Lethally Reliable Predictor

Criminologists have long believed that murder rates will climb when the number of young people grows, especially in areas where unemployment runs high and urban populations are growing.  However, a new study by Rutgers sociologist Julie A. Phillips suggests that the homicide rate may track less closely than previously thought with the size of population centers or with the number or employment status of the young people.  But one all-too-certain portent of murder remains: namely, divorce.

Examining county-by-county data collected between 1970 and 1999, Phillips uncovers a pattern that contradicts rather than confirms conventional wisdom among criminologists. In analyses that she calls “intriguing,” Phillips shows that the statistical relationships between homicide rates on the one hand and unemployment and population size on the other are both negative, so manifesting “effects that run contrary to common theoretical expectations.”  

As most criminologists would expect, Phillips does discern “a positive association between the proportion [of] young [in various areas] and homicide rates within U.S. counties across time.”  But Phillips’s multi-variable analysis establishes that “criminogenic forces, such as poor social conditions…, can alter the association between the relative size of the young population and homicide rates.”  

One particular social measure especially helps Phillips recognize areas with the kind of  “low social control” that looses murderous impulses, even if those impulses are “less heavily concentrated in the young age ranges” in the affected areas than some theorists might have expected.  The indicator of social breakdown that Phillips highlights as a predictor of murder is the divorce rate.  

Unlike elevated unemployment rates and burgeoning population size—both surprisingly linked to lower homicide rates—high divorce rates do augur bloodshed.  In four out of five of Phillips’s statistical models, the county divorce rate emerges as a statistically significant predictor of the homicide rate (p < 0.05 in all four models).  “On average,” Phillips accordingly observes, “higher levels of the percentage of the population divorced are associated with larger homicide rates within counties over time.”  

County coroners, it appears, will often be called on for grim duties wherever the divorce courts are busy.

(Source: Julie A. Phillips, “The Relationship Between Age Structure and Homicide Rates in the United States, 1970 to 1999,” Journal of Research in Crime and Delinquency 43 [2006]: 230-260.)

Lessons from Europe  

The more religious nature of the United States is often attributed to having a religious free-market where no denomination is nationally established, leaving churches to compete for members. So does this mean that if European countries follow the U.S. and disestablish their state churches that Europe might become more religious? Not according to two Dutch scholars, whose study quantifies other factors—including increased participation of women in the labor force and increased religious pluralism—as more responsible for religious decline across the Atlantic.

A sociologist and a theologian at Tilberg University in the Netherlands examined data from the European Values Study, a series of surveys conducted in almost all European countries between 1999 and 2000, to explore how characteristics of individuals and countries influence individual religious beliefs and practices. They found, among factors at the individual level, that women in paid employment were significantly less religious (both in terms of belief and practice) than their peers who stayed at home. In fact, the level of religious belief among employed women was more like those of men, who were found to be less religious than women overall. These consistent patterns were found in almost all countries and were statistically significant (p<.001) in multivariate tests.

Looking at the characteristics of countries, the study found that religious pluralism as measured by the Herfindahl Index—meaning the more religions in a country and the more evenly distributed their market shares—as well as the degree to which people trust the churches in their country were each significantly related to religious belief and to religious practice: The greater the religious diversity of a country, the lower the levels of individual belief and practice; whereas higher levels of public confidence in the church increased each of the two measures.

While differences in the extent of religious belief and practice were found among Protestants, Catholics, and Orthodox, all of these Christians were found to be consistently less religious in both measures than were adherents of “other” religions.

The researchers concede that identifying a pattern to their findings may not be easy. But their analysis regarding the effects of female employment and religious diversity in Europe offers lessons that Americans would be wise to ponder.

(Source: Loek Halman and Veerle Draulans, “How Secular is Europe?” The British Journal of Sociology 57 [June 2006]: 263-289.)

Solving the Fertility Paradox   

Fertility rates have risen in recent decades for unmarried American women.  Fertility rates have likewise risen in recent decades for married American women.  So why has the overall American fertility rate changed very little?  This question recently attracted the attention of a team of demographers at the University of Oregon, and their analysis identifies the national retreat from wedlock as the key to the mystery. 

The researchers begin their work by defining “the apparent paradox” in American fertility data.  They note that between 1974 and 2000 the birth rate among unmarried white women ages 20 to 39 “more than tripled, rising from 13 per 1,000 in 1974 to 46 per 1,000 in 2000” and that during the same period the birth rate among married white women of the same ages “rose by nearly half, from 94 [per 1,000] to 135 [per 1,000].”  The analysts report that “patterns are similar for black women,” unmarried and married.  Nonetheless, the data indicate that 1974 to 2000 was “a period of relatively flat total birth rates.”

What is going on here?  When the Oregon scholars parse the data, they highlight “the effects that marriage behavior has on the composition of married and unmarried women.”  That is, in the researchers’ statistical model  “a decline in marriage will cause increases in the nonmarital birth rate, the marital birth rate, and the nonmarital birth rate relative to the marital birth rate, even though the total birth rate does not change.”  The Oregon scholars characterize recent American fertility data as “remarkably consistent with the various predictions of the [statistical] model.” 

Apparently, when fewer women marry, those who still do wed tend to be distinctively committed to family and childbearing.  Meanwhile, as a growing number of women wed late in life or not at all, an increasing number of them—predictably enough—bear children out of wedlock.  

Of course, even if a society’s overall fertility rate remains unchanged, serious social problems are brewing when wedding bells rarely ring and when more and more children have unmarried parents.

(Source: Jo Anna Gray, Jean Stockard, and Joe Stone, “The Rising Share of Nonmarital Births: Fertility Choice or Marriage Behavior,” Demography 43 [2006]: 241-253.)

Not-So-Gay Gays  

Though very popular among progressive social commentators, the notion that homosexuality is perfectly normal and healthy does not look very credible to researchers who have examined it closely.  The latest evidence linking homosexuality to negative life characteristics appears in a study recently published in the Journal of Youth and Adolescence by a team of researchers from Brock University in Canada. 

Examining data collected from over 3,700 Ontario high school students, the Brock researchers find that adolescents who report “exclusive heterosexual attraction” experience “the most positive status across domains” under investigation: psychological functioning, academic orientation, parental relationships, friendship quality, victimization, school culture, neighborhood quality, and attitudes toward risk.

Though bisexual adolescents report “the most negative results” documented in this study, homosexual adolescents likewise look much worse off than their heterosexual peers.  Compared to exclusively heterosexual peers, students reporting same-sex attraction manifest “significantly worse results” in four areas: psychological functioning, parental relationships, victimization, and neighborhood quality. 

Not all academics will welcome the finding that homosexuality comes with “heightened risk and difficulties” for affected youth.  But the authors of the new study harmonize their research with the prevailing canons of political correctness by calling for “programs aimed at assisting these [homosexual] youth.”  But readers of their study may conclude that what is most needed is the elimination of educational programs that legitimate and even glamorize homosexuality as they invite youth to jeopardize their psychological health by experimenting with unnatural forms of sexuality.

(Source: Michael A. Busseri et al., “Same-Sex Attraction and Successful Adolescent Development,” Journal of Youth and Adolescence 35 [2006]: 563-575.)

Shadow of a Departure, Challenge of a New Arrival  

The arrival of a first child always brings challenges for a couple.  But new research suggests that couples in which both father and mother are in their first marriage cope with this transition more successfully than do couples in which one or both of the partners have been in a previous marriage or cohabiting union.  To understand the distinctively intense difficulties that a new baby brings to parents in a second or third union, readers may turn to a study recently published in the Journal of Divorce & Remarriage by psychologist Geneviève Bouchard of the University of Moncton. 

By scrutinizing data collected from 140 French-Canadian couples all experiencing the arrival of a first child, Bouchard establishes a clear link between parents’ previous relationship history and their ability to adjust to the new arrival.  Data for the mothers in the study indicate that “those who have experienced a divorce or relationship dissolution are globally less conjugally adjusted at both points of assessment than those in first unions.”

Further analysis reveals, however, that “men’s divorce or separation history matters the most.”  Bouchard calculates that “the decline in relationship functioning associated with the transition to motherhood … is about three times larger for women who are married or cohabiting with a man with [the] ‘baggage’ [of a failed previous marriage or cohabiting relationship].”  Further analysis reveals that compared to peers in a first union, fathers who had failed in a previous relationship were more likely to be simply cohabiting with their current partner rather than being married (p < 0.001).”

Bouchard characterizes “relational cost associated with second unions” as “modest” and even as relatively “small,” yet she concedes that the effect in view is “deleterious” and “significant.”  What is more, careful statistical analysis proves that the negative relationship effect linked to a failed previous marriage or cohabitational union cannot be accounted for by the presence or absence of children from the previous union, nor can it be explained away by attending to demographic variables, such as age or length of relationship.

The national epidemic of divorces in recent decades can only mean that for millions of couples, the arrival of a new one has brought extra difficulty and strain.

(Source: Geneviève Bouchard, “Transition to Motherhood and Relationship Quality: Does Divorce or Separation History Matter?” Journal of Divorce & Remarriage 44.1/2 [2005]: 107-116.) 

Violent Homes, Violent Neighborhoods  

Progressives never tire of decrying the evil of domestic violence, particularly that directed against women.   Curiously, however, they rarely say anything about the cultural erosion of the social institution that best shields women from such violence: namely, marriage.  Still, the evidence continues to accumulate showing that marriage matters a good deal in reducing women’s vulnerability to domestic violence.  Indeed, a study recently published in Public Health Reports indicates that a woman seeking safety will want to live in an intact marriage herself—and in a neighborhood filled with intact marriages.

Conducted by researchers at the University of Tennessee and the University of Cincinnati, the new study examines the effects of “contextual risk” on the prevalence and severity of Intimate Partner Violence (IPV).  The Tennessee and Cincinnati scholars calculated the “contextual risk” for IPV for a nationally representative sample of 2,273 couples with children ages 5 to 17, using data collected from these couples in 1990 and 1994 by interviewers and Census officials.  Those calculations highlight the importance of marital status as a predictor of Intimate Partner Violence. 

“As might be expected in a sample of households with school-aged children,” the researchers report, “stably married couples . . . have the lowest rates of I[ntimate]P[artner]V[iolence].” For stably married couples, the researchers calculate an incidence of 16.2% for overall IPV and of 3.5% for IPV involving “physical violence with injury.”   In contrast, the researchers find that “cohabiting couples show the highest rates of  IPV.”  Among cohabiting couples the rate of overall IPV runs more than twice as high as that found among stably married couples (37.5% among “stable cohabiting couples”; 33.6% among “new” cohabiting couples).  The rate of physical violence with injury runs four times as high as that found among stably married couples (16.1% among stable cohabiting couples; 14.1% among new cohabiting couples).

Though the incidence of overall and severe IPV does run higher among newly married or remarried couples than among stably married couples, it still runs far below that observed among cohabiting couples. (The researchers report a rate of overall IPV of 18.7% among newly married or remarried couples and a rate of IPV with physical violence with injury of 7.0%.)

Nor is it just a woman’s own marital status that determines her vulnerability to domestic violence.  The authors of the new study establish that “neighborhood context” also helps determine that vulnerability.  And in determining whether a neighborhood is “advantaged” or “disadvantaged” the researchers look at—among other social and economic characteristics—the fraction of households in the neighborhood that are headed by single parents.  When that fraction rises, the neighborhood becomes more disadvantaged.

The researchers note that, compared to violence-free couples, “couples with IPV are more likely . . . to live in neighborhoods of high disadvantage.”  Among couples who reported Intimate Partner Violence, 27.3% lived in disadvantaged neighborhoods; among couples who reported no IPV, only 18.3%.  Among couples who reported severe domestic violence involving injury, more than a third (35.2%) lived in disadvantaged neighborhoods, compared to less than a fifth (19.1%) of those who reported no severe domestic violence.  Statistical tests identify all of these neighborhood-context effects as significant (p < 0.001 for all neighborhood effects).

Those truly intent on reducing the incidence of domestic abuse are those at work to reverse the national retreat from marriage.

(Source: Greer Litton Fox and Michael L. Benson, “Household and Neighborhood Contexts of Intimate Partner Violence,” Public Health Reports 121 [2006]: 419-427.)

Hitting the Bottle—Hard!  

With good reason, health and educational officials are worried about the effects of heavy binge drinking among the nation’s high schoolers.  According to the authors of a new study on the problem, “Heavy episodic drinking among young people is increasingly perceived as an important social problem.”  The magnitude of the problem is suggested by a 2000 study by the U.S. Department of Health and Human Services in which researchers documented “a heavy episodic drinking prevalence…of 26.2% among 10th graders in a two-week time frame.”  Accordingly, the authors of the new study—all public-health scholars from the University of Rhode Island, Rhode Island Hospital, or Brown University School of Medicine—set out to identify the social and personal characteristics that foster or prevent this problem.

Parsing data collected from 938 Arkansas students enrolled in the 9th, 10th, and 11th grades, the researchers discern some clear patterns.  Students who report heavy episodic drinking are disproportionately “older, male, from families that receive welfare benefits, are not living in intact families, are attending religious services less often, and are more likely to have delinquent friends than those who have not drunk heavily in the month prior to the survey.”

In simple bivariate statistical tests, students living in an intact family are almost half as likely to binge drink as peers not living in an intact family (-46%; p < 0.01).  The protective effect of living in an intact family declines in a more sophisticated multivariate statistical test that takes into account age, sex, welfare receipt, and other background variables, but it is still sizable and significant (-32%; p < 0.05).

Though the effect is smaller than that of family structure, strong attachment to parents also predicts less vulnerability to heavy episodic drinking.  Compared to peers without such attachment, students who reported a strong attachment to their parents were between one-fourth to one-fifth less likely to binge drink (-26% in simple bivariate analysis, p < 0.01; -22% in sophisticated multivariate analysis, p < 0.05).

No doubt, health educators will do their best to combat the problem of binge drinking among adolescents.  However, until Americans end the national retreat from marriage and the national epidemic of divorce, tens of thousands of distressed and rootless young people will continue to take dubious comfort from a bottle.

(Source: Barbara J. Costello, Bradley J. Anderson, and Michael D. Stein, “Heavy Episodic Drinking Among Adolescents: A Test of Hypotheses Derived from Control Theory,” Journal of Alcohol and Drug Education 50.1 [2006]: 35-56.)