Results Generated from the Questionnaire Disseminated Prior to

Results Generated from the Questionnaire Disseminated Prior to the Workshop �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

The Objective of the Questionnaire • To better understand census evaluation activities at the country level • To facilitate the sharing of experience on methods of census evaluation among countries in the region �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

Was PES conducted for 2000 census Yes : 6 No: 8 4 countries did PES for pre-2000 censuses 2 countries did not return the questionnaire • All countries involve National statistical/population census offices in planning and conducting PES • 6 countries where PES team different from census team (Australia, Bolivia, Brazil, Ecuador, Honduras, Peru) • 3 countries used same team for planning census & PES (Panama, Uruguay, Venezuela) �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

Funding • Planned as part of the census programmed in all countries except Venezuela • Funding adequate in 6 countries (Australia, Bolivia, Brazil, Honduras, Panama, Venezuela) �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

Objectives of PES • Census coverage • Characterize missed or over-counted individuals • Estimate undercount for use in estimating official resident population • Learn quality of information collected – content error • Establish ways to correct erroneous census information �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

Institutions responsible for PES and staff used • National statistical office involved in all countries • PES team different from census team in all countries except 4 (Panama, Paraguay, Uruguay, Venezuela) �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

Recruitment and training of PES staff • Three countries selected enumerators and supervisors directly from those used in census (Ecuador, Honduras, Uruguay) • Permanent staff used as supervisors in Australia and Uruguay • Selection criteria includes: Location, experience, academic qualification Recruitment examination • Training periods: 1 -7 days �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

PES sampling designs • • Multi-stage, stratification, approaches Design stages: 1 -3 (mostly 2) – – – Australia: Three stages Bolivia: Two stages with sampling proportional to size Brazil, Peru: Two stages Ecuador, Panama: Two-stages with stratification Honduras, Paraguay, Uruguay: Stratification (one stage only) �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

Sample size for 1 st stage units • In many cases EAs were the primary sampling units • Range of sample sizes 60 EAs in Honduras and 5000 EAs in Australia 1% of EAs in Uruguay and 20% of EAs in Panama 27642 individuals from 7340 households in Paraguay �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

Domains of analysis • Wide range of domains used – National; Rural; Urban – Other domains States/territories, region/province Age, sex, marital status, urban/rural, cities indigenous status, country of birth �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

Matching methods and documentation of rules • Matching method –Manual: 5 (Australia, Brazil, Honduras, Panama, Paraguay) –Computer: 1 (Bolivia) –Combination: 3 (Ecuador, Peru, Uruguay) • Documentation of rules –Yes: 6 (Australia, Brazil, Ecuador, Honduras, Paraguay, Peru) –No: 3 (Bolivia, Panama, Uruguay) �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

Field reconciliation visits Yes: 4 (Australia, Brazil, Ecuador, Honduras) No: 4 (Panama, Paraguay, Peru, Uruguay) Unknown: Venezuela �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

Some of the coverage measures used were • • • Coverage rate only (2 countries – Ecuador, Honduras) Net coverage rate (2 country – Paraguay, Peru) Coverage & net coverage rate (2 countries – Australia, Brazil) Omission rate (1 country - Bolivia) Panama did not calculate due to problems in methodology �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

For content analysis the following were considered • • Estimating undercount and over-count by a variety of demographic categories Counts of misclassification between categories Rate of net difference Index of gross change Index of net change Percentage of identically classified Error of content Agreement �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

Producing report to inform lay users • Planned to produce reports for lay users (5 countries - Australia, Bolivia, Brazil, Paraguay, Peru) • Not planned to produce reports (Ecuador, Honduras, Panama, Uruguay) • All countries that planned to produce reports did so • Most countries 4 -10 months from date of data collection to publication of PES report • The longest period to produce report was taken by Ecuador (about 2 years) �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

To achieve independence for PES: • Different interviewers and supervision arrangements (Australia, Brazil) • No lists created in Census used in PES selection or interviewing processes • Independent planning and execution (Bolivia, Ecuador, Honduras, Peru) • Confidentiality on PES sample area (Brazil) • Different periods (Uruguay) �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

Highlight of lessons learned • Australia: evaluated the potential for computer matching To be introduced in the next round (subject to successful development) • Bolivia: Methodology verification • Brazil: Improve cartography and supervision; undercount rates are used in the next PES sample design • Ecuador: Improve capacity and using imputation techniques �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

Highlight of lessons learned (continued) • Honduras: Need for adequate capacity, independence, budget and logistical support • Paraguay: Need for a separate budget, representative sample for each district and some outsourcing • Peru: Need for larger PES coverage and budget • Uruguay: Independence, high quality personnel, enough time and budget �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

PES for 2010 round • 6 countries are planning to do PES (Australia, Brazil, Ecuador, Honduras, Paraguay, Uruguay) • 5 countries are not planning to do PES (Bolivia, Chile, Costa Rica, Panama, Peru) • Will measure coverage using demographic methods (Costa Rica) �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

Some of the objectives mentioned for PES for 2010 round • Provide an independent check on Census coverage • Identify characteristics of persons missed or over-counted • Produce estimates of undercount for use in estimating resident population • Estimate content error (quality of information) • Measure coverage and quality of information at province level and other domains • Use results to correct census results, calculate population projections the �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys

END Thank You �United Nations Regional Workshop on the 2010 World Programme on Population and Housing Censuses: Census Evaluation and Post Enumeration Surveys