Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Use of weights from Current Population Survey, March Supplement

From   Steve Samuels <>
Subject   Re: st: Use of weights from Current Population Survey, March Supplement
Date   Thu, 15 Apr 2010 14:19:40 -0400

I think Austin's reference is to: Michael Davern, Arthur Jones Jr,
James Lepkowski, Gestur Davidson, Lynn A. Blewett (2007) Estimating
Regression Standard Errors with Data from the Current Population
Survey's Public Use File. Inquiry: Vol. 44, No. 2, pp. 211-224.


On Thu, Apr 15, 2010 at 12:51 PM, Austin Nichols
<> wrote:
> Mónica Guadalupe Chávez Elorza <>:
> Surveys, including the CPS, have pweights.  See help weights and help svy.
> They are also stratified and clustered, though the CPS does not
> identify strata and clusters.
> You can reverse engineer clusters (the most important piece), however.
> Find a paper by Davern et al in Inquiry on CPS clustering comparing
> an approximation to using the actual clustering variable.
> 2010/4/15 Mónica Guadalupe Chávez Elorza <>:
>> I am using Stata 11.0 to analyze the Current Population Survey.
>> Specifically, weights on this survey are nonintegers, therefore when
>> using frequency weigths - fw=marsupwt- sends the following message:
>> may not use noninteger frequency weights.
>> I came out with two options, but which should I used? and Why?
>> 1. To use the importance weight -iw=marsupwt-, which allows the use of
>> noninteger values, or
>> 2. To generate a new weight, rounded to the nearest integer and that
>> for those values that are negatives, values are set to zero.

Steven Samuels
18 Cantine's Island
Saugerties NY 12477
Voice: 845-246-0774
Fax:    206-202-4783

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index