Hello,
Apologies for the potentially basic question—I am generally a qualitative research who is dipping back into some sample statistics that I haven’t encounter in awhile.
I am seeking to generate a sample-level margin of error for my dataset that accounts for the design effects (deff) of pweights. I am aware of the ways to generate margin of error accounting for pweights for single variables. But is there a programmatic way to generate this for your entire sample (under the 50% reported assumption)? If not, I’d appreciate any recommendations for how to account for weights and calculate the most accurate sample margin of error.
I use Stata 15 and my data set is at the individual-level with approximately 3,000 observations. Pweights have been generated that account for non response and sampling design.
Thank you,
Matt
Apologies for the potentially basic question—I am generally a qualitative research who is dipping back into some sample statistics that I haven’t encounter in awhile.
I am seeking to generate a sample-level margin of error for my dataset that accounts for the design effects (deff) of pweights. I am aware of the ways to generate margin of error accounting for pweights for single variables. But is there a programmatic way to generate this for your entire sample (under the 50% reported assumption)? If not, I’d appreciate any recommendations for how to account for weights and calculate the most accurate sample margin of error.
I use Stata 15 and my data set is at the individual-level with approximately 3,000 observations. Pweights have been generated that account for non response and sampling design.
Thank you,
Matt
