Objective 1: The collection, processing, and distribution of time series micro data
This objective will be fulfilled through the following activities:
- Gathering longitudinal data on households, individuals and fields in 42 selected villages during the duration of the project
- Database development and management
- Periodical and synthesis reporting and
- Website development and management.
Data collection: The population of interest for the project embraces rural households in the poverty-ridden, semi-arid and humid tropics of India and Bangladesh. This geographic focus starts in the sandy soils of western Gujarat and extends in a southern and eastern direction through broad swathes of Maharashtra, goes east through northern Karnataka until it reaches at its southern-most point, the Rayalaseema region of Andhra Pradesh, and then runs on a northeast transect through Orissa, Bihar and Jharkhand in India, and ends to the east in the humid tropics of Bangladesh. Six of India's 20 agro-ecologies are represented in the population of interest, which contains several large geographic poverty traps.
Selection of districts and villages: Based on several relevant variables, the following states, districts, taluks/tehsils/mandals, villages and households have been selected for project implementation.
| SAT India |
Andhra Pradesh |
Mahbubnagar |
Aurepalle, Dokur |
70, 50 |
| |
|
Prakasam |
JC Agraharam, Pamidipadu |
40, 40 |
| |
Maharashtra |
Akola |
Kanzara, Kinkhed |
62, 52 |
| |
|
Solapur |
Kalman, Shirapur |
61, 89 |
| |
Karnataka |
Bijapur |
Kapanimbargi, Markabbinahalli |
40, 40 |
| |
|
Tumkur |
Belladamadugu, Tharati |
40, 40 |
| |
Gujarat |
Junagadh |
Karamdichingariya, Makhiyala |
40, 40 |
| |
|
Panchmahal |
Babrol, Chatha |
40, 40 |
| |
Madhya Pradesh |
Raisen |
Papda, Rampura Kalan |
40, 40 |
| East India |
Bihar |
Patna |
Arap, Bhagakole |
40, 40 |
| |
|
Darbhanga |
Inai, Susari |
40, 40 |
| |
Orissa |
Dhenkanal |
Sogar, Chandrasekharpur |
40, 40 |
| |
|
Bolangir |
Anlatunga, Villaikani |
40, 40 |
| |
Jharkhand |
Ranchi |
Dubaliya, Hesapiri |
40, 40 |
| |
|
Dumka |
Dumariya, Durgapur |
40, 40 |
| Bangladesh |
|
Thakurgaon |
Boikunthapur |
40 |
| |
|
Kurigram |
Rasun Shimul Bari |
40 |
| |
|
Bogra |
Dharikamari |
40 |
| |
|
Chuadanga |
Khudiakhali |
40 |
| |
|
Jhenaidah |
Niamatpur |
40 |
| |
|
Patuakhali |
Dakkhin Kabir Kathi |
40 |
| |
|
Madaripur |
Paschim Bahadurpur |
40 |
| |
|
Narsingdi |
Patordia |
40 |
| |
|
Mymensingh |
Konapara, Nishaiganj |
40, 40 |
| |
|
Comilla |
Bhabanipur |
40 |
| |
|
Chandpur |
Begumpur |
40 |
| |
|
|
Total |
1824 |
As in the original VLS, village selection has been purposive and based on modal criteria (complemented by visits to villages) from the analysis of taluka-level (sub-district) information. Respondent households have been randomly chosen from strata based on a village census listing in each selected village.
Following the selection of respondents, the design of the data collection and processing blends methods that have worked well in the past with potential improvements, some of which are being experimented with and others that were incorporated at the beginning of the studies. Several of the substantive and methods-related changes include the following:
- Testing of field-based data entry and Computer-Assisted Personal Interviews (CAPI)
- Geo-referencing of plots
- Assessment of the World Bank’s Living Standards Measurement Study (LSMS) methods to estimate consumption expenditure
- Appointment of women investigators to interview family members on nutrition, health, and gender in each of the selected villages
- Coverage of two near heterogeneous villages with a slightly reduced sample size by one resident investigator per village
- Recruitment of investigators who are agriculture graduates, but who have fewer years of training than those recruited in the past or in a scenario, such as in East India, where the supply of agricultural graduates is scarce
- Inclusion in the sample of respondents households who do not rely primarily on agriculture
- Explicit testing of periodicity in core data collection
- Use of diaries to enhance reliability of recall information and
- Investment in small-scale diagnostic methods research that compares oral estimates of respondents to field measurements and evaluate the determinants of systematic differences between household consumption expenditure and household income.
Objective 2: The assembly and integration of meso data
Activities include the assembly of secondary agricultural data and district-level database development. The assembly and integration of meso level time series data now focuses on commodity area, production, and varietal type, irrigated area, harvest prices, monthly rainfall, fertilizer consumption, agricultural wages, livestock, mechanization, and infrastructure. Additional variables include data on poverty numbers, per capita income, per capita consumption, sectoral shifts, welfare programs, weather-related information, etc.
The district database for India presently covers 512 districts from 19 states. Between 1966 and 2003, 207 new districts were carved out from existing districts. To maintain continuity in the database for time series analysis, the data for the new districts have been apportioned back to their parent districts ie, districts existing in 1966. The data for all districts (unapportioned files) is also maintained for spatial analysis. The geographic coverage has been expanded to include the northeastern states of India. In Bangladesh, such data require assembling at the district level. Digitized maps of the latest district boundaries and 1966 base add value to the database by allowing spatial and temporal analysis.
Objective 3: Data analysis and capacity building
The activities in this multi-purpose objective include:
- The strengthening of in-house institutional analytical capacity to conduct analysis on micro and meso data
- The establishment of a small-grants facility to support special-purpose studies by regional researchers in South Asia
- The dissemination of research results and publications and
- Capacity building in panel data collection and in time series data analysis for social scientists in South Asia.
This project will also deal with the following:
- Agriculture as an engine for growth
- The positive association between economic growth and poverty alleviation
- The impact of technology on poverty reduction
- The impact of climate change
- Nutritional status and income
- Social network architecture through village registries
- The economic, social, and cultural impacts of women’s Self-Help Groups
- The determinants of effectiveness in local governance
- Changes in natural vegetation and soil quality.