Click here to return to the report index
Mapping Indicators of Poverty in West Africa

5. Interpretation

The study aims at presenting data in an easily interpretable format both as graphs and on maps. Our intention is to create a solid base for discussion and to leave interpretation to the readers who may have different perspectives. Nevertheless we provide a general interpretation of the results below. It has to be emphasised that additional factors not analysed in the study are of relevance to the evaluation of poverty. Some factors such as educational attainment levels do not seem to correlate with regional or zonal spatial factors, but rather with political or social conditions inherent in the individual countries.


AGRO-CLIMATIC ZONES:

These zones show a clear north-south trend of decreasing aridity (higher rain-fed agroclimatic potential). Of the HDI indicators, three show a certain degree of correlation with aridity (Child Mortality, Adult Literacy and School Enrolment). One could suggest that the lower the agroclimatic potential, the poorer the people. No trend is visible with 'stunted children'.


LAND DEGRADATATION:

Interpretation of the indices correlated with land degradation is somewhat difficult. No immediate trends are visible. One reason for this may be the fuzziness of the land degradation data, which is rather difficult to map in discrete areas at a generalised scale. We cannot take these results to devalidate the hypothesis 'the more degraded the land the worse is the human development (poverty) level'. A deeper look into land degradation classification will, however, be necessary. The relatively high values of educational indicators in highly degraded land may point to rural areas close to cities, but also here, more in-depth research is needed for verification.


POPULATION DENSITY:

The main observation is the relationship between population density and educational attainment, both for adult literacy and primary school enrolment. This suggests that larger population centers may offer better services for education. Adjacent rural populations may also benefit from these services. No obvious patterns are present for th eindicators stunting and child mortality.


ACCESSIBILITY:

Data was only available for 2 countries, all situated in the more arid part of West Africa. The common-sense logic 'the better accessible the more developed' seems to be confirmed through the life expectancy indicator. The other indicators do not show very clear trends. The general low levels of female literacy in this part of West Africa are striking, regardless of accessibility.


6. Discussion

Some trends can clearly be read from the results. There are, however, limitations to a detailed interpretation. The validity of such a broad approach is debatable, for several reasons:

  • Definition of poverty: The definition of poverty is based on a combination of causes, and the spatial distribution of 'Poor People' will vary considerably depending on the definition chosen. Therefore, the maps cannot be looked at as absolute truth, but in the context of the approach applied.

  • Determination of relevant factors causing poverty: 'Marginal Land' in a strictly biophysical understanding is but one of the many factors causing poverty. This study tried to point at this fact by using both socio-economic and biophysical components in the analysis of factors causing poverty (low human development). If a more comprehensive study was to be carried out more attention should be payed to the analysis of the factors causing poverty, including additional ones such as ethnic patterns, foreign aid, crop use intensity and current agricultural use (cash or subsidence crop), just to name a few.

  • Data availability/data quality: Once the questions regarding the definition of poverty and factors influencing it are solved, the problem of obtaining the data necessary for a solid analysis remains. Existing data are often incomplete, outdated or unreliable. The collection of new data is unrealistically costly and time-consuming. One pragmatic reason for the choice of the parameters for this study was that the data used was available fairly easily and was of reasonable quality ('best available'). In order to improve the reliability of the analysis, more resources are needed to verify the quality and accuracy of the data.

For the results of this study in particular, the considerations above imply:

  • The level of resources available have to be taken into account when looking at the quality of the results. In this case, the time and resource constraints didn't allow for a detailed analysis, which explains the coarseness of the results.

  • The limited time span of this study didn't allow to take into account all eventual causes of a low level of human development.

  • A global study would only make sense, if it is designed broadly enough to allow for in-depth country analysis. Alternatively, detailed case studies for representative countries or regions could be carried out and the results extrapolated.


7. Recommendations

On the whole, the study has certainly provided new insights. We therefore recommend to conduct follow-up activities, in the two outlined directions:

  • Network with international organisations and research institutions involved in poverty mapping, to gain an overview over existing activities and agree on a common methodological approach. An assessment of on-going poverty mapping world-wide is underway (under the UNEP-CGIAR project umbrella).

  • Propose and seek funding for concrete collaborative poverty mapping projects in order to achieve: a wider (global) and/or a more detailed (higher resolution data) analysis.


8. References

Brunner, J., 1997 (unpublished), Human Development vs. Aridity in West Africa Preliminary Results, World Resources Institute.

Corbett, J.D., R.F. O'Brien, E.I. Muchugu, and R.L. Kruska, 1996, Data Exploration Tool, CD-ROM and Technical Documentation, International Center for Research in Agroforestry ICRAF, Nairobi.

Deichmann Uwe, 1996, Asia Population Database Documentation, National Center for Geographic Information and Analysis NCGIA, University of California, Santa Barbara, USA.

Deichmann, U. and L. Eklundh, 1991, Global Digital Datasets for Land Degradation Studies: A GIS Approach, GRID Case Study Series, No.4, Nairobi.

Ehrlich, D. and E. Lambin, 1996, Broad-scale Land Cover Classification and Interannual Climatic Variability, International Journal of Remote Sensing, Vol. 17, pp. 845-862.

Hutchinson, M.R., H.A. Nix, J.P. McMahon, and K.D. Ord., 1996, The Development of a Topographic and Climate Database for Africa, Paper presented at the 2nd International Conference on Integrating GIS and Environmental Modeling, Santa Fe, USA, January 21-25.

Oldeman, L.R., 1991, Global Assessment of Soil Degradation, draft report for the UNEP State of the Environment Report, 1992. International Soil Reference and Information Centre, Wageningen, Netherlands

Scholes, R.J., 1995, The Environmental Determinants of African Terrestrial Biomes. Presented to the MEDIA International School, Nairobi, for special issue of Geo-Science, January 1996

UNDP, (1996), Human Development Report 1996, UN Development Programme, Oxford University Press.

UNEP, 1992, World Atlas of Desertification

U.N. ACC/SCN, 1992, Second Report on the World Nutrition Situation, Vol. 1, Global and Regional Results, Geneva.

van Velthuizen, Harry and Luc Verelst, 1995, Crop Production System Zones of the IGADD Sub-Region, Agrometeorology Working Paper Series No.10, FAO, Rome.