The DHS Program User Forum: Child Health » How do I account for clustering within families?

Home » Topics » Child Health » How do I account for clustering within families?

Show: Today's Messages :: Show Polls :: Message Navigator

Re: How do I account for clustering within families? [message #2009 is a reply to message #2005]

Sat, 12 April 2014 17:01

Reduced-For(u)m
Messages: 292
Registered: March 2013

Senior Member

I think there is a confusion here that stems from different disciplines speaking differently, so let me clarify something.

By "clustering" I mean choosing a specification for the variance/covariance matrix of error terms that accounts for within-cluster heteroskedasticity and serial-correlation. In STATA terms, I mean something like "reg Y X, cluster(clustervar)". This is something that relates to getting your standard errors right, but will not in any way affect point estimates.

I think you're question about "controlling" for within-household effects has to do with point estimates. In that case, you may (and may not) want to include household fixed effects (or dummy variables for each household). This would mean that your estimate is based off within-household differences. It would also limit your effective sample to households with multiple children. When I say "control for household characteristics", I'm usually referring to this, which is about getting the right identifying variation for your model.

But what I was talking about before was "accounting for within-household and within-cluster similarities" in your standard errors (your estimated precision). In that case, you want to cluster at the PSU level because you get all of the benefits of clustering at HH level and the benefits of accounting for the sampling design (though not stratification, which could technically shrink your SEs back down a bit).

What the multi-level models do is, depending on what you choose, something closer to what I call "Random Effects". That is, the V/C matrix on the error terms is parametric in some way, whereas it is "nonparametric" in the clustering case. This gets technical real fast. So let me just repeat the main point:

Household fixed-effects would deal with things like selection (some parents are good, some are bad) or other omitted variable bias problems. Clustering will get the standard errors right. Two distinct problems.

That help? I can try again if it doesn't.

Report message to a moderator

[Message index]

		How do I account for clustering within families? By: vega25 on Fri, 11 April 2014 00:01
		Re: How do I account for clustering within families? By: user-rhs on Fri, 11 April 2014 14:15
		Re: How do I account for clustering within families? By: Reduced-For(u)m on Fri, 11 April 2014 14:25
		Re: How do I account for clustering within families? By: vega25 on Sat, 12 April 2014 01:26
		Re: How do I account for clustering within families? By: user-rhs on Sat, 12 April 2014 09:41
		Re: How do I account for clustering within families? By: Reduced-For(u)m on Sat, 12 April 2014 17:12
		Re: How do I account for clustering within families? By: Reduced-For(u)m on Sat, 12 April 2014 17:01
		Re: How do I account for clustering within families? By: Liz-DHS on Sun, 13 April 2014 18:48
		Re: How do I account for clustering within families? By: Reduced-For(u)m on Sun, 13 April 2014 19:09
		Re: How do I account for clustering within families? By: Liz-DHS on Thu, 17 April 2014 12:42
		Re: How do I account for clustering within families? By: Liz-DHS on Fri, 18 April 2014 13:51
		Re: How do I account for clustering within families? By: Liz-DHS on Thu, 24 April 2014 13:28
		Re: How do I account for clustering within families? By: vega25 on Wed, 03 December 2014 20:33
		Re: How do I account for clustering within families? By: user-rhs on Wed, 03 December 2014 20:43
		Re: How do I account for clustering within families? By: ahmed89o on Sat, 09 July 2016 17:14
		Re: How do I account for clustering within families? By: user-rhs on Sat, 09 July 2016 22:35
		Re: How do I account for clustering within families? By: ab803 on Thu, 21 September 2017 21:37
		Re: How do I account for clustering within families? By: Reduced-For(u)m on Sun, 24 September 2017 15:20
		Re: How do I account for clustering within families? By: ab803 on Mon, 25 September 2017 00:43
		Re: How do I account for clustering within families? By: Reduced-For(u)m on Mon, 25 September 2017 14:26
		Re: How do I account for clustering within families? By: ab803 on Tue, 26 September 2017 01:24
		Re: How do I account for clustering within families? By: Reduced-For(u)m on Thu, 28 September 2017 16:44

Previous Topic:	Ethiopia treatment sources: more detail on country-specific labels?
Next Topic:	analysis of preceding and succeeding birth interval in children's data file

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Thu Dec 12 13:55:11 Coordinated Universal Time 2024