We conducted a study of post-partum haemorrhage among women (a risk that may lead to maternal death in some cases). Our key respondents were: ASHA workers, anganwadi workers, NGO functionaries working in the field of health, medical officers working in health facilities from block level to district hospitals, in-charges of reproductive and child health programmes, and block level programme managers. These people know a lot about the causes of maternal deaths in the field and also about what all that needs to be done to improve the situation covering larger issues such as strengthening medical facilities, improving access to transportation, women's education, strengthening community support and introduction of new drugs like introduce Mag Sulf in the programme . This helped in identifying several intervention variables.
EXPERIMENTAL METHOD
Experimental data can be used for testing specific hypotheses regarding policy initiatives. Usually this implies:
The idea is to test whether something (experiment or intervention) has an effect. If the data are collected using appropriate scales one may also use the analysis of variance and covariance techniques for analysing data.
The logic of having control and experimental groups is simple. Sometimes the outcome may actually not be due to the intervention or independent variable. It may be due to historical and environmental factors, biographical characteristics of the subjects or other factors. Therefore to ensure that the dependent variable is explained by the independent variables only, we need to compare results of intervention in experimental group with changes in the dependent variable in one another group which is similar to the experimental group in all respects except that the intervention is not made there. |
