The United Nations Economic Commission for Europe publication "Conference of European Statisticians Recommendations for the 2010 Censuses of Population and Housing" lists some quality elements in the three stages of the census process; design, implementation and evaluation. The elements relate to: accuracy, timeliness, accessibility, interpretability and coherence. Israel conducted an integrated census in 2008 with no post enumeration phase since it was assumed that the integrated census with its two samples would provide the necessary information to evaluate the census operations. The evaluation process also took advantage of the fact that the census was an integrated one, which incorporates administrative sources and field operations. The census data collection operation was fully computerized with the use of laptops for data collection and daily data transmission to a central database of collected data. These features made it possible to monitor the field operation with specific tools which took advantage of the technology. To date, the full evaluation process has not yet been completed; it will be completed by the end of 2009. Nevertheless, most of the evaluation elements are completed.

Paper presented at the joint UNECE/Eurostat meeting on Population and Housing Censuses, Geneva, 28-30 October 2009.

The Slovene example how to improve the census count in a register-based census

Jan Beise — Mon, 28 Sep 2009 10:32:00 GMT

By the Statistical Office of the Republic of Slovenia, 2009.

The paper focuses on the two main activities which took place in 2007 and 2008 at the Statistical Office of the Republic of Slovenia in order to improve the census count in the register based Census of Population and Housing in Slovenia in 2011. The first activity was focused on the identification of potential so-called ˝administrative survivors˝ in the Central Population Register (hereinafter CPR) among population aged 80 years or over. The second activity was the introduction of the census usual residence concept in the quarterly population statistics.

Due to the ageing of populations in the developed countries and the increase in the number of the persons aged 80 or over there is a tendency among national statistical institutes to disseminate also the population by the oldest age-groups (90 or more, 100 or more) and even more, to cross-tabulate these age groups with socio-economic topics. It is well known that the uncertainty and unreliability of the population estimates and population counts are higher for the oldest age-groups. Therefore at Slovene national statistical institute the analysis of potential ˝administrative survivors˝ in the age group 80 or over was done in order to improve the quality of the census count at the 2011 register-based census. On the basis of several analyses estimations were made that the centenarians in the quarterly population statistics are overestimated by about 15-20 per cent and the nonagenarians by about 2-3 per cent.

Paper presented at the joint UNECE/Eurostat meeting on Population and Housing Censuses, Geneva, 28-30 October 2009.

Accuracy evaluation of Nuts level 2 hypercubes with the adoption of a sampling strategy in the 2011 Italian Population Census

Jan Beise — Mon, 28 Sep 2009 10:31:00 GMT

By the National Institute of Statistics, Italy, 2009.

Italian National Institute of Statistics (Istat) is considering using sampling techniques in order to adopt a short/long form strategy for the 2011 Italian Population Census. The choice is based on a simple random design for the selection of private household samples from population registers and the calibrated estimators. Because the adoption of a sampling strategy causes the introduction of sampling errors, tests and studies have been conducted in order to evaluate the efficiency of sampling estimates and the accuracy of dissemination hypercubes.

The planning of the 2011 Italian Population Census has taken in account both the critical points of the last census and the possibility to introduce methodological innovations according to the international recommendations. In order to improve the efficiency of the survey operations, to reduce the workload of the municipalities and to minimize the statistical burden for the people involved in the enumeration, many solutions have been taken into consideration. The most important are related to the use of population registers, to a mailing for the census forms and to a mixed mode of data collection mainly based on mail and web.

Introducing sampling strategy in the next Census round implies not only savings and a smaller amount of data to be managed also provides an opportunity to improve the overall quality of data. It will be possible to set up and execute more checks on census forms and follows up in the field to reduce the non-sampling errors. Another advantage of the sampling strategy is improvements to the Census timeliness, which is a constraint since 2011 Census data have to be provided to Eurostat on 1 April 2014.

Paper presented at the joint UNECE/Eurostat meeting on Population and Housing Censuses, Geneva, 28-30 October 2009.

A quantitative approach to the European census quality reporting

Jan Beise — Mon, 28 Sep 2009 10:30:00 GMT

By the Statistical Office of the European Communities, 2009.

The Council of the European Union and the European Parliament adopted in 2008 a regulation on population and housing censuses. For the first time, a European Union (EU) regulation defines the concepts to be used in the census exercises, the topics to be covered and the data sources, and fixes the reference year for the first round on 2011. Further legislative measures – currently under preparation – are foreseen to define the technical specifications and breakdowns of the topics, the programme of statistical data and metadata, the technical format for data transmission, the modalities and structure of the quality reports. The methodological elements of this Regulation acknowledge the Conference of European Statistician (CES)Recommendations for 2010 Round of Population and Housing Censuses.

The Regulation (EC) No 763/2008 attributes great importance to quality issues. Inparticular, it devotes one entire article to the quality assessment, defining its dimensions,prescribing the submission of quality reports from the Member States to Eurostat and stating that"The Commission (Eurostat) shall assess the quality of the data transmitted.” Hence, for the firsttime, the population and housing censuses in the EU will have a formal, structured andcomparable quality reporting. Further, quality shall be assessed and this is a significant stepforward in comparison to simple reporting.

Paper presented at the joint UNECE/Eurostat meeting on Population and Housing Censuses, Geneva, 28-30 October 2009.

Setting the Standard for Quality Control in the United States Decennial Census

Jan Beise — Fri, 11 Jul 2008 04:29:00 GMT

Paper by US Census Bureau (2004)

The Census Bureau has a long-standing tradition of high standards of quality for its census activities and the data collected during the census. To maintain those standards, we have implemented quality control (QC) operations to ensure enumeration and address listing/mapupdating operations followed established procedures. In the past, the process of designing these QC operations has been somewhat ad hoc in nature. That is, once the census operations were designed, appropriate QC operations were then planned and designed to check the quality of the work being done. Because the QC planning began after the production operations had already been designed, there was limited opportunity to integrate the QC into the production work, andso the capabilities and effectiveness of the QC may have been limited. This limitation was mostly manifested by our inability to get real-time information on the QC and provide feedback to the production operation to facilitate corrective actions and continuous improvement.

We are undertaking an effort to build the planning of QC into the production planning so that the QC operations are designed at the same time as the production operations. This enables the QC design to envision all of the vulnerabilities within an operation and address those vulnerabilities appropriately. For QC to be truly effective, it must address all aspects of an operation that maybe subject to error or failure. With this new culture of concurrent planning, we hope to enhance our ability to ensure the quality of census processes and products.

Census Quality Evaluation: Considerations from an international perspective

Jan Beise — Thu, 22 May 2008 10:10:00 GMT

Paper by by Bernard Baffour and Paolo Valente (UNECE)

Regardless of the methodology used and of the measures that can be taken to assure the best quality, the population census cannot be a perfect operation and there will always be some errors affecting the quality of the census results in terms of coverage and content. For this reason, the agencies responsible for the census should always conduct an evaluation of the quality of the census and provide an assessment of the census coverage and content errors. This is particularly important considering the huge resources invested in the census and the key role played by the census results for the national statistical systems and as primary source of statistical data for the countries. Measuring the census quality, however, is not an easy task, in particular when non-traditional census methods are adopted, for which established census evaluation techniques like the Post Enumeration Survey cannot be applied. In this paper, some issues concerning census quality evaluation are discussed for each of the most common census methodologies,considering the implications on the different dimensions of quality. Some considerations are also presented on census quality evaluation from an international perspective, and on the role that InternationalOrganisations can play in this field.

Paper presented at the Joint UNECE/Eurostat Meeting on Population and Housing Censuses, Geneva, 13-15 May 2008.

Quality control in Finnish Censuses in 1970-2000

Jan Beise — Thu, 22 May 2008 10:06:00 GMT

Paper by Statistics Finland

This document presents briefly the history of the register-based population censuses conducted in Finland from 1970 to 2000. The document focuses on methods and measures taken to control the quality of census operations, and on the evaluation of the quality of census results.

Paper presented at the Joint UNECE/Eurostat Meeting on Population and Housing Censuses, Geneva, 13-15 May 2008.

Quality assessment of the register-based Slovenian census 2011

Jan Beise — Thu, 22 May 2008 09:42:00 GMT

By Statistical Office of the Republic of Slovenia.

The use of the administrative data as a means for the costs and burden reduction is a very popular practice in the modern official statistics. In Slovenia already the 2002 Census was partly based on administrative data sources, while the next, 2011 Census is planned to be fully register based. The quality assessment of the census processes and results is an important obligation of statistical offices. In the document some considerations are presented about the modifications of the quality assessment concepts when moving from a conventional to a register-based census.The first part of the document is devoted to some general reflections, while in the second part the focus is on the plans for the quality assessment of the Slovenian 2011 register-based census.

Paper presented at the Joint UNECE/Eurostat Meeting on Population and Housing Censuses, Geneva, 13-15 May 2008.

[Paper also available in French and Russian]

Sample results expected accuracy in the Italian population and housing census

Jan Beise — Thu, 22 May 2008 09:05:00 GMT

This paper by the National Institute of Statistics (Istat), Italy, discusses considerations by Istat to improve the efficiency of the surveys operations and reduce the statistical burden. Istat is considering the use of sampling techniques (based on long form and short form) among many others innovations under evaluation for the next census. In large municipalities (population over 10,000 inhabitants) detailed information would be collected only on a sample of the population (using the long form) whole for the rest of the population only limited information, consisting mainly in demographic variables, would be collected (using the short form). In smaller municipalities, a traditional approach is planned where the long form is submitted to the whole population. Tests and studies were conducted in order to evaluate the efficiency of sampling estimates. Some preliminary results of these tests are presented in this paper. These results seem to encourage the use of sampling techniques in order to adopt a short/long form data collection strategy for the 2011 Italian population census.

Paper presented at the Joint UNECE/Eurostat Meeting on Population and Housing Censuses, Geneva, 13-15 May 2008.

[Paper also available in French and Russian]

2011 Census Quality Strategy for England and Wales

Jan Beise — Fri, 14 Mar 2008 09:53:00 GMT

This paper by the British Office of National Statistics (2006) sets out a structure for managing and delivering quality in the 2011 Census. It is a high-level document and as such presents a general framework for developing and implementing processes to define, manage, and measure the quality of the 2011 Census. It is not intended as a specification, or to define specific quality standards or how they will be measured. Lessons about quality learned from the 2001 Census are discussed since they have an important role in defining the approach to managing quality in 2011.

The 2000 Round of Population and Housing Census in the ESCAP Region: Lessons Learned and Emerging Issues

Jan Beise — Thu, 13 Mar 2008 10:47:00 GMT

This discussion paper by Luisa T. Engracia (UNESCAP) reviews the experiences of countries in the ESCAP region in the implementation of the 2000 Round of Population and Housing Census and tries to bring out some important issues that came to light in the process and which are likely to bear on the next round, such as: quality of census data; use of information technology; timeliness of data; analysis, utilization and dissemination of census data; and indicators for monitoring Millennium Development Goals in the context of the 2010 round of population and housing census.

Paper presented at the Expert Group Meeting on Population and Housing Censuses9-10 December 2004, Bangkok, Thailand.

A Comprehensive View of Capture Data Quality

Jan Beise — Mon, 21 Jan 2008 05:52:00 GMT

This paper by Peter Smith and others presents a comprehensive view of data quality and how these approaches will be used in the US 2010 Census. The overall data quality problem is discussed in terms of five key characteristics (Inventory Control, Questionnaire Integrity, Image Quality, Processing Integrity, and Data Capture Accuracy) and four response modes (Paper, Telephone, Internet, and hand-held device). Selected characteristics and modes are discussed in terms of their relevance, issues, quality interpretation and techniques to address data quality. For each characteristic and mode topics to be addressed include response data characteristics, modeling, processing definition, metrics collection and analysis approach.

Paper presented at the Satellite Meeting on Innovative Methodologies for Censuses in the New Millennium of the 56th Session of the ISI, Lisboa, 31 August - 2 September 2007.

Towards an integrated model of management and quality control for population and housing census at the local administrative level

Jan Beise — Mon, 10 Sep 2007 07:42:00 GMT

This paper by the National Statistical Institute of Portugal presents the quality assurance and control programme that was instituted for the 2001 census in order to ensure effective control of the whole production process. It discusses what alert indicators were used and how the involvement of the local government in census operations makes it possible to detect problems with data collection and take corrective measures in a more efficient and timely manner.

Paper presented at the Joint UNECE-Eurostat Expert Group Meeting on the Implementation of the CES Census Recommendations for Register-Based Censuses Astana (Kazakhstan), 7-8 June 2007.

[available also in Russian]

Census planning and quality assurance strategy in a register-based census

Jan Beise — Mon, 10 Sep 2007 07:35:00 GMT

This paper by Statistics Sweden describes the introduction of the population register-based census in Sweden. The census of 1990 was the last time that Sweden used any questionnaires in population and housing census. All subsequent censuses will be completely based on data that are available in numerous registers. The paper discusses a quality assurance strategy for a register-based census.

Paper presented at the Joint UNECE-Eurostat Expert Group Meeting on the Implementation of the CES Census Recommendations for Register-Based Censuses Astana (Kazakhstan), 7-8 June 2007.

[available also in Russian]

Quality management in the French census

Jan Beise — Mon, 10 Sep 2007 07:28:00 GMT

This paper by the Institut National de la Statistique et des Études Économiques (INSEE) deals with the redesigned census (rolling census) in France. The redesigned census is based on a five-year cycle rolling data collection and use of sample surveys in municipalities with 10000 inhabitants or more. The paper describes how quality for each of the operations of the census is assessed and operationally managed both at the level of the municipality and also by the national statistical institute.

Paper presented at the Joint UNECE-Eurostat Expert Group Meeting on the Implementation of the CES Census Recommendations for Register-Based Censuses Astana (Kazakhstan), 7-8 June 2007.

[available also in Russian]

Quality assurance during census field operations

Jan Beise — Mon, 10 Sep 2007 07:20:00 GMT

This paper describes quality assurance during census field operations. Quality assurance in this context is said to be rather directed towards identifying problem enumerators within the enumeration workforce than systematic or process errors. The paper addresses the issues of the role of supervisors, observing interviews, checking households already enumerated, checking coverage of the enumeration area, and of checking completed census forms.

The paper is an excerpt from: United Nations (2001). Handbook on Census Management for Population and Housing Censuses. Studies in Methods, No. 83, Revision 1.

Register-based 2002 Census of Population, Households and Housing in Slovenia: organization and data processing in the view of outsourcing

Jan Beise — Fri, 07 Sep 2007 10:24:00 GMT

This paper by the Statistical Office of the Republic of Slovenia describes the experience made in Slovenia by outsourcing some important part of the census activities including services in the preparatory phase, organization of field enumeration and in particular in the area of data processing. The outsourcing was deemed as success in the context of the Slovenian census despite initial concerns about quality and confidentiality of the data.

Paper presented at the 2006 plenary session of the Conference of European Statisticians (CES) in Paris (13-15 June 2006).

[also available in French and Russian]

Evaluating the 2001 UK Census (United Kingdom)

Jan Beise — Wed, 22 Aug 2007 06:29:00 GMT

This paper by the the Office for National Statistics (ONS) in England and Wales reports about the Census Evaluation Programme of the 2001 UK Census. The aim of this evaluation programme has been to get to a systematic and continuous review of procedures and operations of the UK 2001 Census programme. Each stage of the process of planning and implementation has been subject to rigorous analysis, and the evaluation and findings will form an essential part of the metadata/support material compiled. The paper focuses on a number of the more significant issues that arose during the 2001 Census, and, in particular, those that arose from the innovations introduced, and which may have implications for planning any future census.

Paper presented at the Joint ECE-EUROSTAT Work Session on Population and Housing Censuses (Ohrid, The former Yugoslav Republic of Macedonia, 21-23 May 2003).

Quality assurance for census data processing

Jan Beise — Mon, 06 Aug 2007 11:31:00 GMT

This paper addresses the issue of quality assurance in the context of processing of census data. It describes the philosophy of (total) quality management and develops the concept of a quality management framework to be established at a processing center.

The paper is an excerpt from: United Nations (2001). Handbook on Census Management for Population and Housing Censuses. Studies in Methods, No. 83, Revision 1.

Quality assurance in the context of printing of forms and other census documents

Jan Beise — Mon, 06 Aug 2007 11:08:00 GMT

This paper describes quality assurance in the context of printing of forms and other documents, a preparatory task of census management.

The paper is an excerpt from: United Nations (2001). Handbook on Census Management for Population and Housing Censuses. Studies in Methods, No. 83, Revision 1.

Quality assurance in the census context

Jan Beise — Mon, 06 Aug 2007 11:00:00 GMT

This paper addresses the issue of census related quality assurance in the areas of relevance, cost, timeliness, and data quality (data accuracy). The paper sees the focus of quality assurance in preventing errors from reoccurring, detecting errors easily and informing the workers so that they do not continue. It describes the role of managers in this process and discusses the possibilities of quality improvement of different aspects of census activities.

The paper is an excerpt from: United Nations (2001). Handbook on Census Management for Population and Housing Censuses. Studies in Methods, No. 83, Revision 1.