Abstract
Different single-item and multiple-item scales are used as subjective indicators of well-being in the international arena.
However, very few cross-cultural studies exist into subjective indicators of well-being among adolescent populations. In this
study, three different multi-item scales, variations of these scales and several single items –all of them previously used
separately in international research- were tested together on 12 to 16-year-old adolescents in 4 different countries with
Latin-based languages (Argentina, Brazil, Chile and Spain). The scales are the PWI (Cummins, Eckersley, van Pallant, Vugt,
Misajon, et al., Social Indicators Research, 64, 159–190, 2003), SWLS (Diener, Emmons, Larsen & Smith, Journal of Personality Assessment, 49, 71–75, 1985) and BMSLSS (Seligson, Huebner & Valois, Social Indicators Research, 61, 121–145, 2003). The single-item scales explored are: one on overall life satisfaction (OLS), one on happiness with overall life (HOL),
Fordyce’s single-item scale, two items from Russell’s scale (2003) on core affects (CAS), one on happiness and another on satisfaction and the optional item 6 of the BMSLSS. The performance
of these scales is analyzed by taking into account overall scores from the pooled sample and scores in each country. Special
attention has been paid to any trends in scores for each subjective well-being measure across age, as up until now diverse
results have been obtained in different countries when using different instruments. In order to contribute to the debate on
happiness versus satisfaction with life specifically in adolescents’ populations, multi-group structural equation models with factor invariance constraints have
been used to model together the PWI, SWLS and BMSLSS in Brazil, Chile and Spain, with Argentina excluded due to the sample
size being too small. The HOL, OLS and age were used as predictors of the three scales. The strongest relationships were between
the BMSLSS and the HOL, the SWLS and the OLS, and between the PWI and the OLS. Age showed low but significant negative correlations
with all three scales. A second order factor analysis model has also been tested, with some limitations. Results show directions
for future exploration of a second order latent variable related to the 3 multiple-item scales, which would represent the
“macro-construct” of positive life suggested by some authors. The interest in using these scales and items for cross-country
comparison is discussed.
However, very few cross-cultural studies exist into subjective indicators of well-being among adolescent populations. In this
study, three different multi-item scales, variations of these scales and several single items –all of them previously used
separately in international research- were tested together on 12 to 16-year-old adolescents in 4 different countries with
Latin-based languages (Argentina, Brazil, Chile and Spain). The scales are the PWI (Cummins, Eckersley, van Pallant, Vugt,
Misajon, et al., Social Indicators Research, 64, 159–190, 2003), SWLS (Diener, Emmons, Larsen & Smith, Journal of Personality Assessment, 49, 71–75, 1985) and BMSLSS (Seligson, Huebner & Valois, Social Indicators Research, 61, 121–145, 2003). The single-item scales explored are: one on overall life satisfaction (OLS), one on happiness with overall life (HOL),
Fordyce’s single-item scale, two items from Russell’s scale (2003) on core affects (CAS), one on happiness and another on satisfaction and the optional item 6 of the BMSLSS. The performance
of these scales is analyzed by taking into account overall scores from the pooled sample and scores in each country. Special
attention has been paid to any trends in scores for each subjective well-being measure across age, as up until now diverse
results have been obtained in different countries when using different instruments. In order to contribute to the debate on
happiness versus satisfaction with life specifically in adolescents’ populations, multi-group structural equation models with factor invariance constraints have
been used to model together the PWI, SWLS and BMSLSS in Brazil, Chile and Spain, with Argentina excluded due to the sample
size being too small. The HOL, OLS and age were used as predictors of the three scales. The strongest relationships were between
the BMSLSS and the HOL, the SWLS and the OLS, and between the PWI and the OLS. Age showed low but significant negative correlations
with all three scales. A second order factor analysis model has also been tested, with some limitations. Results show directions
for future exploration of a second order latent variable related to the 3 multiple-item scales, which would represent the
“macro-construct” of positive life suggested by some authors. The interest in using these scales and items for cross-country
comparison is discussed.
- Content Type Journal Article
- Pages 1-28
- DOI 10.1007/s12187-011-9119-1
- Authors
- Ferran Casas, Institut de Recerca sobre Qualitat de Vida, Universitat de Girona, Pl. S. Domènec, 9, 17071 Girona, Spain
- Jorge Castellá Sarriera, Institut de Recerca sobre Qualitat de Vida, Universitat de Girona, Pl. S. Domènec, 9, 17071 Girona, Spain
- Daniel Abs, Institut de Recerca sobre Qualitat de Vida, Universitat de Girona, Pl. S. Domènec, 9, 17071 Girona, Spain
- Germà Coenders, Institut de Recerca sobre Qualitat de Vida, Universitat de Girona, Pl. S. Domènec, 9, 17071 Girona, Spain
- Jaime Alfaro, Institut de Recerca sobre Qualitat de Vida, Universitat de Girona, Pl. S. Domènec, 9, 17071 Girona, Spain
- Enrique Saforcada, Institut de Recerca sobre Qualitat de Vida, Universitat de Girona, Pl. S. Domènec, 9, 17071 Girona, Spain
- Graciela Tonon, Institut de Recerca sobre Qualitat de Vida, Universitat de Girona, Pl. S. Domènec, 9, 17071 Girona, Spain
- Journal Child Indicators Research
- Online ISSN 1874-8988
- Print ISSN 1874-897X