Summary: In quasi-random samples of adopted Korean children, shared environment explains about 14-34% of educational outcomes, with one estimate for the top grade showing about 15.7% shared environment and about 44.3% heritability. Swedish estimates of multiple relatives indicate that income is about 27% heritable and about 5.3% due to shared environment, meaning that genes have more influence, but family environment still matters. A shared family environment has zero impact on major social outcomes, so the claim that "parents don't matter" is incorrect. Siblings raised together are consistently more similar than siblings raised separately in terms of education, cognitive and social-emotional skills, suggesting the influence of a shared environment. Adopted children raised together who are not related show remarkable similarities in education, cognitive and social-emotional skills, indicating that the family environment creates similarities even without shared genes. According to Swedish data, years of education have an estimated impact of a shared environment of about 16.4%. Adjusting studies of twins for mating still leaves a non-twin component of shared environment for education of about 10% (with a larger twin-specific shared influence). Adoption studies show that education correlates with both biological and adoptive parents, so the similarity between parents and offspring is not purely genetic. In quasi-random samples of adopted Korean children, shared environment explains about 14-34% of educational outcomes, with one estimate for the top grade showing about 15.7% shared environment and about 44.3% heritability. Many studies indicate that parental education has an impact on children's education beyond genetics, but much less than the raw parent-child correlation. Parents' field of study has a causal effect on children's choice of major and occupation, according to evidence on enrollment thresholds, indicating a direct influence of parents on life choices. Birth order reliably predicts differences in education and earnings, and evidence suggests that this is due to the postnatal environment rather than genetics. Income shows a non-zero component of shared environment in twin/sibling/adoption studies, typically in the range of ~5-20% depending on country and design. Swedish estimates of multiple relatives indicate that income is ~27% hereditary and ~5.3% due to shared environment, meaning that genes have more influence, but family environment still matters. Violent crime and delinquent behavior show a significant share of shared environment in Swedish registration studies (often around 14% and sometimes 16-30% depending on the method), in addition to a larger genetic influence. Swedish adoption data show that the breakdown of the adoptive family (divorce, death, illness) increases the criminality of adoptees, indicating environmental causes that cannot be explained by shared genes. Regarding these results, the general rule of thumb adopted in the article is that genetic influence is roughly twice as strong as the influence of shared environment, but shared environment clearly has >0.