I am looking for studies or discussion papers which demonstrate that household-level education is a better predictor or measure of SES vs. individual/respondent-level education. Preferably as it relates to health behaviours (e.g., physical activity, diet).