One of the biggest challenges in Marketing Study and Survey/Poll is how to make the survey results represent the general population. The solution to this challenge is assigning the marginal weights to each respondent.

The objective of the marginal weighting process is that the sum of the weights should be close to the whole population. It is not simply make the total weight sum and population close, but also the gaps bewteen the aggregated respondent weights and the population by geography and by demographics should be as close as possible. Therefore, the weighted survey results from the respondents can be considered as if coming from the whole population.

The benefits of this weighting process are not just in that it provides an expansion to the whole population, but also in that it can adjust some distortions in the respondent distributions. For example, it often happens that general male/female distribution is 49%/51% but the respondent distribution was 53%/47%. Marginal weighting process can eliminate this kind of distortion by assigning low weights to male respondents and high weights to female respondents.

SM Research has conducted multiple marginal weighting processes for the clients, including some well-known national surveys and companies. We use a special designed multiple converging iteration method, which can assign the weights by one selected geographic level (normally at the Census Subdivision or City level) and the several key demographic variables (the most common variables are Age & Gender, Households by Size, Household Income).