IS NGO DEVELOPMENT ASSISTANCE MISTARGETED? AN EPISTEMOLOGICAL APPROACH

Abstract

An empirical analysis of the relationship between “objectively assessed need” and the formation of NGOs in Andhra Pradesh, India, suggests that the former is not a significant trigger for the latter. The formation of NGOs appears to be a response to perceptions of need that are so far removed from the local level that they fail to prioritize amongst levels and locations of poverty. In the absence of objective feedback mechanisms, the global—and therefore indiscriminate—perspective of international funding sources may help explain the lack of correspondence between need and NGO location, combined with the inevitably limited and blinkered knowledge of regional disparities that can be obtained through local NGOs. This preliminary “epistemological” approach to the question of the sub-optimal formation of NGOs is buttressed by consideration of the factors that may bring perception, reality, and organizational responses into better alignment when natural disasters occur, namely the factors that make natural disasters compelling news stories, unlike chronic underdevelopment.

Notes

1. Based on Planning Commission of India database on Voluntary Organizations, May 2004 and FCRA reports of Home Ministry, Government of India. All foreign aid figures are drawn from the Annual Reports of the Receipts of the Foreign Contribution by Voluntary Agencies, a report published by the Ministry of Home Affairs, Government of India on the international funds received by all registered NGOs and other nonprofit organizations. The government has used the terms NGOs and Voluntary Agencies interchangeably in its publications.

2. Organizational Websites.

3. Annual Reports of the Receipts of the Foreign Contribution by Voluntary Agencies, Ministry of Home Affairs, Government of India.

4. CAPART is an autonomous agency under the Ministry of Rural Development for providing funding and capacity building support to NGOs across the country. From the database, the year of formation and location of over 5,000 NGOs in AP were identified and after data cleaning, 4,979 NGOs were used for the analysis. Assumptions pertaining to the area of operation of the NGO and the programmatic areas have been made in the analysis. Both these assumptions have been verified through interviews with key informants and also through a phone survey. The assumptions are (1) that the NGO works in the mandal/district in which it is registered; the few that are large would be working in contiguous mandals/districts; and (2) that the NGO operates in more than one programmatic area simultaneously.

5. Sources of Data: Literacy Rate, Primary Census Abstract, 1991 for district-level indices, and Primary Census Abstract 2001 for mandal-level indices. Infant-mortality rate, per-capita income and average per-capita landholding data from a Centre for Media Studies Report. The report extrapolated infant-mortality rate at the district level based on decline in state level infant-mortality rate during 1981–91 obtained from the Sample Registration Bulletin, Registrar General of India, 1993 and Census of India 1981. Per-capita income was for 1987–88 and was computed as the sum of the per-capita value of agricultural output of 35 major crops and the per-capita value of production in industrial and small-scale sectors obtained from the Second All India Census of Small Scale Industrial Units, Ministry of Industry & Profile of Districts, 1993, Centre for Monitoring Indian Economy. Per capita average landholding was obtained from the Profile of Districts, 1993, Centre for Monitoring Indian Economy.

6. Naxalism has been a historically significant phenomenon in AP. The fact that it has a significant effect on NGO formation was discovered during field work. Naxalite areas were identified with the help of key informants and were corroborated through published reports on Naxalite activity in the state. We noticed that most of the districts that ranked lower on development indicators were also areas with strong Naxalite presence. It was possible that the effect of Naxalism was confounding with the effect of the level of development need. Therefore, a chi-square test was performed. The test could be done only for infant-mortality rate and per-capita income due to data limitations. The results of the chi-square test are significant in both cases, suggesting the confounding effect of Naxalism. In the case of infant-mortality rate, the chi square=639.285 (df=2, p=0.000) and for per-capita income, chi square=307.651 (df=2, p=0.000).

7. Table 1. The Statistical Insignificance of Need-Based Indicators for NGO Location

The table uses anova to determine whether samples from two or more groups come from populations with equal means (Hair et al. 1998). The greater the F statistic, the greater the likelihood that the two samples come from different populations. The F statistic is statistically significant only when the figure shown in column 4 is 0.05 or less.

8. Tukey's HSD Mean Difference=420.71, p=0.010.

9. r=0.284, SE =13.481.

10. See n7 above.

11. chi square=944.052, p=0.000.

12. Clustering is a phenomenon where a large number of NGOs have been formed in very close proximity to a large, pioneering, and established NGO in a manner similar to the industrial clusters. We discuss this phenomenon among the NGOs in AP separately elsewhere as part of a larger work.

13. r=0.244.

Additional information

Notes on contributors

Rahul Varman

warmly thank Jeffrey Friedman for useful suggestions on an earlier draft.