A life course perspective on predictors of midlife socioeconomic status

BACKGROUND: Previous studies have found paternal occupation, childhood intelligence, and educational attainment to be important predictors of socioeconomic status (SES) later in life. However, these factors only explain part of the variance in SES and thus, it is important to identify other predictors of SES and trajectories of influence from early childhood to adulthood.

OBJECTIVES: To analyze predictors of SES attainment during the life course from early childhood to midlife with special emphasis on identifying direct and indirect effects on midlife SES of early childhood, late childhood and young adult characteristics.

METHODS: This study uses questionnaire and national registry data, including data on parental social background, intelligence, education, and midlife SES for 6,294 members of the Metropolit 1953 Danish Male Birth Cohort. The study sample included cohort members with information on intelligence at age 12 who were living in Denmark at age 50. Using structural equation modelling, direct and indirect mediated effects on midlife SES were estimated for early childhood, late childhood, and young adult characteristics.

RESULTS: Educational attainment, intelligence, parental education, and father's occupational class had the strongest influences on midlife SES. A prediction model only including education and intelligence could only be slightly improved by the inclusion of other predictors, from 53.5% to 54.1% explained variance in midlife SES. Educational attainment was a particularly strong predictor of status attainment. Other early and late childhood factors had relatively weak direct effects, but significant indirect effects. Thus, it was possible to identify trajectories of influence from early childhood to midlife.

CONCLUSION: Young adult education and intelligence were the strongest predictors of midlife SES. Early and late childhood factors influence young adult characteristics, but over the life course the direct effects of early life variables tend to decrease, and the effects on midlife SES become mediated and indirect.