Higher education means a better life

The argument

The benefits of university are not just economic. Those with a university degree are more likely to perform well in other metrics that indicate a happy, stable life. Those with university degrees are more likely to have a successful first marriage. They are also less likely to be living on the streets. They are also healthier than non-college educated workers. College graduates are less likely to smoke, more likely to do regular exercise, and less likely to be obese.

Context

Those that have completed higher education have happier, more stable lives.

Premises

Counter-arguments

The correlations are real and the causal claim does not follow from them. Graduates differ from non-graduates before they enrol — in family income, parental education, school quality and childhood health — and every one of those independently predicts marital stability, housing security, exercise and not smoking. Separating the effect of the degree from the effect of who obtains one requires designs the argument does not invoke, such as comparisons of applicants just above and below admission thresholds, or of siblings; studies using those methods generally find smaller effects on non-economic outcomes than the raw gaps suggest. The homelessness example illustrates the difficulty especially clearly. Living on the streets tracks mental illness, addiction, experience of care and the absence of family support, all of which also strongly predict not completing a degree — so the arrow plausibly runs from prior disadvantage to both outcomes rather than from education to housing. The question is finally whether college is worth it, which is comparative and requires the cost side: tuition paid, debt carried, years of forgone earnings, and the outcomes of the substantial group who enrol and do not complete, who bear the costs without the credential and are absent from graduate-versus-non-graduate comparisons altogether. Averages across all graduates also conceal wide variation by institution and field of study, which is exactly what the sibling position that it depends is built on.

Rejecting the premises

[Rejecting P1] The comparison is between groups that already differed in family income, parental education, school quality and health, all of which independently predict marital stability, housing security, exercise and smoking — so the gaps do not establish that the degree produced them. [Rejecting P2] Homelessness tracks mental illness, addiction, care experience and absent family support, which also predict not completing a degree, so causation plausibly runs from prior disadvantage to both outcomes. [Rejecting P3] Whether college is worth it is comparative and requires the costs — tuition, debt, forgone earnings — and the outcomes of those who enrol without completing, who bear the costs without the credential and are absent from these comparisons.