The Thinking 

The Marriage Gap

December 7, 2010


THE NATIONAL MARRIAGE PROJECT has released its latest report, “When Marriage Disappears: The Retreat from Marriage in Middle America,” showing once again that family stability has declined significantly in the middle and lower classes, while rising modestly among the highly successful.  According to the study’s summary:

The children of highly educated parents are now more likely than in the recent past to be living with their mother and father, while children with moderately educated parents are far less likely to be living with their mother and father.

Specifically, the percentage of 14-year-old girls with highly educated mothers living with both their parents rose from 80 to 81 percent from the 1970s to the 2000s, but the percentage of 14-year-old girls with moderately educated mothers living with both parents fell from 74 to 58 percent. And the percentage of 14-year-old girls with the least-educated mothers living with both parents fell from 65 to 52 percent.

Overall, then, the family lives of today’s moderately educated Americans increasingly resemble those of high-school dropouts, too often burdened by financial stress, partner conflict, single parenting, and troubled children.

This is not surprising. Sexual liberation is far more damaging at the lower end of the economic scale, which also has been more affected by declines in male earning power. A young woman with a high school education and no college is much more likely in an era of sexual permissiveness to become a single mother while she is very young, and thus dramatically lessen her chances of stable marriage, than a college-educated woman. If she does marry, the life of the two-income family is more stressful than it is for those with nannies, cleaning services and restaurant meals. Our family standards include what G.K. Chesterton called “a plutocratic assumption.”

The report talks about growing social conservatism among the highly educated. But the evidence for this is mostly in more stable marriages (and the increases are modest). There does not seem to any evidence that the highly educated are explicitly rejecting pre-marital sexual permissiveness, which is especially destructive among the less affluent. However, among the highly educated, 50 percent now favor stricter divorce laws, according to the report.

When we talk about “highly-educated” and”moderately educated,” we obviously are not just talking about education, but intelligence. The most intelligent have destroyed the ideals and social strictures that keep the lives of the less intelligent ordered and relatively stable.

                                                       — Comments —

Mabel Le Beau writes:

The statement taken on its own that “a young woman with a high school education and no college is much more likely in an era of sexual permissiveness to become a single mother while she is very young, and thus dramatically lessen her chances of stable marriage, than a college-educated woman” might mean a0. young woman should attend college to have a happier marriage, or b) young women should go to college and wait until after their education is complete before marrying to have have children, or c) the era of sexual permissiveness promotes single motherdom, or d. marriage while young or marriage without benefit of college education results in less stable marriages.

But, why? Is availability of access to education the plutocratic exercise? Perhaps, it doesn’t necessarily follow that levels of increased intelligence = greater enrollment in higher education facilities +/- [promiscuity, single motherhood, or living in less stable marriages]

I’m inclined to think kids enrolled in college or training facilities are more likely to be busy and stay out of trouble. Along with the notion of less than full-utilization of educational resources, I have a half-formed idea that all that increased level of technology that puts people out of previously manual (blue-collar) work situations to free-up time allows for increased time to get into trouble if society isn’t fully integrated in the various levels of technology.

If we’re considering young women and education, maybe we can go out on limb and suggest a measure of the society’s level of family instability is the amount of socially accepted promiscuity i.e. child prostitution, wherein the most vulnerable members of society are victims of those that seek the services or solicit those devalued as humans to be seen as commodities?

If a young woman has a choice about entering prostitution, then she has a choice in marrying young. If women are ‘forced’ to marry young, therefore we can assume there are high levels of prostitution, and less opportunity for women to choose education.

If a society, then, wants more family stability then perhaps those that solicit the services of a child prostitute might be controlled e.g. automatic castration for the perpetrators. Is this a feminist view to discuss the educational choices available to women of child-bearing age?

Laura writes:

No, as far as your first point, I was not suggesting that more education was the answer. That’s always the feminist solutions to out-of-wedlock childbearing.

There are a number of things that differentiate women of lower intelligence and education (these things unfortunately are correlated in a society that places a high premium on female achievement and certification) from their more highly educated peers:

1. Less educated women tend to want children sooner. This seems to be because they are, as you say, less distracted and have fewer economic incentives to put off childbearing. Their attention is not diverted; motherhood is on their minds earlier.

2. Less intelligent women seem to have a greater abhorrence of abortion. This is something I would like to back up with statistics, but I cannot do so at this time. It seems to be generally true that there are increasing numbers of modertaely-educated middle class women who see no problem with pre-marital sex but who reject abortion. The anti-abortion movement has been successful in that sense. However, it has not provided the ideals that might motivate a woman to hold off on sex altogether until she is married. Those ideals can only be established with a complete rejection of feminism.

3. Less intelligent women, in a culture of sexual permissiveness, are not able to exercise restraint on their own.

I don’t fully follow your argument about prostitution. I think what you are saying is that the weaker members of society are hurt more by the dehumanization of sexuality. This is true. But women are not being forced into this. They just are not given any ideals by society’s cultural leaders that might divert them from unmarried sex. Motherhood and wifeliness are heavily sentimentalized and given plenty of shallow, materialistic affirmation, but they are not considered truly important by our culture.

Jesse Powell writes:

There is a great deal of confusion about the correlation between education levels and family breakdown.

When looking at the overall level of education in a society or when looking at the overall educational attainment of women versus men what you are looking at is the health of society overall; societies with low levels of economic development and therefore low levels of education overall tend to be societies with low levels of family breakdown and low levels of divorce (this is not an absolute unbreakable connection, by the way, but it is historically true at the current point in time). In the same way, societies where men achieve significantly higher levels of educational attainment are societies where men pursue their traditional roles as breadwinners and women pursue their traditional roles as homemakers. These societies are healthier in their family formation and have lower divorce rates. Looked at from this angle women with lower education levels, either because the society overall has low educational levels or because women earn less education than men do, have lower divorce rates. This creates an association between higher education among women and higher divorce rates. 

When looking at a particular society at a given point in time where the overall educational level of attainment among women is fixed whether a particular woman is educated or not is a class marker or a status level marker for the individual woman in question. In this situation a highly educated woman is a woman of a higher social class and of a higher intelligence level and being privileged or more functional in your upbringing or in your abilities as an individual is associated with higher functioning in one’s family life. Since divorce represents dysfunctional behavior and staying married represents functional behavior in a given society at a particular point in time women of a higher status level and a higher educational level will have a lower divorce rate. This creates an association between higher education among women and lower divorce rates. 

These conflicting associations I think is what causes the confusion as to whether women with higher educational levels get divorced more often or not. 

On the more broad issue of whether society is bifurcating by class in terms of its family life, an argument that I have seen promoted by social conservatives before, I am not so sure about that. My impression is that the upper class is experiencing family breakdown at about the same rate as the lower class is, the upper class just started at a higher level of family functioning and so remains at a higher level of family functioning in comparative terms. [Laura writes: Out-of-wedlock births have tripled for women with high education levels since 1960, but remain relatively low. Divorce rates among the highly educated have stabilized since the 1970s, according to the National Marriage Project. But they still remain high. Both divorce rates and out-of-wedlock births have grown significantly for those in the relatively uneducated middle and lower classes.] I know when looking at the specific issue of the out-of-wedlock birth ratio among college educated women that that indicator has grown among college educated women in much the same way that it has grown among women overal. I haven’t studied in detail the claims being made in the new National Marriage Project report so I can’t argue intelligently on that subject right now; I can at least say that looking at divorce rates of particular demographics today compared to the 1970s is a bit tricky in trying to draw conclusions as to what is going on by class level in America’s family life. I say this because of the anomalie in divorce rates in the United States, the extreme spike in divorce rates from 1967 to 1976 followed by the extended period of stabilization that persists to this day. I’ve seen charts indicating that divorce has gone down in younger generations compared to older generation. My theory is that a divorce fad took place from 1967 to 1976 and that the stabilization of divorce since then is based on a decline in the level of the divorce fad and an increase in what might be called “underlying divorce.” It may well be that the divorce fad was more popular among the upper classes. If this is so, and the level of divorce attributable to the divorce fad is declining, then it would make sense that in comparative terms upper class marriage is holding together better than lower class marriage when looking at data since the 1970s. However, if this is what is going on it does not indicate a greater stability of upper class marriage on a fundamental level; it could still be the case that the “underlying divorce rate” among both the upper and lower classes is growing at a continuous steady pace.

Jesse Powell writes:

The report boldly asserts “Among the affluent, marriage is stable and may even be getting stronger. Among the poor, marriage continues to be fragile and weak. But the most consequential marriage trend of our time concerns the broad center of our society, where marriage, that iconic middle-class institution, is foundering.” This is the theme of the entire report, that family life among the “highly educated,” meaning Bachelor’s Degree of higher, is doing just fine while the broad middle that is “moderately educated,” meaning they have earned a High School Degree but have not completed a Bachelor’s Degree, is facing new and profound deterioration in its family life. The category of “least educated” means those who have not graduated from High School. 

I think the fundamental mistake being made in this analysis is that education level is acting as an indicator of success and competence, it does not consist of a “class” that is homogenous and able to perpetuate itself. The “When Marriage Disappears” report gives the impression that there is a class out there called “the college educated” and that they have all these advantages that keep their families intact. Implicitly, it seems to me, there is the assumption that this “class” will pass on its advantages to its children and therefore will be able to sustain itself indefinitely as perhaps an island of privilege impervious to the wreckage being visited upon unfortunate others. 

The problem is, the college educated are not an island separated from the rest of society; the “highly educated” consist of many people who were born to less educated parents and many children who were born to highly educated parents fall out of the highly educated group by not achieving the level of education their parents achieved; there are many entrances into and exits out of the “college educated” group. Not only is the college educated population fluid and interactive with the other classes of society but the size of the college educated population itself changes; it has grown greatly during the past 20 years. It is this interactive and changeable nature of the college educated population that has led to the apparent family stability of the “highly educated” class. 

Basically, it is my view that family formation competence has been shifted or transferred from the “least educated” and “moderately educated” groups to the “highly educated” group and that this is why the “highly educated” have fared so well in terms of family stability. All of the different social classes are undergoing deterioration in their abilities to maintain healthy family life but the more competent elements of the less educated classes have moved up the education hierarchy and therefore have transferred their family forming competencies away from the less educated groups and towards the more educated groups. 

When looking at the statistics provided in the “When Marriage Disappears” report I was not able to look into the source data they were using because the National Survey of Family Growth 2006-08 information is not yet publicly available in a usable form. I was however able to find alternative sources of information that confirm the basic story of what is happening in America’s family life in regards to educational attainment. I will use these alternative sources that I found to explain in more detail what is happening in regards to family stability and education. 

The below table is comparable to Figure 6 given in the “When Marriage Disappears” report that was cited as part of the study’s summary:

Comparable Data from the Current Population Survey – America’s Families and Living Arrangements
Children under 18 years old with Married Parents according to the highest educational attainment of either parent 



% of Total


% of Total

Least Educated 61.1% 17.0% 49.2% 11.1%
Moderately Educated 74.7% 59.5% 59.4% 52.5%
Highly Educated 89.8% 23.5% 87.6% 36.3%
Total 76.0% 100.0% 68.5% 100.0%

Sources: US Census Bureau, Current Population Survey, Household and Family Characteristics: March 1993; US Census Bureau, Current Population Survey, 2010 Annual Social and Economic Supplement

The above table shows a mild decline in the proportion of children with a college educated parent living in an intact home, with their parents still married to each other. The decline among the “less educated” and “moderately educated” groups is much more severe. This is consistent with what the original “When Marriage Disappears” report says. What I think is significant though is the major reduction in the number of children living with “least educated” parents, from 17.0% to 11.1%, and the major increase in the number of children living with “highly educated” parents, from 23.5% to 36.3%. What this tells me is that the “least educated” and “highly educated” classes are not comparable to each other comparing March 1993 to June 2010. The meaning of being “least educated” is different if in 1993 you are in the bottom sixth of the distribution but in 2010 you are in the bottom ninth. Same thing holds true on the opposite side of the spectrum; the meaning of being “highly educated” is different if in 1993 you are in the top fourth of the population but in 2010 you are not even in the top third. So it is true, the least educated and the moderately educated do much worse than the highly educated in terms of the decline in marriage but the composition of the groups in question is not comparable over time so you are not comparing like to like. The numbers support the hypothesis that a large proportion of the moderately educated group advanced themselves to highly educated status thereby transferring part of the skill base away from the moderately educated class into the highly educated class, leaving the moderately educated group that is left behind worse off and the highly educated group receiving the newcomers better off in comparison.

Next I will look at the relationship status of women who have given birth in the past year, according to the educational attainment of the mother:

June 2008, women who gave birth in the past year; by educational attainment and living arrangement 





% of Total

Least Educated 37.1% 11.9% 51.0% 17.7%
Moderately Educated 55.8% 14.0% 30.1% 54.6%
Highly Educated 89.3% 3.5% 7.2% 27.6%
Total 61.8% 10.7% 27.5% 100.0%

Source: June 2008, Current Population Survey, Census Bureau, Fertility in America 2008

Looking at the stability of family life according to education level one sees a different story when looking at out-of-wedlock births, a story that contradicts the idea that the highly educated are stable in their family life while those with less education are the ones at risk. Looking at data from the American Community Survey one sees that the out-of-wedlock birth ratio has risen for all educational groups during the past 5 years, but it has risen especially quickly for the highly educated. I will mention, there is no “co-habiting” option in the American Community Survey; those who co-habit can choose either the single or married options when reporting their relationship status. 

2004 and 2009 American Community Surveys
Out-of-wedlock ratio according to education level 

  2004 2009
Least Educated 50.8% 55.6%
Moderately Educated 32.5% 42.0%
Highly Educated 6.3% 9.0%

Sources: 2004 American Community Survey 

Finally, on the issue of the “highly educated” class being able to sustain its family stability over the long term, as if it is separated from what is going on in the rest of society, I want to point out that women with less education have more children than women with higher levels of education. In June 2008 the Total Fertility Rate (TFR) of women who were “least educated”, looking at the number of children a woman has had in the 40 to 44 year old age category, was 2.452; for those who were “moderately educated” their TFR was 1.930; and for those who were “highly educated” their TFR was 1.673.  (Source: Current Population Reports – Fertility of American Women: 2008)

Also, it must be remembered, not every child born to college educated parents will become college educated themselves; even if the college educated class was stable in its family life the children of college educated parents would still face danger as they may not make it into the college educated strata themselves. Looking at income mobility between generations, in 1992 in the United States, a parent whose income was in the top quartile, roughly comparable to the college educated class, would have a son whose income would remain in the top quartile 42% of the time; 28% would drop to the third quartile, 19% would drop to the second quartile, and 12% would drop straight down to the bottom quartile. So, the moral of the story is, even if you are a parent fortunate enough to be college educated and in a stable marriage, your kids may not be so lucky.


Share:Email this to someoneShare on Facebook0Tweet about this on TwitterPin on Pinterest0Share on Google+0