skip to content
When should you adjust standard errors for clustering? Preview this item
ClosePreview this item
Checking...

When should you adjust standard errors for clustering?

Author: Alberto Abadie; Susan Athey; Guido Imbens; Jeffrey M Wooldridge; National Bureau of Economic Research,
Publisher: Cambridge, Mass. : National Bureau of Economic Research, 2017.
Series: Working paper series (National Bureau of Economic Research), no. 24003.
Edition/Format:   eBook : Document : EnglishView all editions and formats
Summary:
In empirical work in economics it is common to report standard errors that account for clustering of units. Typically, the motivation given for the clustering adjustments is that unobserved components in outcomes for units within clusters are correlated. However, because correlation may occur across more than one dimension, this motivation makes it difficult to justify why researchers use clustering in some  Read more...
Rating:

(not yet rated) 0 with reviews - Be the first.

Subjects
More like this

Find a copy online

Links to this item

Find a copy in the library

&AllPage.SpinnerRetrieving; Finding libraries that hold this item...

Details

Material Type: Document, Internet resource
Document Type: Internet Resource, Computer File
All Authors / Contributors: Alberto Abadie; Susan Athey; Guido Imbens; Jeffrey M Wooldridge; National Bureau of Economic Research,
OCLC Number: 1011532085
Notes: "November 2017"
Description: 1 online resource (28 pages).
Series Title: Working paper series (National Bureau of Economic Research), no. 24003.
Responsibility: Alberto Abadie, Susan Athey, Guido W. Imbens, Jeffrey Wooldridge.

Abstract:

In empirical work in economics it is common to report standard errors that account for clustering of units. Typically, the motivation given for the clustering adjustments is that unobserved components in outcomes for units within clusters are correlated. However, because correlation may occur across more than one dimension, this motivation makes it difficult to justify why researchers use clustering in some dimensions, such as geographic, but not others, such as age cohorts or gender. This motivation also makes it difficult to explain why one should not cluster with data from a randomized experiment. In this paper, we argue that clustering is in essence a design problem, either a sampling design or an experimental design issue. It is a sampling design issue if sampling follows a two stage process where in the first stage, a subset of clusters were sampled randomly from a population of clusters, and in the second stage, units were sampled randomly from the sampled clusters. In this case the clustering adjustment is justified by the fact that there are clusters in the population that we do not see in the sample. Clustering is an experimental design issue if the assignment is correlated within the clusters. We take the view that this second perspective best fits the typical setting in economics where clustering adjustments are used. This perspective allows us to shed new light on three questions: (i) when should one adjust the standard errors for clustering, (ii) when is the conventional adjustment for clustering appropriate, and (iii) when does the conventional adjustment of the standard errors matter.

Reviews

User-contributed reviews
Retrieving GoodReads reviews...
Retrieving DOGObooks reviews...

Tags

Be the first.
Confirm this request

You may have already requested this item. Please select Ok if you would like to proceed with this request anyway.

Linked Data


Primary Entity

<http://www.worldcat.org/oclc/1011532085> # When should you adjust standard errors for clustering?
    a schema:MediaObject, schema:Book, schema:CreativeWork ;
   library:oclcnum "1011532085" ;
   library:placeOfPublication <http://id.loc.gov/vocabulary/countries/mau> ;
   schema:about <http://experiment.worldcat.org/entity/work/data/4612686064#Topic/correlation_statistics> ; # Correlation (Statistics)
   schema:about <http://experiment.worldcat.org/entity/work/data/4612686064#Topic/regression_analysis> ; # Regression analysis
   schema:about <http://experiment.worldcat.org/entity/work/data/4612686064#Topic/cluster_analysis> ; # Cluster analysis
   schema:about <http://experiment.worldcat.org/entity/work/data/4612686064#Topic/economics_statistical_methods> ; # Economics--Statistical methods
   schema:author <http://experiment.worldcat.org/entity/work/data/4612686064#Person/athey_susan> ; # Susan Athey
   schema:author <http://experiment.worldcat.org/entity/work/data/4612686064#Person/wooldridge_jeffrey_m_1960> ; # Jeffrey M. Wooldridge
   schema:author <http://experiment.worldcat.org/entity/work/data/4612686064#Person/imbens_guido> ; # Guido Imbens
   schema:author <http://experiment.worldcat.org/entity/work/data/4612686064#Person/abadie_alberto> ; # Alberto Abadie
   schema:bookFormat schema:EBook ;
   schema:datePublished "2017" ;
   schema:description "In empirical work in economics it is common to report standard errors that account for clustering of units. Typically, the motivation given for the clustering adjustments is that unobserved components in outcomes for units within clusters are correlated. However, because correlation may occur across more than one dimension, this motivation makes it difficult to justify why researchers use clustering in some dimensions, such as geographic, but not others, such as age cohorts or gender. This motivation also makes it difficult to explain why one should not cluster with data from a randomized experiment. In this paper, we argue that clustering is in essence a design problem, either a sampling design or an experimental design issue. It is a sampling design issue if sampling follows a two stage process where in the first stage, a subset of clusters were sampled randomly from a population of clusters, and in the second stage, units were sampled randomly from the sampled clusters. In this case the clustering adjustment is justified by the fact that there are clusters in the population that we do not see in the sample. Clustering is an experimental design issue if the assignment is correlated within the clusters. We take the view that this second perspective best fits the typical setting in economics where clustering adjustments are used. This perspective allows us to shed new light on three questions: (i) when should one adjust the standard errors for clustering, (ii) when is the conventional adjustment for clustering appropriate, and (iii) when does the conventional adjustment of the standard errors matter."@en ;
   schema:exampleOfWork <http://worldcat.org/entity/work/id/4612686064> ;
   schema:inLanguage "en" ;
   schema:isPartOf <http://experiment.worldcat.org/entity/work/data/4612686064#Series/nber_working_paper_series> ; # NBER working paper series ;
   schema:isPartOf <http://experiment.worldcat.org/entity/work/data/4612686064#Series/working_paper_series_national_bureau_of_economic_research> ; # Working paper series (National Bureau of Economic Research) ;
   schema:name "When should you adjust standard errors for clustering?"@en ;
   schema:productID "1011532085" ;
   schema:publisher <http://experiment.worldcat.org/entity/work/data/4612686064#Organization/national_bureau_of_economic_research> ; # National Bureau of Economic Research,
   schema:url <http://dx.doi.org/10.3386/w24003> ;
   schema:url <http://www.nber.org/papers/w24003> ;
   wdrs:describedby <http://www.worldcat.org/title/-/oclc/1011532085> ;
    .


Related Entities

<http://experiment.worldcat.org/entity/work/data/4612686064#Organization/national_bureau_of_economic_research> # National Bureau of Economic Research,
    a schema:Organization ;
   schema:name "National Bureau of Economic Research," ;
    .

<http://experiment.worldcat.org/entity/work/data/4612686064#Person/abadie_alberto> # Alberto Abadie
    a schema:Person ;
   schema:familyName "Abadie" ;
   schema:givenName "Alberto" ;
   schema:name "Alberto Abadie" ;
    .

<http://experiment.worldcat.org/entity/work/data/4612686064#Person/athey_susan> # Susan Athey
    a schema:Person ;
   schema:familyName "Athey" ;
   schema:givenName "Susan" ;
   schema:name "Susan Athey" ;
    .

<http://experiment.worldcat.org/entity/work/data/4612686064#Person/imbens_guido> # Guido Imbens
    a schema:Person ;
   schema:familyName "Imbens" ;
   schema:givenName "Guido" ;
   schema:name "Guido Imbens" ;
    .

<http://experiment.worldcat.org/entity/work/data/4612686064#Person/wooldridge_jeffrey_m_1960> # Jeffrey M. Wooldridge
    a schema:Person ;
   schema:birthDate "1960" ;
   schema:familyName "Wooldridge" ;
   schema:givenName "Jeffrey M." ;
   schema:name "Jeffrey M. Wooldridge" ;
    .

<http://experiment.worldcat.org/entity/work/data/4612686064#Series/nber_working_paper_series> # NBER working paper series ;
    a bgn:PublicationSeries ;
   schema:hasPart <http://www.worldcat.org/oclc/1011532085> ; # When should you adjust standard errors for clustering?
   schema:name "NBER working paper series ;" ;
    .

<http://experiment.worldcat.org/entity/work/data/4612686064#Series/working_paper_series_national_bureau_of_economic_research> # Working paper series (National Bureau of Economic Research) ;
    a bgn:PublicationSeries ;
   schema:hasPart <http://www.worldcat.org/oclc/1011532085> ; # When should you adjust standard errors for clustering?
   schema:name "Working paper series (National Bureau of Economic Research) ;" ;
    .

<http://experiment.worldcat.org/entity/work/data/4612686064#Topic/correlation_statistics> # Correlation (Statistics)
    a schema:Intangible ;
   schema:name "Correlation (Statistics)"@en ;
    .

<http://experiment.worldcat.org/entity/work/data/4612686064#Topic/economics_statistical_methods> # Economics--Statistical methods
    a schema:Intangible ;
   schema:name "Economics--Statistical methods"@en ;
    .

<http://experiment.worldcat.org/entity/work/data/4612686064#Topic/regression_analysis> # Regression analysis
    a schema:Intangible ;
   schema:name "Regression analysis"@en ;
    .


Content-negotiable representations

Close Window

Please sign in to WorldCat 

Don't have an account? You can easily create a free account.