Sample weighting to match population?

[tobytang]

Dear all,

I am running an OLS over some field data, whose questions score 1-10.

My estimation is like this:
question1 c age gender education

Meanwhile, I also want to weigh my sample so that it's distributed closer to the country's population. I wish the data be two-dimensionally weighted by population age and gender. For example, 0.2% of my sample is male aged 0-14 and from census data, such group comprises 0.6% of the entire population.

I would like to know what to do to manipulate my data.

Thank you all in advance!

Toby

[startz]

If you look at the options tab after an OLS you will see it lets you specify a variable that contains weights for each observation.