Skip to main content

13 to 14 December 2022 | By invitation only

Novotel Manila Araneta City, Manila, Philippines
group photo

ESCAP holds Technical R Training on LNOB analysis in the Philippines

 

On 13-14 December 2022, ESCAP conducted a two-day hybrid statistical training in the Philippines in collaboration with the Philippine Statistics Authority (PSA). The aim of the training was to build national capacity to strengthen the evidence base through “Leaving No One behind’ analysis on SDG indicators.

The training was attended by close to 40 people, among them statisticians and technical specialists representing PSA, the National Economic Development Authority (NEDA) and the Department of Science and Technology – Food and Nutrition Research Institute (DOST-FNRI).

On the first day, the training covered SDG 7 with a focus on access to clean and affordable energy. Trainees explored the inner workings of the statistical LNOB code that feeds cutting-edge results on inequality at national and sub-national levels to the ESCAP LNOB platform. In addition to reproducing results, trainees also had an opportunity to see how trees would change based on model parameters.

After mastering the statistical code, the trainees on the second day focused on two other indicators currently available on the ESCAP LNOB platform, including completion of secondary education (SDG 4.4) and prevalence of intimate partner violence against (SDG 5.2). To showcase the versatility of the code, ESCAP trainers took participants to alternative data sets and jointly produced novel results for malnutrition-related outcomes based on National Nutrition Survey 2015 and employment-related outcomes based on Labour Force Survey 2020.

At the end of the training, participants had obtained a strong new tool to help them design inclusive and effective policies that leave no one behind and evaluate progress over time.

=========================================================================

  1. Background

Leaving no one behind (LNOB) is the central, transformative promise of the 2030 Agenda for Sustainable Development and its Sustainable Development Goals. LNOB means moving beyond assessing average and aggregate progress, towards ensuring progress for all population groups at a disaggregated level.

To support governments and the United Nations system in the Asia-Pacific region, the United Nations Economic and Social Commission for Asia and the Pacific (ESCAP) has developed a user-friendly diagnostic tool called Leave No One Behind (LNOB) platform. The tool is used to improve the understanding of how various circumstances intersect and create inequalities in access to basic opportunities covered by the Sustainable Development Goals. Building on empirical methodologies such as Classification and Regress Trees (CART) and the Dissimilarity index (D-index), it uses data and statistics at national and subnational levels to identify groups left furthest behind and the circumstances they share.

In this technical R training, ESCAP took participants through the inner workings of its LNOB platform, using the statistical R code that prepares, analyzes and produces evidence on 16 SDG indicators based on the Philippines Demographic and Health Survey (DHS) from 2017. In addition to replicating the results hosted in LNOB platform, participants learned how to operationalize the LNOB methodology on indicators other than those currently available on the platform and to carry out LNOB analysis, suing other nationally representative surveys such as National Nutrition Survey (2015) and Labor Force Survey (2020).

  1. Training Objectives

The primary objective of this statistical training was to strengthen national capacities in building the evidence base to developing policies that reduce inequality of opportunity and accelerate the implementation of the Sustainable Development Goals, by making sure that those left behind are also included.

At the end of the training, technical officials and specialists were able to map out inequalities in access to opportunities and prevalence of barriers for which they had data and identified furthest behind and furthest ahead and quantify the level of inequality at national and sub-national levels through the D-Index.

  1. Target Audience

The statistical training drew participation from technical officials and specialists in ministries, departments and agencies, academia, think-thanks, research institution and the United Nations system entities. Given the complexity of the code and the short duration of the training, a basic level of understanding of and experience with statistical software programmes such as R and Stata were needed to follow the instructions. We recommended that participants follow our guidelines for installing and preparing the R Studio environment before the training begins. It might be beneficial, but was not required, for participants to review basic R skills which could be found by a simple search on the internet.

  1. Organization and Participation

The statistical training was organized physically at the Novotel Manila Araneta City in Manila, on 13-14 December 2022. Local transportation for participants was covered by ESCAP. The meeting was conducted in English. The tentative programme of the training is available on the Programme tab.  

13 Dec 2022
09:00
09:15
Opening remarks
  • ESCAP representative
  • PSA Representative
  • UN RCO Representative
09:15
09:30
Self-introduction of experts
09:30
11:10
Session 1: Theory behind LNOB Methodology

This session reviews the theoretical background of LNOB methodology and provide a working understanding of the Classification and Regress Trees (CART) and D-index methodologies.

  • Theory I: CART: Presented by Selsah Pasali, ESCAP
  • Theory II: D-index: Presented by Areum Han, ESCAP
  • Q&A
11:15
11:30
Coffee/tea break
11:30
12:30
Session 2: Setting up R Studio working environment

This session will focus on setting up R studio on each participants computer and become familiar with R and its different working environments. Participants will also be guided through basic R syntax which will prepare them for the LNOB code in the training.

12:30
13:00
Session 3: Getting Ready for the LNOB Analysis

This session reviews the set-up for the LNOB code, organizes input files and gives background to using R for LNOB analysis.

13:00
14:00
Lunch
14:00
15:30
Session 4: Reproducing LNOB results at the national Level

This session begins the LNOB R code where our independent (circumstances) and dependent (indicators) variables are defined. Using DHS data, participants will create a focused data frame which includes data for the defined variables, with the intention of learning how to sort through and manipulate large amounts of data in R. Participants will also be guided to explore different ways to use R to produce simple results including access rates of different population groups. For this first indicator we will focus on access to Basic drinking water.

Once data are prepared, participants will run the CART algorithm and D-index calculations at the national level for the Basic drinking water indicator and its defined circumstances. Results will be verified on the https://lnob.unescap.org/ platform.

15:30
15:45
Coffee/tea break
15:45
16:45
Session 5: Reproducing LNOB results and the use of additional factors

In continuation of our national results, participants will learn how to add additional factors to the CART algorithm for the defined indicator, access to Clean Fuel (SDG 7.1.2). This includes regional level analysis and additional circumstances that may be considered, including but not limited to Religion, Ethnicity and Language.

16:45
17:00
Closing of Day 1 by ESCAP
12 Dec 2022
09:00
10:30
Session 6: Complex LNOB Indicators

In this session participants will experience running the full analysis, including at national and regional levels, and by using additional circumstances, for the more complex indicator, Completion of secondary education (SDG 4.1.1). The composition of the circumstances will be discussed in more detail.

10:30
10:45
Coffee/tea break
10:45
12:15
Session 7: LNOB Barrier Indicators

The LNOB platform treats Opportunities and Barriers differently. This session will focus on exploring two barriers from the LNOB platform, Any type of Violence against Women in the past 12 months (SDG 5.2.1). Since child anthropometrics are not available in DHS 2017, prevalence of stunting (SDG 2.2.2) will be investigated based on National Nutrition Survey (2015). Here the participant will learn how to adapt the LNOB methodology to apply it to barriers while maintaining comparability with opportunities.

12:15
13:15
Lunch break
13:15
14:45
Session 8: New dataset and new indicator

The LNOB platform uses MICS and DHS as its primary data source, given its comparability across Asia and the Pacific. However, the LNOB methodology can be applied to any dataset. This session will use the Philippine Labor Force Survey (LFS) 2020 to run the CART algorithm and D-Index calculations for several indicators that proxy decent employment.

14:45
15:00
Coffee/tea break
15:00
15:45
Session 9: Next steps, when to use the LNOB analysis

Voluntary National Reviews, Common Country Assessments, national studies are just a few examples of how the LNOB methodology can be integrated. This session will discuss ways the methodology has been used and can be used in the future.

15:30
16:30
Session 10: Q&A, Discussion and Feedback

Before closing the training, participants will have a chance to ask questions on the methodology, the code, or general questions about LNOB work. This session will also give a chance for participants to provide feedback or ask for any additional inputs from the ESCAP team.

16:30
17:00
Closing Remarks by ESCAP

for more information, please contact

Social Development Division +66 2288 1234 [email protected]
Print this page