Smoking's Relationship to Education Disentangled

Gallup's annual update on consumption habits in the U.S. confirms the continuing education gap in self-reported smoking. Overall, 16% of the U.S. adult population reports smoking a cigarette within the last week, but that jumps to 28% among those with a high school degree or less -- and shrinks to 5% among those with postgraduate education.

This educational gap is not a new phenomenon, at least not in recent decades. Gallup trend data from the past two decades show a consistent educational gap -- although the good news is that smoking has decreased across all groups. In 2001-2003, for example, 31% of adults with a high school education or less smoked, compared with 11% among those with postgraduate education.

By way of comparison, this is the inverse of the pattern we observe with alcohol consumption. Drinking (at least occasionally) is significantly higher among college graduates and those with postgraduate education than those with high school or less. Self-reported use of marijuana shows less difference by education, although it is slightly lower among postgraduates than among all others.

What are the reasons behind the educational gap in smoking?

One seemingly plausible explanation might center on the assumption that Americans with more education are more likely to be exposed to or to become more appreciative of scientific data showing the correlation between smoking and negative health outcomes. But there isn't a lot of support for that hypothesis in the data.

For example, a few years ago, Gallup asked Americans about the harmfulness of smoking. About nine in 10 Americans with college degrees said that smoking is very harmful. But the percentage who say smoking is very harmful among those with high school degrees or less is 81% -- lower, but not substantially so. In other words, if all Americans suddenly became college graduates, the perception of the harmfulness of smoking would rise marginally, but not significantly. Most Americans with lower levels of education are already aware of smoking's negative effects on health, so going to college isn't going to have much effect on that front.

Other, more complex research studies also cast doubt on the hypothesis that increasing those getting college degrees in American society would significantly decrease smoking. Smoking behavior appears to begin before a young person has a chance to get higher education. Childhood factors apparently lead to educational attainment and smoking behavior without a necessary causal link between the two. The environment in which one grows up -- including parents and peers -- appears to be the significant factor for smoking and educational attainment. If everyone in their late teen years suddenly went off to college to get a BA degree, it would be too late for most to affect their smoking behavior.

Here's the conclusion from one study published in the International Journal of Epidemiology¹:

A substantial portion of the education differential in smoking that has been repeatedly observed is attributable to factors shared by siblings that contribute to shortened educational careers and to lifetime smoking trajectories. Reducing disparities in cigarette smoking, including educational disparities, may therefore require approaches that focus on factors early in life that influence smoking risk over the adult life span.

And another important study reached a similar conclusion²:

Differences in smoking by completed education are apparent at ages 12-18, long before that education is acquired…. Overall, educational inequalities in smoking are better understood as a bundling of advantageous statuses that develops in childhood, rather than the effect of education producing better health.

The key is the final sentence above -- the fact that smoking and education are both caused by a third factor (advantageous statuses), rather than a pattern by which education produces a lower probability of smoking.

One of these advantageous statuses is parental smoking behavior. Data show that young people whose parents smoke are much more likely to smoke themselves. As one report in the medical journal Pediatrics³ summarized: "Children of current and former smokers face an elevated risk of smoking." Another factor is peer influence and peer pressure. Young people who grow up in an environment where smoking is normative among their peers are undoubtedly more likely to take up the practice themselves.

Education and Other Health-Related Factors in Americans' Lives

We can put smoking in the broader context of the correlation between education and other health-related factors in life. Most seriously, educational attainment is significantly related to mortality. Those with lower levels of educational attainment are significantly more likely to die early than those with higher levels of educational attainment. And research shows that a significant part of this mortality gap is related to what health researchers call "preventability causes," including smoking.

Princeton economists Anne Case and Sir Angus Deaton (the latter is a Gallup Senior Scientist) have been leading investigators of the relationship between education and mortality, most recently in their important and insightful book, Deaths of Despair and the Future of Capitalism, published last year. Case and Deaton summarize, "We have seen the divide between those with and without a four-year college degree, with a whole range of bad outcomes up to and including death being visited on those with less education." And as Case and Deaton concluded in an update published this year, "Without a 4-year college diploma, it is increasingly difficult to build a meaningful and successful life in the United States."

Smoking can thus best be characterized as one of a group of behaviors that are more common among those with lower levels of education than those with more education -- behaviors that affect health and can significantly affect long-term mortality.

Certainly, most smokers are already aware of the negative consequences of their habit. Gallup research shows a modest gap between smokers and nonsmokers in acceptance (or willingness to admit) of the harmful effects of smoking, but 66% of smokers say smoking is very harmful to health and another 17% say it is somewhat harmful. Other Gallup data show that two-thirds or more of smokers would like to quit but haven't been able to at the time of the interview. Further education for these smokers would seemingly only confirm what they already know.

Bottom Line

Data on smoking and education confirm that educational inequality in U.S society today is significantly related to health outcomes. The intergenerational transmission of this inequality occurs as Americans are growing up, with some young people moving toward a path of higher educational attainment and better health outcomes, and others toward less education and worse health outcomes.

This is not surprising. It would be much more unusual if data showed that each new generation of children essentially starts from scratch when they establish their socioeconomic and health trajectories. Clearly children raised in privileged households benefit from a host of advantages beyond income, while just as clearly on the other side of the coin, children raised in less privileged environments get fewer advantages. As discussed in this article, one of these advantages is avoidance of exposure to tobacco and another is seeing college attendance as the norm.

How society intervenes in this process and attempts to shield children in less privileged homes from exposure to tobacco - as well as level the playing field when it comes to other advantages -- is extremely difficult to figure out -- even if this is considered a desirable objective. The most logical course of action would be to improve the circumstances in which children whose parents are less advantaged are raised, perhaps including the types of early pre-schooling and childcare now being discussed in Washington.