Poonam Smith-Sreen. Accountability in Development Organisations: Experiences of Women's Organisations in India.
Soomro, Ghulam Yasin
Poonam Smith-Sreen. Accountability in Development Organisations: Experiences of Women's Organisations in India. New Delhi: Sage Publications, 1995. 297 pp. Hardcover. Indian Rs 295.00.
There has been a transition in NGO activities over time, from the basic welfare orientation to various fields of development. This has been mainly due to the failure of the previous development strategies that emphasised on rates of economic growth, on the assumption that the top-down approach would initiate the process of trickle-down benefits to the poor. The assumed linkages were conceived without a knowledge of the working of the inter-linkages of cultural, social, political, and economic structural processes which led to the failure of such strategies. There appeared to be a need to evolve a workable methodology of development which could replace the top-down strategy. As a consequence of this search, a strategy of grassroots mobilisation of the recipient population was formulated so that they could take part in the process of development in terms of problem identification, project implementation, and its monitoring and evaluation. This strategy is popularly known as community participation. There are two important elements in this participatory approach: a two-way active communication between the community and the NGO to build trust and understanding, and the empowerment of the community towards decision-making for their own needs. This participatory approach is still viable and is a source of social change and development in many countries. The author of this book promises an advancement in this type of development strategy and suggests an alternative model and its method of accountability.
The success of an NGO operation in any field can only be measured by comparing the stated objectives against the absolute achievements. Accountability not only deals with financial matters by devising a procedure that ensures the allocation of funds for the planned objectives but also with the effectiveness of the project and the amount of benefits to be received by the client population. The author has rightly pointed out that the use of accountability in evaluative research includes impact accountability, coverage accountability, service delivery accountability, efficiency accountability, and fiscal and legal accountability. As for the accountability of the NGOs, the author specifically look for three types of accountability parameters which relate to Financial Matters, Project Outcomes, and Social Accountability.
The book raises these questions and attempts to provide an answer to bridge the information gap on this very crucial operational aspect of voluntary organisations, especially those which concentrate their efforts to promote the status of women by introducing income-generation schemes for economically poor women.
Divided into eight chapters overall, in the first four chapters the author develops the model of accountability, methodology, and collection of data. The fifth chapter provides examples of the application of the accountability model by undertaking four case-studies of the NGO work in India. The last three chapters offer a discussion of the experiences learnt and suggest useful applications for the sponsors.
The author appears to be a proponent of the organisation's strategic management framework which emphasises three organisational components, i.e. structure, strategies, and process. Within the structural component, the author argues that the participatory approach has not been very fruitful because it is very complex; hence it is very difficult to measure and standardise problems related to such elusive concepts as empowerment and self-reliance and their indirect effects. She, therefore, suggests the use of the concept of member-accountability. The basic argument for structural change is to make the NGOs responsible to the clients according to the needs and expectations--to the level that the clients not only perceive as the benefits accruing to them but also in the role they play in influencing the NGO activities. The concept of member-accountability holds the NGO responsible to its clients for the outcome of all decisions made by the respective management and has thus been defined: "A member-accountable organisation may be defined as one which works with its members to continuously attempt to meet the current and potential expectations of those members (page 51). The author then develops a framework to assess an NGO which is accountable to its members, based on three sets of variables pertaining to an NGO's response to its clients' problems and suggestions, power-sharing, and promptness in providing service and support.
However, there are some ambiguities which will require further discussion. The author has not advanced an argument for the abandonment of the participatory approach in favour of member-accountability, which is justified by more than the weak premise that it is difficult to measure its unobservable or indirect effects. She seems to be legitimising the fulfilment of client expectations without qualifying them. If the people attach high expectations to an NGO operation which may be unable to meet the expectations raised due to funding difficulties, the project would be declared a failure if examined by the yard-stick of member-accountability. The clients will need to be educated for the project goals, even to empower them to participate in decision-making for their own needs. A valid criticism of the participatory approach has been that it is looked at as a movement aimed at bringing about a radical social transformation and a complete restructuring of society by involving people in the Third World countries who are mostly uneducated in the decisionmaking process. Community contribution in these instances would be worthless and people can be influenced to endorse the ideology of the unregistered NGOs out of ignorance. But the reasons for the abandonment of the participatory approach, which is now viewed as a development methodology, have not been based on empirical findings. The interesting thing is that the alternative strategy essentially appears to be analogous to the participatory approach and is only conceptually different in that the idea seems to have been borrowed from the working of the co-operatives.
The author has used questionable data collection methods, which include standardised interviews administering a semi-structured questionnaire with many open-ended questions. The standard technique of interviewing utilises structured questionnaires where the wording, sequence, and the number of questions is already designed and they are not normally modified by the researcher. However, Phillips (1971) has argued that un-standardised interviews are administered through the semi-structured questionnaire where the researcher can exercise freedom of rewording the question for in-depth probing, and that this type of data collection methodology is appropriate for the case-studies, which is what the author appears to be doing. A multi-method approach was used to conduct two interviews with each respondent to assess accountability and the social and economic benefits. The author then devised indices by assigning a five-point scaling factor.
For every kind of social and economic project, the evaluation design, method, and techniques have to be conceived based primarily on the accessible information. The data collection for the evaluation of a project comes from three sources, namely, survey, case-study, and experimentation. Which data collection method is suitable for a specific project depends largely on the project objectives and the researcher's strategy whether to apply a randomised or a non-randomised technique. The author chose case-studies because the major objective appears to be the testing of hypotheses relating to the member-accountability aspects. The data, however, should reflect two aspects: it should be representative of the study area, and it should be statistically reliable.
The author's strong point appears to be a well-planned sampling design and the weak point appears to be the execution of it, especially the use of interviewing techniques. She tries to eschew any expected criticism by asserting the limitations of every similar project: "These realities of social science research have been frowned upon by the laboratory-oriented scientist because it is argued that such changes tend to reduce the objectivity of research. However, it may be argued that every research design is driven by the biases of the particular researcher. The field can be more real than the sterility of the controlled laboratory which is created by the world-view of the researcher" (page 89).
Although field research allows flexibility, it is not at the expense of the basic principles, which include the use of chance in interviewing the respondent and prior identification of the groups before sampling. Frequent changes in the sampling design and the interviewing techniques suggest that the author pre-determined an approach that would support the researcher's point of view. This creates doubt about the representativeness of the projects of income generation in India. Thus problems like the following are admitted: "There is usually no one way of preparing one random sample in the field where few NGOs keep accurate records or even any records at all. One can also be confronted by a myriad of different kinds of records in the same NGO. Field judgements will have to be used to suit the needs of the individual study for creating a representative sample" (page 91). Such problems legitimately led the author to change the sampling methodology.
The author maintains that a systematic sampling was used but the modifications were made to include members from different years and with different authority, and then the stratification was done on an urban-rural basis. The author did not realise, however, that the actual stratification sampling method had become very complex in the process. The idea obviously was to combine the information gathered from the survey, participant and non-participant observation, and group discussions to provide appropriate answers. Yet, this is at the cost of selectivity bias and loss in representativeness. One may recall Babbie's (1973) caution "that no one can observe everything; some selectivity is inevitable. To the extent that such selectivity is uncontrolled, the researchers runs the risk of amassing a biased set of observations, just as the inept survey researcher may select a biased sample of respondents" (page 109).
Smith-Sreen is quite aware of the sampling problems but has been selective in providing support to the beliefs already held by bringing in references that fit the situation in the best way. Thus, quite familiar as she is with the literature on survey research methods, including the work of Babbie, who has been heavily quoted, she looks elsewhere when it does not serve her purpose simply to support her subjective viewpoint. The result is that her data collection methods appear to be somewhat purposive, thus limiting their generalisibility.
The analysis of the results is weak. It is concluded on the basis of Pearson's correlation that there appeared to be an association between the economic benefits and member-accountability. But these findings may turn out to be spurious once the statistical controls are applied, which require application of advanced techniques of statistical analysis. However, before resorting to statistical analysis, it would be advisable, especially in the case of the scaling of variables, to run a reliability analysis; to evaluate how well the scaled items meet the standard internal consistency criterion.
The case-studies of income generation are based on four projects of different sizes. There is no problem if these projects are analysed separately, but when these projects are compared as is done in Chapter Five, a weighting of the samples appears to be in order.
It may be said finally that the model developed by the author is activity-specific and may be useful for the evaluation of income-generating NGOs but it is not generalisable for a comprehensive substitution--for the evaluation of the NGOs in fields like health, population, and education. Robust and empirically verifiable measurement of NGO accountability can be performed more efficiently by employing impact evaluation, which includes the component of cost-effectiveness and mostly takes care of the accountability concerns.
Ghulam Yasin Soomro
Pakistan Institute of Development Economics, Islamabad.
REFERENCES
Babble, Earl R. (1973) The Practice of Social Research. Belmont, California: Wadsworth Publishing Company, Inc.
Phillips, Bernard S. (1971) Social Research: Strategy and Tactics. New York: The Macmillan Company.
