Two: How?

The study used a “comparative” or “cross sectional” design, where the effects on exposed children were sought to be established in comparison with control groups in similar demographic groups. This would mean that the children in the study groups and the control groups would be similar in all important and relevant socio-economic, genetic and cultural aspects. They would only differ in that the children in the control group were significantly less exposed to pesticides. It speaks much for the situation that it was next to impossible to locate control groups where there was no exposure to pesticides.
Challenges faced by the researchers
The design of a study like this cannot be a rigorous, experimental one because the health of human beings cannot be experimented on. Therefore it was decided that a cross-sectional approach would be adopted. Children from the cotton belt who were exposed to high levels of pesticides would be compared to children from areas where there was less exposure to pesticides.

In India, farmers in the country are known to use mixtures of pesticides, the permutations and combinations of such mixtures run into the thousands. In Punjab, the current study came across farmers using mixtures of up to four pesticides in one spray. The cocktail situation makes it difficult to calculate the impact. Inferences had to be drawn based on the known effects related to the broad classes of pesticides and consumption quantities in the study districts (mostly organophosphorus pesticides, but also mixtures of innumerable varieties).

How do we know that the child is not performing better than another child in school because s/he has strong parental support? How do we know that children in a village are doing better because their specific cultural background places high value on formal education? How do we eliminate for factors such as gender or exposure to media? These and other confounding variables needed to be eliminated to ensure that the only difference between the control group and the study group lay in their level of exposure to pesticides.

The strong underlying understanding of the situation is drawn from similar studies done earlier. Therefore, reliance on a tested methodology and known effects were used as the main pillars.

The study realizes that a multitude of factors can be used to discount the findings here. Therefore, doing it across six different states was useful, since strong common trends across the states were used to draw conclusions. For instance, in one state the control location had satellite television access and the children performed better here; in another state, the study location had access to satellite television but its corresponding control location did better in this instance too. It was obvious then that what the children had in common was their level of exposure to pesticides.

The biggest challenge was in drawing the boundaries of the study since the impacts could be many and unknown.