Measuring hedonia and eudaimonia as motives for activities: Cross-national investigation through traditional and Bayesian structural equation modeling
Two major goals of this paper were, first to examine the cross-cultural consistency of the factor structure of the Hedonic and Eudaimonic Motives for Activities (HEMA) scale, and second to illustrate the advantages of using Bayesian estimation for such an examination. Bayesian estimation allows for...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2014-09-01
|
Series: | Frontiers in Psychology |
Subjects: | |
Online Access: | http://journal.frontiersin.org/Journal/10.3389/fpsyg.2014.00984/full |
id |
doaj-db5ec6165ba7496bbe46c8d83f4fd59a |
---|---|
record_format |
Article |
spelling |
doaj-db5ec6165ba7496bbe46c8d83f4fd59a2020-11-24T22:39:32ZengFrontiers Media S.A.Frontiers in Psychology1664-10782014-09-01510.3389/fpsyg.2014.0098498884Measuring hedonia and eudaimonia as motives for activities: Cross-national investigation through traditional and Bayesian structural equation modelingAleksandra eBujacz0Aleksandra eBujacz1Joar eVittersø2Veronika eHuta3Lukasz Dominik Kaczmarek4Adam Mickiewicz UniversityUniversity of TromsøUniversity of TromsøUniversity of OttawaAdam Mickiewicz UniversityTwo major goals of this paper were, first to examine the cross-cultural consistency of the factor structure of the Hedonic and Eudaimonic Motives for Activities (HEMA) scale, and second to illustrate the advantages of using Bayesian estimation for such an examination. Bayesian estimation allows for more flexibility in model specification by making it possible to replace exact zero constraints (e.g. no cross-loadings) with approximate zero constraints (e.g. small cross-loadings). The stability of the constructs measured by the HEMA scale was tested across two national samples (Polish and North American) using both traditional and Bayesian estimation. First, a three-factor model (with hedonic pleasure, hedonic comfort and eudaimonic factors) was confirmed in both samples. Second, a model representing the metric invariance was tested. A traditional approach with maximum likelihood estimation reported a misfit of the model, leading to the acceptance of only a partial metric invariance structure. Bayesian estimation - that allowed for small and sample specific cross-loadings – allowed for the metric invariance model. The scalar invariance was not supported, therefore the comparison between latent factor means was not possible. Both traditional and Bayesian procedures revealed a similar latent factor correlation pattern within each of the national groups. The results suggest that the connection between hedonic and eudaimonic motives depends on which of the two hedonic dimensions is considered. In both groups the association between the eudaimonic factor and the hedonic comfort factor was weaker than the correlation between the hedonic pleasure factor and the eudaimonic factor. In summary, this paper explained the cross-national stability of the three-factor structure of the HEMA scale. In addition, it showed that the Bayesian approach is more informative than the traditional one, because it allows for more flexibility in model specification.http://journal.frontiersin.org/Journal/10.3389/fpsyg.2014.00984/fullMeasurement invarianceWell-beingBayesian structural equation modelinghedoniaeudaimonia |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Aleksandra eBujacz Aleksandra eBujacz Joar eVittersø Veronika eHuta Lukasz Dominik Kaczmarek |
spellingShingle |
Aleksandra eBujacz Aleksandra eBujacz Joar eVittersø Veronika eHuta Lukasz Dominik Kaczmarek Measuring hedonia and eudaimonia as motives for activities: Cross-national investigation through traditional and Bayesian structural equation modeling Frontiers in Psychology Measurement invariance Well-being Bayesian structural equation modeling hedonia eudaimonia |
author_facet |
Aleksandra eBujacz Aleksandra eBujacz Joar eVittersø Veronika eHuta Lukasz Dominik Kaczmarek |
author_sort |
Aleksandra eBujacz |
title |
Measuring hedonia and eudaimonia as motives for activities: Cross-national investigation through traditional and Bayesian structural equation modeling |
title_short |
Measuring hedonia and eudaimonia as motives for activities: Cross-national investigation through traditional and Bayesian structural equation modeling |
title_full |
Measuring hedonia and eudaimonia as motives for activities: Cross-national investigation through traditional and Bayesian structural equation modeling |
title_fullStr |
Measuring hedonia and eudaimonia as motives for activities: Cross-national investigation through traditional and Bayesian structural equation modeling |
title_full_unstemmed |
Measuring hedonia and eudaimonia as motives for activities: Cross-national investigation through traditional and Bayesian structural equation modeling |
title_sort |
measuring hedonia and eudaimonia as motives for activities: cross-national investigation through traditional and bayesian structural equation modeling |
publisher |
Frontiers Media S.A. |
series |
Frontiers in Psychology |
issn |
1664-1078 |
publishDate |
2014-09-01 |
description |
Two major goals of this paper were, first to examine the cross-cultural consistency of the factor structure of the Hedonic and Eudaimonic Motives for Activities (HEMA) scale, and second to illustrate the advantages of using Bayesian estimation for such an examination. Bayesian estimation allows for more flexibility in model specification by making it possible to replace exact zero constraints (e.g. no cross-loadings) with approximate zero constraints (e.g. small cross-loadings). The stability of the constructs measured by the HEMA scale was tested across two national samples (Polish and North American) using both traditional and Bayesian estimation. First, a three-factor model (with hedonic pleasure, hedonic comfort and eudaimonic factors) was confirmed in both samples. Second, a model representing the metric invariance was tested. A traditional approach with maximum likelihood estimation reported a misfit of the model, leading to the acceptance of only a partial metric invariance structure. Bayesian estimation - that allowed for small and sample specific cross-loadings – allowed for the metric invariance model. The scalar invariance was not supported, therefore the comparison between latent factor means was not possible. Both traditional and Bayesian procedures revealed a similar latent factor correlation pattern within each of the national groups. The results suggest that the connection between hedonic and eudaimonic motives depends on which of the two hedonic dimensions is considered. In both groups the association between the eudaimonic factor and the hedonic comfort factor was weaker than the correlation between the hedonic pleasure factor and the eudaimonic factor. In summary, this paper explained the cross-national stability of the three-factor structure of the HEMA scale. In addition, it showed that the Bayesian approach is more informative than the traditional one, because it allows for more flexibility in model specification. |
topic |
Measurement invariance Well-being Bayesian structural equation modeling hedonia eudaimonia |
url |
http://journal.frontiersin.org/Journal/10.3389/fpsyg.2014.00984/full |
work_keys_str_mv |
AT aleksandraebujacz measuringhedoniaandeudaimoniaasmotivesforactivitiescrossnationalinvestigationthroughtraditionalandbayesianstructuralequationmodeling AT aleksandraebujacz measuringhedoniaandeudaimoniaasmotivesforactivitiescrossnationalinvestigationthroughtraditionalandbayesianstructuralequationmodeling AT joarevittersø measuringhedoniaandeudaimoniaasmotivesforactivitiescrossnationalinvestigationthroughtraditionalandbayesianstructuralequationmodeling AT veronikaehuta measuringhedoniaandeudaimoniaasmotivesforactivitiescrossnationalinvestigationthroughtraditionalandbayesianstructuralequationmodeling AT lukaszdominikkaczmarek measuringhedoniaandeudaimoniaasmotivesforactivitiescrossnationalinvestigationthroughtraditionalandbayesianstructuralequationmodeling |
_version_ |
1725708387743694848 |