Weighting

In order to make the sample look more like the broader population, sampling is used. It is a very common method that has been used for a long time now. When the sample we have is not quite a representation of the real world or the broader population of interest, then weighting methods can be used for making the sample to look more like the broader population. To do this larger weights are added to the individuals who are underrepresented in the sample and a lower weight to those who are over-represented.

• Example: If the female population is underrepresented, say 20%, whereas in general the female population is almost half i.e 50% then probably we would want to increase the weight assigned to females in this case.

"Assign a weight" to every individual is the strategy of this method.

Note: The aim of this method is to remove the sampling bias.

Problems:

The issue with using this method is that, in a case when the propensity score is very low(almost 0) in that case the weight is drastically increased.