scholarly journals An ordinal model for analysis of years of education

2021 ◽  
Author(s):  
Benjamin Domingue ◽  
Klint Kanopka ◽  
Sam Trejo ◽  
Jeremy Freese

Years of education is a commonly used outcome variable in many lifecourse studies. We argue that such studies may derive additional insights from a treatment of years of education as an ordinal outcome rather than the standard treatment using the linear model. Via simulation, we show that the ordinal approach performs well if the linear model is actually the true model while, in the reverse scenario, estimates from the linear model may be somewhat suboptimal when the ordinal model is the true model. We use data from the Health and Retirement Study to illustrate additional insights that are readily derived from application of the ordinal model and offer a suggested workflow for future analysis.

2020 ◽  
Vol 4 (5) ◽  
Author(s):  
Roland J Thorpe ◽  
Ryon Cobb ◽  
Keyonna King ◽  
Marino A Bruce ◽  
Paul Archibald ◽  
...  

Abstract Background and Objectives Among the multiple factors posited to drive the health inequities that black men experience, the fundamental role of stress in the production of poor health is a key component. Allostatic load (AL) is considered to be a byproduct of stressors related to cumulative disadvantage. Exposure to chronic stress is associated with poorer mental health including depressive symptoms. Few studies have investigated how AL contributes to depressive symptoms among black men. The purpose of the cross-sectional study was to examine the association between AL and depressive symptoms among middle- to old age black men. Research Design and Methods This project used the 2010 and 2012 wave of the Health and Retirement Study enhanced face-to-face interview that included a biomarker assessment and psychosocial questionnaire. Depressive symptoms, assessed by the endorsement of 3 or more symptoms on the Center for Epidemiological Studies—Depression 8-item scale, was the outcome variable. The main independent variable, AL, score was calculated by summing the number values that were in the high range for that particular biomarker value scores ranging from 0 to 7. black men whose AL score was 3 or greater were considered to be in the high AL group. Modified Poisson regression was used to estimate prevalence ratios (PRs) and corresponding 95% confidence intervals (CIs). Results There was a larger proportion of black men in the high AL group who reported depressive symptoms (30.0% vs. 20.0%) compared with black men in the low AL group. After adjusting for age, education, income, drinking, and smoking status, the prevalence of reporting 3 or more depressive symptoms was statistically significant among black men in the high AL group (PR = 1.61 [95% CI: 1.20–2.17]) than black men in the low AL group. Discussion and Implications Exposure to chronic stress is related to reporting 3 or more depressive symptoms among black men after controlling for potential confounders. Improving the social and economic conditions for which black men work, play, and pray is key to reducing stress, thereby potentially leading to the reporting of fewer depressive symptoms.


2019 ◽  
Vol 3 (Supplement_1) ◽  
pp. S194-S194
Author(s):  
Kexin Yu ◽  
Kexin Yu ◽  
Shinyi Wu ◽  
Iris Chi

Abstract Internet is increasingly popular among older adults and have changed interpersonal interactions. However, it remains controversial whether older people are more or less lonely with internet use. This paper tests the longitudinal association of internet use and loneliness among older people. One pathway that explains the association, the mediation effect of social contact, was examined. Data from the 2006, 2010 and 2014 waves of Health and Retirement Study was used. Hierarchical liner modeling results showed internet use was related to decreased loneliness over 12-year period of time (b=-0.044, p<.001). Internet use was associated with more social contact with family and friends overtime (b=0.261, p<.001), social contact was related to less perceived loneliness longitudinally (b=0.097, p<.001). The total effect of internet use on loneliness is -0.054 and the mediated effect is -0.025. The findings imply that online activities can be effective for reducing loneliness for older people through increased social contact.


Sign in / Sign up

Export Citation Format

Share Document