Project

General

Profile

Support #877

weights for pooled cross-sections over waves (a)-(f)

Added by Nico Ochmann almost 2 years ago. Updated over 1 year ago.

Status:
Closed
Priority:
High
Assignee:
Category:
Weights
Target version:
Start date:
11/10/2017
Due date:
% Done:

100%

Estimated time:

Description

Dear Peter,
I have a few more questions with regard to my original support #758.
1. When I generate my new weighting variable, newwgt, 600 observations in my estimation sample are assigned a zero meaning they are dropped. Why do you guys assign zero weights in w_indinus, w_indinub, etc. ?
2. The sample mean of my newwgt is 1.1 with a s.d. of 0.58. Does this look reasonable to you as in theory the mean should be 1.
3. Does USoc normalize weights so that N observations in weighted data equal N observations in unweighted data? It seems to me that you guys do that.
4. Given the normalization, how would I find the number of weighted and unweighted observations with Stata for my estimation sample?

History

#1 Updated by Peter Lynn almost 2 years ago

  • Assignee changed from Peter Lynn to Nico Ochmann
  • % Done changed from 0 to 70

Nico,

1. The provision of weights requires the ability to estimate probabilities of continuing to respond over multiple waves. This is true of cross-sectional weights as well as longitudinal ones, as they are derived from the longitudinal ones (how this was done is described in section 3.8.3.10 of the User Guide). In consequence, a person in a household where there is no person who has been enumerated at every wave up to wave w will get a weight of zero. Such people should not be given a weight, as the weights for all other sample members are calculated in a way that compensates for these "missing" people.
2. If you are pooling all people with a non-zero value of the respective weight in each of the six waves, the mean of newwgt would be 1.0. But it looks like you have dropped some cases of some kind?
3. Yes, this is the final step in the derivation of each weight, as described in the User Guide. Thus, the mean of each weight, amongst cases with a non-zero value, is always 1.0.
4. sum newwgt if newwgt>0 - 'Obs' is the unweighted number of observations. total newwgt if newwgt>0 - 'Total' is the weighted number of observations.

Peter

#2 Updated by Stephanie Auty almost 2 years ago

  • Status changed from New to Feedback
  • Private changed from Yes to No

#3 Updated by Andrew Brown almost 2 years ago

Hi

Could I ask a related question to this?

For those cases not assigned a weight because 'a person in a household where there is no person who has been enumerated at every wave up to wave w will get a weight of zero. Such people should not be given a weight, as the weights for all other sample members are calculated in a way that compensates for these "missing" people'

How could/ should they be included in any analysis as SPSS 'makes them invisible' - could they be assigned the mean weight of 1?

Many thanks

Andrew

#4 Updated by Stephanie Auty almost 2 years ago

From: Nico Ochmann [mailto:]
Sent: 10 November 2017 18:24
To:
Subject: RE: [Understanding Society User Support - Support #877] weights for pooled cross-sections over waves (a)-(f)

Hi Peter,

thank you very much for the quick and informative reply.

I appreciate your help over and over again.

Best wishes.

Nico

#5 Updated by Stephanie Auty almost 2 years ago

  • Status changed from Feedback to Resolved
  • % Done changed from 70 to 100

#6 Updated by Stephanie Auty over 1 year ago

  • Status changed from Resolved to Closed

Also available in: Atom PDF