世界发展银行-How-to-Frequently-and-Accurately-Measure-Poverty-and-Forest-Dependence__19页_416kb

Summary of "How to frequently and accurately measure poverty and forest dependence?"

Core Content

This paper introduces Forest-SWIFT, a new data collection tool that enables the timely and frequent measurement of poverty and forest dependence in rural forest villages. Developed as an extension of the traditional SWIFT (Survey of Well-being via Instant and Frequent Tracking) methodology, Forest-SWIFT uses a 15-question country-specific mini-survey to estimate both poverty and forest dependence. It was piloted in Turkey, a country with a significant forest-dwelling population and high poverty rates.

The tool combines consumption-based poverty estimation and forest income analysis to provide a comprehensive understanding of how rural forest communities are affected by poverty and their reliance on forest resources. The use of cross-validation and stepwise regression ensures that the models are robust and efficient, reducing the need for lengthy and costly household surveys.

Main Views

• Poverty and forest dependence are closely linked, with poorer households showing higher relative dependence on forest income.
• Traditional surveys often fail to capture the reality of forest-dwelling communities due to their remote location and limited coverage.
• Forest-SWIFT provides a cost-effective and timely alternative, using a smaller set of questions to collect accurate data.
• The tool improves the frequency of data collection, which is essential for tracking changes in poverty and its relation to forest activities.
• Data quality and consistency are maintained through rigorous validation techniques, including backward imputation and testing for normality of the error term.
• The pilot in Turkey showed promising results, with poverty estimated at 23.2% and forest dependence at 15% using the $7 per day (2011 PPP) poverty line.

Key Information

Poverty and Forest Dependence in Turkey

• Forest area: 22.34 million hectares, about 30% of Turkey’s land cover.
• Population in or near forests: ~7 million in 2016.
• Poverty in rural areas:
  • 2013: 34.9% of rural households were poor using the $7 per day (2011 PPP) poverty line.
  • 2017: 23.2% of rural forest households were estimated as poor.
• Forest dependence:
  • 2016: 28% among the lowest income quintile, 8% among the highest.
  • 2017: Estimated at 15%, using the same poverty line.
• Forest income:
  • 2016: Average forest income was TL 1223.82 (in 2016 PPP).
  • 2017: Forest income increased, and the average relative contribution to total income was 19%.

Methodology Overview

• Consumption model:
  • Based on the 2013 Household Budget Survey (HBS).
  • Uses 14 variables out of 23 potential explanatory variables.
  • Estimates poverty at 34.9% using a $7 per day poverty line.
• Forest income model:
  • Based on the 2016 Socio-Economic Household Survey (SEHS).
  • Uses 10 variables out of 25 potential explanatory variables.
  • Estimates forest dependence at 15% in 2017.
• Data collection:
  • Conducted via CAPI (Computer Assisted Personal Interviewing).
  • Survey included 20 questions on forest-related activities and household roster.
  • Survey duration was three weeks, with each household interviewed in less than 20 minutes.

Challenges and Considerations

• Questionnaire design influenced response rates, with a reduced number of forest products in Forest-SWIFT increasing participation.
• Assets like freezers and solar panels, provided through social assistance programs, were no longer correlated with wealth and thus excluded from models.
• Model accuracy was improved by using multiple imputation to generate 1000 estimates per household.
• Data comparison was limited due to the use of different poverty lines and the focus on consumption rather than income.

Conclusion

Forest-SWIFT is a valuable tool for tracking poverty and forest dependence in remote forest villages. It allows for more frequent and efficient data collection, which is essential for policy development and monitoring. The pilot in Turkey demonstrated its effectiveness and reliability, though further refinement is needed to ensure comparability with national datasets and to account for changing socio-economic conditions.