INTRODUCTION TO SMALL AREA ESTIMATION …

1 ASIAN DEVELOPMENT BANKINTRODUCTION TO SMALL area ESTIMATION TECHNIQUESA Practical Guide for National Statistics OfficesMAY 2020 ASIAN DEVELOPMENT BANKINTRODUCTION TO SMALL area ESTIMATION TECHNIQUESA Practical Guide for National Statistics OfficesMAY 2020 Creative Commons Attribution IGO license (CC BY IGO) 2020 Asian Development Bank6 ADB Avenue, Mandaluyong City, 1550 Metro Manila, PhilippinesTel +63 2 8632 4444; Fax +63 2 8636 rights reserved. Published in 2020. ISBN 978-92-9262-222-0 (print); 978-92-9262-223-7 (electronic); 978-92-9262-224-4 (ebook) Publication Stock No. TIM200160-2 DOI: The views expressed in this publication are those of the authors and do not necessarily reflect the views and policies of the Asian Development Bank (ADB) or its Board of Governors or the governments they represent.

2 ADB does not guarantee the accuracy of the data included in this publication and accepts no responsibility for any consequence of their use. The mention of specific companies or products of manufacturers does not imply that they are endorsed or recommended by ADB in preference to others of a similar nature that are not making any designation of or reference to a particular territory or geographic area , or by using the term country in this document, ADB does not intend to make any judgments as to the legal or other status of any territory or work is available under the Creative Commons Attribution IGO license (CC BY IGO) By using the content of this publication, you agree to be bound by the terms of this license.

3 For attribution, translations, adaptations, and permissions, please read the provisions and terms of use at # CC license does not apply to non-ADB copyright materials in this publication. If the material is attributed to another source, please contact the copyright owner or publisher of that source for permission to reproduce it. ADB cannot be held liable for any claims that arise as a result of your use of the contact if you have questions or comments with respect to content, or if you wish to obtain copyright permission for your intended use that does not fall within these terms, or for permission to use the ADB to ADB publications may be found at : In this publication, $ refers to United States dollars.

4 ADB recognizes Korea as the Republic of design by Rhommel Tables, Figures, and Boxes ivForeword viiAcknowledgments ixAbbreviations xCHAPTER I: INTRODUCTION What is SMALL area ESTIMATION and Why Do We Need It?

5 2 CHAPTER II: DEVELOPING A SMALL area ESTIMATION PLAN Goal or Purpose of SMALL area ESTIMATION Variable of Interest Level of Disaggregation and Data Requirements Approach to SMALL area ESTIMATION and Choosing a Specific 16 Technique or Model Quality Assessment of the SMALL area Estimates Dissemination Strategy for Presentation of the SMALL area Estimates 18 CHAPTER III: DATA MANAGEMENT USING R Overview of R and RStudio Fundamentals of R Data Manipulation Using dplyr and tidyr Linear Regression in R R Packages for SMALL area ESTIMATION 45 CHAPTER IV: APPROACHES IN SMALL area ESTIMATION Direct Survey ESTIMATION SMALL area ESTIMATION Using Auxiliary Information SMALL area ESTIMATION Using Regression-Based Models 60 CHAPTER V: VISUALIZING SMALL area ESTIMATES USING R 81 CHAPTER VI: CONCLUSION 83 APPENDIXES 851.

6 Description of R packages Identified and Used in the Illustrations 862. Data Files Identified and Used in the Illustrations 873. Model Building 89 REFERENCES 95 TABLES, FIGURES, AND BOXESTABLES Basic Arithmetic Operators in R Logistical and Relational Operators in R R Packages for Importing and Exporting Data Files from 33 Different Applications R Commands for Examining Data Set R Commands for Basic Statistics Population and Magnitude of Poor Population for Each 50 Municipality in Province X Structure of Data Set for Synthetic ESTIMATION Summary of Different SMALL area ESTIMATION Methods Availability of Disaggregated Data from the Sustainable Development Goals among 6 Asian Development Bank United Nations Economic and Social Commission for Asia and the Pacific Member Recommended Sample Size for Different Levels of

7 Geographic Disaggregation Graphical User Interface for Downloading R Installing RStudio Opening the Fourth Panel of RStudio Four Main Panels of RStudio Shortcut Tools of the Editor Window in RStudio Console Window in RStudio Files Tab in RStudio Plots Tab in RStudio Packages Tab in RStudio Help Tab in RStudio Environment Tab in RStudio History Tab in RStudio Install Packages Window Installing Multiple Packages in RStudio List of Packages in RStudio 27iv || Loading the Package in RStudio Accessing Help Tab in RStudio Running Help Command in RStudio Navigating the Folder for Setting the Working Directory Setting the Working Directory Importing Data Sets in Environment Tab Illustration of Weight Reallocation from Neighboring Subdomains SMALL area ESTIMATION Process SMALL area ESTIMATION of Poverty in the Philippines and Thailand Difference Between Accuracy and Precision in Survey Sampling 17 Tables, Figures, and BoxesFOREWORDFrom 2000 to 2015, the Millennium Development Goals (MDGs) influenced global development strategies by setting concrete, time-specific, and measurable targets.

8 By 2015, the MDGs had achieved substantial progress in poverty reduction and other areas of socio- economic development. In education and health, for instance, the number of out-of-school children of primary school age and the mortality rate for children aged under 5 years had decreased since 1990. Although data for the MDGs generated intercountry comparisons across various social and economic metrics, the absence of granular data meant that they fell short in showing how disparities within each country differed over time. This offered scarce empirical evidence on which sector of a country s population advanced or trailed behind in relation to the MDGs, and provided insufficient data to inform the development of appropriate programs for vulnerable segments of the population.

9 To address this concern, the 2030 Sustainable Development Agenda pledged that no one will be left behind, and called for more granular data by measuring specific Sustainable Development Goal (SDG) indicators for various clusters of the population ( , based on income level, ethnicity, geographic area and other groups relevant to the national context).Many techniques can generate granular-level SDG data, and each strategy requires different levels of accuracy and data specifications. For survey-based estimates, data granularity implies that the survey sufficiently represents samples from each subgroup of the population. However, most national statistics offices (NSOs) in developing nations are resource-constrained and may not be able to conduct large enough surveys to generate reliable estimates for various subgroups of the population.

10 In such cases, SMALL area ESTIMATION methodologies can provide more reliable granular level estimates by borrowing strength from other data collection vehicles with more comprehensive coverage, thus artificially increasing the survey sample document serves as a step-by-step guide on how to implement basic SMALL area ESTIMATION methods and highlights important considerations when executing each technique. Brief discussions of underlying theories and statistical principles are complemented with practical examples to reinforce the readers learning process. Due to increasing popularity of usage of R among development statisticians and researchers, software implementation using R is also demonstrated throughout this guide.