Wedlock or Lockup?

Johnny Cash was no criminologist.  But lyrics made famous by the country singer appear in a study of the effects of marriage recently published in Criminology by scholars at Harvard and the University of Maryland.  For when Johnny Cash sang, “Because you’re mine, I walk the line,” he was anticipating by five decades the results of a sophisticated study into how marriage prevents crime.

Deploying sophisticated statistical techniques, the authors of the new study scrutinize data collected from 500 high-risk young men from adolescence to age 32.  These new techniques allow the Harvard and Maryland scholars to calculate that “being married is associated with an average reduction of approximately 35 percent in the odds of crime compared to nonmarried states for the same man.” 

The researchers characterize their findings as “robust” and see in them evidence “supporting the inference that states of marriage causally inhibit crime over the life course.”

In explaining why “marriage [is] important in the process of desistence from crime,” the authors of the new study suggest that “marriage has the potential to ‘knife off’ the past from the present in the lives of disadvantaged men,” thus allowing these men to experience “identity transformation” as they move toward “structured routines that center more on family life and less on unstructured time with peers.”  In short, the researchers see marriage transforming a difficult man “from a hell raiser to a family man.”

As a reformed hell raiser himself, Johnny Cash would wholeheartedly endorse the findings of this new study.

(Source: Robert J. Sampson, John H. Laub, and Christopher Wimer, “Does Marriage Reduce Crime?  A Counterfactual Approach to Within-Individual Effects,” Criminology 44 [2006]: 465-502.)

School Readiness: A Phony Issue?  

Although Congress claimed in 1965 that Head Start would reduce juvenile delinquency, poverty, and dependency, the preschool initiative that today costs $7 billion a year has yet to demonstrate any lasting boost to the educational outcomes of at-risk children. Nevertheless, welfare-state advocates are now pushing for public preschool for all children, claiming it will dramatically improve educational achievement. Yet a study by the Goldwater Institute of Arizona warns that the public should not fall for the curve ball this time, finding not only that pre-K programs offer questionable educational value, but also that no crisis even exists to justify such action.

Among the extensive findings pulled together by institute president and early education authority Darcy Olsen, the most riveting is her observation that the huge expansion of early education since 1965 has not yielded rising outcomes of elementary school students. In 1965, only five percent of three-year-olds and 14 percent of four-year-olds were enrolled in pre-K programs. Today, those figures are 39 percent and 66 percent, respectively. Yet statistics from the National Assessment of Educational Progress show how fourth-grade reading, science, and math scores have stagnated since the early 1970s and in some cases fallen, even as the nation has tripled spending in education, increased teachers’ salaries, and reduced class sizes. Nevertheless, even in these subject areas, American fourth-graders still outperform their peers in France, Italy, and Germany, countries that have the kind of universal pre-K system that some want here.

In her review of the empirical research, Olsen finds that formal early education at best yields only short-term effects with at-risk students, effects of which “fade out” by grade three, and at worst yields adverse effects with mainstream children. Even where a program might be beneficial, like the often cited Carolina Abecedarian Project, its applicability to the preschool question is limited, as this costly intervention enrolled at-risk children at the age of six months in an all-day, five-days-a-week, and twelve-months-a-year program for four and half years.

Nor are conventional preschool programs any more promising. Olsen notes that after ten years and spending $1.15 billion making preschool free for all four-year-olds in the Peach State, scores on the Georgia Kindergarten Assessment Program remain essentially unchanged since 1993 (when the experiment began) and that differences between students (who were in preschool and those who were not) are not statistically significant. She also wonders what crisis initiatives like Georgia’s are intended to address, citing studies from the Early Childhood Longitudinal Study, K Cohort, that reveal—on the basis of the “readiness” standards of Goals 2000—that close to 90 percent of children already start kindergarten well-prepared and with a strong foundation.

These findings lead Olsen to lament that the current debate has little to do with the cost or effectiveness of preschooling: “At heart is the question of in whose hands the responsibility for young children should rest. On that question, plans to entrench the state further into early education cannot be squared with a free society that cherishes the primacy of the family over the state.”

(Source: Darcy Olsen and Jennifer Martin, “Assessing Proposals for Preschool and Kindergarten: Essential Information for Parents, Taxpayers, and Policymakers,” Policy Report No. 201, February 8, 2005, Goldwater Institute, Phoenix, Arizona.)

Blue Laws Not So Blue  

State and local ordinances restricting commerce on Sunday were fairly widespread in America from colonial times until 1961. That’s when the Supreme Court opened the door to constitutional challenges to “blue laws,” after which most were repealed on merchants’ claims that they imposed hardships on the public. But now comes a study by two noted economists that suggests rather than spreading a case of the blues, the secular fencing off of one day a week for worship, rest, and family yielded important social dividends.

Using datasets from the General Social Survey (GSS), Jonathan Gruber of MIT and Daniel M. Hungerman of Notre Dame found that the repeal of Sunday-closing laws in 16 representative states between 1955 and 1991 triggered “a very strong reduction” in the frequency of church attendance in those same states between 1973 and 1998. Then looking at data from the Consumer Expenditure Survey and state-by-state spending by congregations of four representative Protestant denominations between 1950 and 2000, they found that repeal triggered significant declines in giving to religious organizations, as well as in church budgets. Finally, looking at the National Longitudinal Survey of Youth (NLSY), the economists found that repeal yielded increases in alcohol and illegal drug use among a cohort of young people, ages 14 to 21.

Quantifying the strong and “striking” correlation between blue laws and church attendance, their analysis found that the change in state laws reduced attendance by about five percent of the median, what they term a sizeable effect roughly one-third as large as the well-established higher rate of attendance among women relative to men. They also observed a downshift in attendance-frequency categories, including a 15 percent decline in GSS respondents claiming to attend church weekly and increases in those claiming rare attendance. In addition, the sizeable decline in giving amounted to a 13 percent reduction in giving by individuals, as well as a statistically significant 6.3 percent drop in per-member spending by congregations associated with the United Methodist Church, the Southern Baptist Convention, the Missouri Synod Lutheran Church, and the United Church of Christ.

Their analysis of the NLSY data, which tracks church attendance since 1979 and substance use between 1982 and 1994, found large and significant negative effects of repeal among church-attending youth that extend beyond a cut back in attendance. They found that church-going teens (relative to their peers that do not attend church) were 5.5 percent more likely to drink, which lowered the gap in heavy drinking between the religious and nonreligious youth by 50 percent. In addition, they were 11 percent more likely to use marijuana and 3.6 percent more likely to use cocaine, effects which closed the gaps completely in these behaviors between churched and unchurched youth.

Specification tests, which included controls for a state’s socioeconomic characteristics and changes in other types of social participation, confirmed that these findings were not driven by declines in religiosity that may have been occurring before repeal or by declines in membership and giving to nonreligious organizations after repeal. So while opening the mall on Sunday might deliver short-term gains to the corporate bottom line, these findings offer hard evidence that the move to turn the Lord’s Day into just another hectic day of buying and selling was not possible without a critical loss of social capital.

(Source: Jonathan Gruber and Daniel M. Hungerman, “The Church vs. the Mall: What Happens When Religion Faces Increased Secular Competition?” National Bureau of Economic Research, Working Paper 12410, July 2006.)

The End of an Era?  

Has the movement of women and especially married mothers into the labor force run its course? Not only did the trend slow considerably during the 1990s, but the latest data reveal a modest decline in the labor force participation rates of women during the first half of the present decade. While reluctant to conclude that the five-year drop represents a reversal of the dramatic increases of the 1970s and 1980s, economists at the University of Houston and the Federal Reserve Bank in New York nonetheless consider it “the most striking change” of the 2000s and do not anticipate seeing married women increasing their labor force participation above the record levels set during the late 1990s.

Using micro data of the March version of the Current Population Survey of the U.S. Census Bureau, Chinhui Juhn and Simon Potter point out that married women—and especially married mothers in their childbearing years—were the driving force behind the rise in the overall female labor force participation rate. Among married mothers, the participation rate jumped from 38.9 percent in 1969 to 65.8 percent in 1989. Although their rate rose to 70.3 percent in 1999, it dropped to 68.4 percent in 2004.

In contrast, the rate among never-married mothers jumped dramatically during the 1990s, from 63 to 75.4 percent (due to the impact of the 1996 welfare reform legislation and the expanded Earned Income Tax Credit). But the overall rise in their rate of labor force participation was far less dramatic, from 62.9 percent in 1969 to 71.7 percent in 2004. Looking at these marital status differences, former Howard Center fellow Brian Robertson observed the anomaly that married mothers, at least up until the 1990s, seem more driven to work outside the home than did unmarried mothers.

Seeking to identify “supply-side” explanations for what happened during the 1970s and 1980s, the researchers put a lot of stock in divorce rates, noting how divorce increases the proportion of unmarried women—who are more likely to work outside the home—as well as how the increased risk of divorce increases married women’s use of employment as a hedge against marital instability. They discount the relative decline in men’s earnings as an explanation, as the increases in the percentage of working mothers has been greatest among wives of top-earning males. They unfortunately fail to consider the decline of the family wage as the statistical norm for men’s wages across the income spectrum, which Robertson credits with placing new economic pressures on married-couple families.

Other than citing the “unsustainably high levels” of labor force participation during the late 1990s and existing “weak market labor conditions,” the economists have little to say about what fueled the slowdown of female labor force participation in the 1990s and its decline in the 2000s. But if that decline continues, that might provide fodder for greater insight and signal a return to sanity for the American family.

(Source: Chinhui Juhn and Simon Potter, “Changes in Labor Force Participation in the United States,” Journal of Economic Perspectives 20 [Summer 2006]: 27-46.)

Status at Birth Sets the Pace  

As the transition to formal schooling sets a trajectory that influences a child’s long-term prospects, scholars continue to look for risk factors for children making that transition. Recently Shannon Cavanagh and Aletha Huston at the University of Texas found that family instability—or changes in parents’ relationships with their “partners” between birth and the end of kindergarten—increases children’s behavioral problems in the first grade. Perhaps more important, they also discovered that parental marital status at birth plays a critical role in reducing or exacerbating that instability.

Looking at longitudinal data on a sample of more than 1,000 children drawn from Study of Early Child Care and Youth Development of the National Institute of Child Health and Human Development (NICND), the research team discovered that only 25 percent of the sample suffered from a change in their parents’ intimate relationships. However, less than 20 percent of children born to their natural and married parents experienced any family instability, compared to more than half of those born to single parents and nearly two thirds of those born to cohabiting parents.

Likewise their multivariate tests found direct links between family instability and family structure at birth to three measures of problem behaviors. Family instability was modestly related to teacher reports of externalizing behavior and to observed disruptive behavior with peers (each correlation, p<.10) and significantly related to observed disruptive behavior with teachers (p<.001). Family structure was related to the last two measures, as children born to single parents engaged in more disruptive behavior with peers (p<.001) and with teachers (p<.01), as were children born to cohabiting parents (p<.10 for behavior with peers, p<.05 for behavior with teachers). These patterns held even when controlled for household environment factors, including maternal depression, maternal sensitivity, Home Observation for Measurement of the Environment (HOME) scores, and family income.

Furthermore, the children born to married parents seemed less affected by subsequent family instability. While the extent of family instability did influence their behavior as measured by teacher reports of externalizing behavior, the number of parental transitions did not affect the two classroom measures of disruptive behavior as it did with their peers born to single and cohabiting parents.

The researchers hope their findings help the public “better understand the ways families change.” Yet they ironically shy away from highlighting what their study clearly shows: that the changing structure of the American family itself poses greater risks to children than any factors related to it.

(Source: Shannon E. Cavanagh and Aletha C. Huston, “Family Instability and Children’s Early Problem Behavior,” Social Forces 85 [September 2006]: 551-580.)

No Support  

Feminists have been assuring Americans for decades that children do fine when Mom takes a job outside the home.  Such assurances, however, do not look very plausible in the light of a study on maternal employment published in The Journal of Marriage and Family by sociologists from the University of Maryland and Northern Illinois University.

To assess the effects of maternal employment on family life, the Maryland and Northern Illinois scholars set to work parsing data collected in 1995-1996 from a probability sample of 2,246 English speaking Americans, ages 25 to 74, who had all lived with both biological parents from birth to age 16.  The data provide much to sober and little to reassure advocates of maternal employment. 

The data analyzed in this new study show that “adults whose mothers worked for pay during all or most of their childhood reported a lower level of maternal support during childhood compared with those whose mothers stayed home.”   Nor did the husbands of employed mothers compensate for the reduced support these women gave their children.  Far otherwise.  The researchers conclude that adults who grew up with a mother employed outside the home “reported a lower level of support from their fathers, compared with those whose mothers stayed at home.” 

The deficit in parental attention evident in employed-mother households manifests itself not only in a lack of support for children, but also in a lack of discipline of children.  The authors of the new study find that adults who grew up with a mother employed outside the home experienced “a lower level of discipline by their mothers than those who had homemaker mothers.”  And again, just as they saw fathers imitating rather than compensating for employed mothers too busy to give their children support, even so researchers see no evidence that fathers are compensating for this deficit in discipline on the part of employed mothers.  Rather, the data indicate that in employed-mother households, “children reported a lower level of discipline by their fathers, compared with those whose mother stayed at home.”

Tragically, many children—or at least many sons—who are not receiving the formative guidance of discipline in employed-mother homes are suffering through the psychologically damaging trauma of abuse.  Statistical analysis reveals that the sons of employed mothers experienced “higher levels of fathers’ verbal or physical assaults” than did peers with homemaker mothers.

The authors of the new study interpret their findings as corroboration of a “role-strain perspective” on maternal employment.  This perspective predicts both that “employment may prevent mothers from engaging in sound parenting practices (e.g., warm support, consistent discipline, and little or no harshness)” and that “the absence of a homemaker may also lead fathers to be less effective in parenting.”

The empirical support for a role-strain perspective comes as no great surprise for the researchers.  What does come as a surprise for the researchers is the remarkable consistency of the maternal-employment effects across generations.  The authors of the study acknowledge that they began their research expecting the data collected from the Baby Bust generation (born between 1960 and 1970) to look quite different from the data collected from the older generations, who grew up during a time when maternal employment was less common and regarded less favorably.  “Unexpectedly,” the researchers admit, “we did not find any cohort differences.”

Whether born in the Thirties, the Forties, the Fifties, or the Sixties, children have experienced maternal employment as deprivation and trial.  That deprivation and trial continue today for the children of employed mothers.

(Source: Kei M. Nomaguchi and Melissa A. Milkie, “Maternal Employment in Childhood and Adults’ Retrospective Reports of Parenting Practices,” Journal of Marriage and Family 68 [2006]: 573-594.)