Scholarlink

<p>In recent decades, the concept of subjective well-being (SWB) has gained considerable importance and is increasingly preferred as a complement to GDP per capita as a&#xa0;measure of societal prosperity. However, for valid comparisons of this latent construct&#xa0;across different countries, regions, or cultures, establishing its measurement invariance&#xa0;is crucial. This study uses EU-SILC data from 2013 to assess whether measurement&#xa0;invariance holds for SWB with a bifactor-(S-1) model. The model incorporates cognitive, eudaimonic and affective indicators across 32 European countries. The analysis&#xa0;employs multi-group confirmatory factor analysis with WLSMV estimation, which is&#xa0;well-suited for latent factors based on ordinal indicators. Measurement invariance is&#xa0;evaluated through a four-step process: (1) configural invariance, (2) threshold invariance, (3) metric, and (4) scalar invariance. The results indicate that the proposed&#xa0;bifactor-(S-1) SWB model, is fully scalar invariant in 30 European countries, covering&#xa0;nearly all of the 32 countries included in the analysis. This finding supports factually&#xa0;accurate cross-country comparisons of covariances, regression coefficients, and averages&#xa0;of latent constructs.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Assessing Measurement Invariance of Subjective Well-Being Across European Countries

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月11日

查看全文
  • Simon Röck,
  • Lukas Kleinheinz,
  • Janette Walde

摘要

In recent decades, the concept of subjective well-being (SWB) has gained considerable importance and is increasingly preferred as a complement to GDP per capita as a measure of societal prosperity. However, for valid comparisons of this latent construct across different countries, regions, or cultures, establishing its measurement invariance is crucial. This study uses EU-SILC data from 2013 to assess whether measurement invariance holds for SWB with a bifactor-(S-1) model. The model incorporates cognitive, eudaimonic and affective indicators across 32 European countries. The analysis employs multi-group confirmatory factor analysis with WLSMV estimation, which is well-suited for latent factors based on ordinal indicators. Measurement invariance is evaluated through a four-step process: (1) configural invariance, (2) threshold invariance, (3) metric, and (4) scalar invariance. The results indicate that the proposed bifactor-(S-1) SWB model, is fully scalar invariant in 30 European countries, covering nearly all of the 32 countries included in the analysis. This finding supports factually accurate cross-country comparisons of covariances, regression coefficients, and averages of latent constructs.