I have designed a small survey for my work place of 2000 people. 200 people were selected based on gender (Male female), age group (4 25-year age groups) and working grade (a,b,c).

Out of my sample of 200, 120 surveys were returned completed.

I was recommended to first create a design weight, and weight the achieved sample (120) back up to my desired sample (200) and then weight this result back up to the population (2000).

I have consulted the following https://bookdown.org/jespasareig/Book_How_to_weight_a_survey/design-weights.html which expressed a design weight can be calculated by 1/prob of the observation being chosen.

enter image description here

This is my data:

This is my attempt at creating design weights for the survey sample and achieved sample for the groups:

enter image description here

I need to create weights which reflect gender * age group * grade.

Could anyone recommend any sources that would explain how to calculate the design weight, and explain the weight process above (achieved->sample->population) in more detail?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.