Can we weight satisfaction score with importance ranks across life domains?

TY - JOUR

T1 - Can we weight satisfaction score with importance ranks across life domains?

AU - Wu, Chia Huei

PY - 2008/5

Y1 - 2008/5

N2 - The main purpose of this study was to investigate the utility of importance weighting when importance ranks were considered as the weighting values by (1) examining the range-of-affect hypothesis in the within-subject context and (2) comparing performances of weighted and unweighted satisfaction scores in predicting overall judgment of subjective well-being. Participants were 167 undergraduates at National Taiwan University. The mean age was 19.80 years (SD = 1.98). They were first asked to complete the measurements for global life satisfaction and overall QOL and then completed a QOL questionnaire for rating satisfaction, perceived have - want discrepancy on 12 life domains and ranking importance on these domains. Hierarchical linear modeling with a random-coefficients regression model was applied to examine the range-of-affect hypothesis in the within-subject context. Correlation analysis was applied to evaluate performances of weighted and unweighted satisfaction scores in predicting overall judgment of subjective well-being. Results of this study supported the range-of-affect hypothesis, showing that the relationship between item have - want discrepancy and item satisfaction is stronger for high importance items than low importance items for a given individual. Correlation analysis found that the four weighted satisfaction scores computed from the algorithms proposed by Hsieh (Social Indicators Research 61:227-240, 2003) were not superior to unweighted satisfaction score in predicting overall QOL and global life satisfaction. All these findings suggested that weighting satisfaction scores with importance ranks may not have theoretical basis and empirical contribution.

AB - The main purpose of this study was to investigate the utility of importance weighting when importance ranks were considered as the weighting values by (1) examining the range-of-affect hypothesis in the within-subject context and (2) comparing performances of weighted and unweighted satisfaction scores in predicting overall judgment of subjective well-being. Participants were 167 undergraduates at National Taiwan University. The mean age was 19.80 years (SD = 1.98). They were first asked to complete the measurements for global life satisfaction and overall QOL and then completed a QOL questionnaire for rating satisfaction, perceived have - want discrepancy on 12 life domains and ranking importance on these domains. Hierarchical linear modeling with a random-coefficients regression model was applied to examine the range-of-affect hypothesis in the within-subject context. Correlation analysis was applied to evaluate performances of weighted and unweighted satisfaction scores in predicting overall judgment of subjective well-being. Results of this study supported the range-of-affect hypothesis, showing that the relationship between item have - want discrepancy and item satisfaction is stronger for high importance items than low importance items for a given individual. Correlation analysis found that the four weighted satisfaction scores computed from the algorithms proposed by Hsieh (Social Indicators Research 61:227-240, 2003) were not superior to unweighted satisfaction score in predicting overall QOL and global life satisfaction. All these findings suggested that weighting satisfaction scores with importance ranks may not have theoretical basis and empirical contribution.

KW - Hierarchical linear modeling

KW - Importance

KW - Quality of life

KW - Satisfaction

KW - Weighting

UR - http://www.scopus.com/inward/record.url?scp=41549143223&partnerID=8YFLogxK

U2 - 10.1007/s11205-007-9180-9

DO - 10.1007/s11205-007-9180-9

M3 - Article

AN - SCOPUS:41549143223

SN - 0303-8300

VL - 86

SP - 469

EP - 480

JO - SOCIAL INDICATORS RESEARCH

JF - SOCIAL INDICATORS RESEARCH

IS - 3

ER -