Performance evaluation of recent information criteria for selecting multilevel models in Behavioral and Social Sciences
Autor(a) principal: | |
---|---|
Data de Publicação: | 2014 |
Outros Autores: | , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
Texto Completo: | http://hdl.handle.net/1822/65475 |
Resumo: | This study was designed to find the best strategy for selecting the correct multilevel model among several alternatives taking into account variables such as intraclass correlation, number of groups (m), group size (n), or others as parameter values and intercept-slope covariance. First, we examine this question in a simulation study and second, to illustrate the behavior of the criteria and to explore the generalizability of the findings, a previously published educational dataset is analyzed. The results showed that none of the selection criteria behaved correctly under all the conditions or was consistently better than the others. The intraclass correlation somewhat affects the performance of all selection criteria, but the extent of this influence is relatively minor compared to sample size, parameter values, and correlation between random effects. A large number of groups appears more important than a large number of individuals per group in selecting the best model (m >= 50 and n >= 20 is suggested). Finally, model selection tools such as Akaike's Information Criterion (AIC) or the conditional AIC are recommend when it is assumed that random effects are correlated, whereas use of the Schwarz's Bayesian Information Criterion or the consistent AIC are advantageous for uncorrelated random effects. |
id |
RCAP_6e3797911b40ac23f870ba05356d42ef |
---|---|
oai_identifier_str |
oai:repositorium.sdum.uminho.pt:1822/65475 |
network_acronym_str |
RCAP |
network_name_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository_id_str |
7160 |
spelling |
Performance evaluation of recent information criteria for selecting multilevel models in Behavioral and Social SciencesModel selectionMultilevel modelsInformation criteriaMonte-Carlo studyCiências Sociais::PsicologiaSocial SciencesThis study was designed to find the best strategy for selecting the correct multilevel model among several alternatives taking into account variables such as intraclass correlation, number of groups (m), group size (n), or others as parameter values and intercept-slope covariance. First, we examine this question in a simulation study and second, to illustrate the behavior of the criteria and to explore the generalizability of the findings, a previously published educational dataset is analyzed. The results showed that none of the selection criteria behaved correctly under all the conditions or was consistently better than the others. The intraclass correlation somewhat affects the performance of all selection criteria, but the extent of this influence is relatively minor compared to sample size, parameter values, and correlation between random effects. A large number of groups appears more important than a large number of individuals per group in selecting the best model (m >= 50 and n >= 20 is suggested). Finally, model selection tools such as Akaike's Information Criterion (AIC) or the conditional AIC are recommend when it is assumed that random effects are correlated, whereas use of the Schwarz's Bayesian Information Criterion or the consistent AIC are advantageous for uncorrelated random effects.Asociación Española de Psicología Conductual (AEPC)Universidade do MinhoVallejo, GuillermoTuero-Herrero, ElliánNúñez, José CarlosRosário, Pedro20142014-01-01T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfhttp://hdl.handle.net/1822/65475engVallejo, G., Tuero-Herrero, E., Núñez, J. C., & Rosário, P. (2014). Performance evaluation of recent information criteria for selecting multilevel models in behavioural and social sciences. International Journal of Clinical and Health Psychology, 14(1), 48–57. https://doi.org/10.1016/S1697-2600(14)70036-51697-260010.1016/S1697-2600(14)70036-5info:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2023-07-21T12:06:32Zoai:repositorium.sdum.uminho.pt:1822/65475Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T18:57:14.603545Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse |
dc.title.none.fl_str_mv |
Performance evaluation of recent information criteria for selecting multilevel models in Behavioral and Social Sciences |
title |
Performance evaluation of recent information criteria for selecting multilevel models in Behavioral and Social Sciences |
spellingShingle |
Performance evaluation of recent information criteria for selecting multilevel models in Behavioral and Social Sciences Vallejo, Guillermo Model selection Multilevel models Information criteria Monte-Carlo study Ciências Sociais::Psicologia Social Sciences |
title_short |
Performance evaluation of recent information criteria for selecting multilevel models in Behavioral and Social Sciences |
title_full |
Performance evaluation of recent information criteria for selecting multilevel models in Behavioral and Social Sciences |
title_fullStr |
Performance evaluation of recent information criteria for selecting multilevel models in Behavioral and Social Sciences |
title_full_unstemmed |
Performance evaluation of recent information criteria for selecting multilevel models in Behavioral and Social Sciences |
title_sort |
Performance evaluation of recent information criteria for selecting multilevel models in Behavioral and Social Sciences |
author |
Vallejo, Guillermo |
author_facet |
Vallejo, Guillermo Tuero-Herrero, Ellián Núñez, José Carlos Rosário, Pedro |
author_role |
author |
author2 |
Tuero-Herrero, Ellián Núñez, José Carlos Rosário, Pedro |
author2_role |
author author author |
dc.contributor.none.fl_str_mv |
Universidade do Minho |
dc.contributor.author.fl_str_mv |
Vallejo, Guillermo Tuero-Herrero, Ellián Núñez, José Carlos Rosário, Pedro |
dc.subject.por.fl_str_mv |
Model selection Multilevel models Information criteria Monte-Carlo study Ciências Sociais::Psicologia Social Sciences |
topic |
Model selection Multilevel models Information criteria Monte-Carlo study Ciências Sociais::Psicologia Social Sciences |
description |
This study was designed to find the best strategy for selecting the correct multilevel model among several alternatives taking into account variables such as intraclass correlation, number of groups (m), group size (n), or others as parameter values and intercept-slope covariance. First, we examine this question in a simulation study and second, to illustrate the behavior of the criteria and to explore the generalizability of the findings, a previously published educational dataset is analyzed. The results showed that none of the selection criteria behaved correctly under all the conditions or was consistently better than the others. The intraclass correlation somewhat affects the performance of all selection criteria, but the extent of this influence is relatively minor compared to sample size, parameter values, and correlation between random effects. A large number of groups appears more important than a large number of individuals per group in selecting the best model (m >= 50 and n >= 20 is suggested). Finally, model selection tools such as Akaike's Information Criterion (AIC) or the conditional AIC are recommend when it is assumed that random effects are correlated, whereas use of the Schwarz's Bayesian Information Criterion or the consistent AIC are advantageous for uncorrelated random effects. |
publishDate |
2014 |
dc.date.none.fl_str_mv |
2014 2014-01-01T00:00:00Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://hdl.handle.net/1822/65475 |
url |
http://hdl.handle.net/1822/65475 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
Vallejo, G., Tuero-Herrero, E., Núñez, J. C., & Rosário, P. (2014). Performance evaluation of recent information criteria for selecting multilevel models in behavioural and social sciences. International Journal of Clinical and Health Psychology, 14(1), 48–57. https://doi.org/10.1016/S1697-2600(14)70036-5 1697-2600 10.1016/S1697-2600(14)70036-5 |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Asociación Española de Psicología Conductual (AEPC) |
publisher.none.fl_str_mv |
Asociación Española de Psicología Conductual (AEPC) |
dc.source.none.fl_str_mv |
reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação instacron:RCAAP |
instname_str |
Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
instacron_str |
RCAAP |
institution |
RCAAP |
reponame_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
collection |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository.name.fl_str_mv |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
repository.mail.fl_str_mv |
|
_version_ |
1799132360988426240 |