31 October 2016

Weird Stat Reports On Blogger

The statistics feature on blogger for this blog has been showing very weird traffic patterns for the last week or so. Usually, most traffic comes from Google and a few sites that link to this one. But, in the last week, almost all of the traffic (more than 7,000 page views, a large share of which to pages I haven't visited for years) has allegedly been linked to a variety of pages at this site, with particularly heavy views of a number of political posts from the period from 2005 to 2012.

The feature at this blog and its sister blog that notes the latest post at the other blog has also been acting up and not functioning at time.

I have no idea what is going on and it doesn't seem to have implicated functionality, but I put it out there for those who may be more knowledgable about these things.

30 October 2016

Winning The Battle But Losing The War?

The backers of a pipeline across the Dakotas have truly screwed the pooch with the intense law enforcement response to protests in the face of some doubtful efforts by supporters to push the project through over opposition.

At this point, even if they are successful, they are going to have a pipeline that can't move and can't possibly be guarded over all of its thousands of miles expanse, that large numbers of people will feel justified in sabotaging for decades to come. As a practical matter, the project will become an uninsurable albatross that will be out of service due to intentional efforts to destroy it more often than it is up and running.

I'm not endorsing that course of action, but it increasingly seems inevitable and makes the efforts of backers of the pipeline appear to be nothing but folly.

29 October 2016

Global Music

One of the joyous things about living in the 21st century is the access that I have, between YouTube, Sirius XM and Spotify, to popular music in languages other than English. Just in the last couple of weeks I've listened to pop music in French, Japanese, Latin, Hebrew, Arabic, Spanish, Finnish, Swedish, German and Chinese (from Hong Kong).

It was one of the things I dreamed might be possible someday when I was high school that managed to actually happen by the time that my own kids are in high school (other things I dreamed of that have also come to be that actually did include Smart Cars, continuously variable transmissions, and electric cars).

American Men Pitch In More At Home Than Asian Counterparts

This will come as no surprise to anyone familiar with the situation.
On average, married men in the United States spend 167 minutes per day in home production, whereas Japanese husbands spend only 40 minutes and Korean husbands spend 48 minutes.
The difference holds even after adjusting for the labor market status of the man and the woman. 
That is from Daiji Kawaguchi and Soohyung Lee, “Brides for Sale: Cross-Border Marriages and Female Immigration,” Economic Inquiry.  Here are various copies and drafts of the paper.
Via Marginal Revolution.

28 October 2016

Pro Se Parties In Colorado's State Courts

Statistics are available regarding the percentage of litigants in Colorado state courts who were not represented by attorneys (i.e. pro se) in the 2015 fiscal years:

* In domestic relations cases in Colorado, 75% of filing parties and 75% of responding parties are not represented by attorneys. About 67% of domestic relations cases have no party that is represented by an attorney.

* In civil cases in the general jurisdiction district court (excluding tax lien filings and foreclosure cases), 16% of filing parties and 52% of responding parties are not represented by attorneys. About 12% of these cases have no party who is represented by an attorney.

* In civil cases in the limited jurisdiction county court (jurisdictional limit $15,000), 17% of filing parties and 98% of responding parties are not represented by attorneys. About 14% of these cases have no party who is represented by an attorney.

The court process in domestic relations cases and general jurisdiction civil case procedures are simply not equipped to handle pro se parties, so this is a real problem.

Jury Trials Still Scarce

At the Founding, juries decided all but the most minor criminal cases. But by 1962, jury trials accounted for only 8.2% of cases tried in federal court. And by 2013, that number had more than halved, dropping to 3.6%. The numbers in state courts are even more bleak—and, in most cases, come on top of the absence of grand juries. And in civil cases, as will surprise absolutely no one, juries decided only 5.5% of federal cases in 1962—and 0.8% by 2013. 
There are lots of explanations, obvious and otherwise, for these trends. But whereas conventional narratives of the jury’s demise have emphasized the inefficiency, cost, incompetence, and inaccuracy of the jury, the real culprits, Thomas argues, are each of the branches of government, which have “seized the domain of the jury.” As Thomas explains, “the executive charges, convicts, and sentences, despite juries indicting, sentencing, and convicting in the past. The legislature can set damages, although only the jury historically had that power. The judiciary circumvents juries by resolving cases via mechanisms such as the motion to dismiss, summary judgment, judgment of acquittal, and judgment as a matter of law, procedures nonexistent at our Constitution’s founding.” And all of this is on top of what Thomas consciously excludes from her discussion, the move (sanctioned, if not affirmatively encouraged, by all three branches) toward non-trial settlement—whether through plea bargains in the criminal context or alternative dispute resolution in the civil context.

The facts

Criminal Juries

Grand juries, particularly in federal cases, as I've recounted in a previous post, do virtually nothing to protect individual rights and are an anachronism.

Contrary to the quoted material above, both trials in general, and jury trials in particular, are much more common in state criminal cases than in federal criminal cases. This is largely as a result of tough mandatory minimum sentences and sentencing guidelines that federal prosecutors can utilize and because federal prosecutors have more control over which cases to prosecute and which to leave to state prosecutors (taking only the stronger cases, most of the time). The fact that the penalty for seeking a jury trial is so high is what leads to very high plea bargaining rates in federal cases. Criminal jury trials are all but dead as a result in federal court, but are alive, if not well, in state court.

For example, in 2006 in Colorado, in state courts there were 1,776 criminal or quasi-criminal jury trials (plus about 35 quasi-criminal parental rights termination jury trials). In state court, two-thirds of traffic cases that go to trial were tried by juries, 69% of misdemeanor trials were before juries, and 97% of felony trials were before juries. In federal court, there were 22 criminal jury trials, just 59% of criminal trials were to juries, and the percentage of cases going to trial is lower than in state court (at least for felonies).

About 2% of state felony cases and 1.25% of misdemeanor cases in Colorado in 2006 went to trial, with 2% of state felony cases and about 0.8% of state misdemeanor cases resolved with jury trials. About 0.34% of state traffic cases go to trial, with about 0.22% resolved by jury trials.  There are roughly the same number of traffic cases and limited jurisdiction civil cases filed each year in state courts in Colorado, but there were 386 jury trials in traffic cases and just 17 in limited jurisdiction civil cases in 2006.

Nationwide, only about 6% of state criminal cases and 3% of federal criminal cases are not plea bargained, and some of the rest of dismissed prior to trial.

Less than 2% of criminal jury trials take place in federal court, but about 9% of people in prison at any given time in the United States were convicted in federal court for violating federal criminal laws, and less than 1% of death sentences are federal. The proportion of criminal cases brought in federal court is lower than 9%, however, because misdemeanor and petty offense prosecutions are disproportionately brought in state courts and most criminal prosecutions involve misdemeanors and petty offenses.

Civil Juries

The situation in civil cases, which 15% of all jury trials are conducted, is more comparable between the federal and state system. There were just 17 limited jurisdiction civil trials in state court (v. 1236 bench trials and 3485 small claims bench trials) with jury trials making up just 1% of state court trials and where trials themselves are very rare.  About 54% of civil trials in federal court (43 jury trials and 36 bench trials) and 50% civil trials in general jurisdiction state courts were to juries (277 jury trials and 280 bench trials), with about 75% in tort cases) - both of these categories of cases are more likely to go to trial but trials are still rare in civil cases across the board.

About 1% of general jurisdiction court civil cases went to trial in Colorado's state courts in 2006, so less than 0.5% of them were resolved in jury trials. About 0.7% of limited jurisdiction court civil cases in Colorado's state courts went to trial and only 0.01% of them were resolved in jury trials.

Federal jury trials are more common as a share of federal civil actions than in state court, as roughly 97% of all civil cases are filed in state courts and 89% of the civil cases filed in federal court were bankruptcies, a typical year, so just 0.3% of the non-bankruptcy civil cases in the state are filed in federal court, yet 13% of civil jury trials are conducted in federal court. But, this is largely a function of debt collection cases and routine minor tort cases which are unlikely to go to trial, being brought predominantly in state courts. Jury trial rates in cases of comparable significance in federal court settle at fairly comparable rates to those in state courts, which makes sense because the procedural framework for civil cases is almost identical in state and federal courts.

On the civil side, the issues involved with jury trials are quite different. Much of what makes civil procedure inefficient and hard to reform can ultimately be traced back to the fact that key decisions have to be deferred to a final trial before a jury that is unlikely to ever actually happen.

* * * *

Still, in the face of a blatant case of jury nullification in the prosecution of an armed militia that had engaged in treason (although the charges brought were less serious), it is hard to be too attached to the jury trial as an instrument of justice.
PORTLAND, Ore. — The leaders of an armed group who seized a national wildlife refuge in rural Oregon were acquitted Thursday in the 41-day standoff that brought new attention to a long-running dispute over control of federal lands in the U.S. West. A jury found brothers Ammon and Ryan Bundy not guilty a firearm in a federal facility and conspiring to impede federal workers from their jobs at the Malheur National Wildlife Refuge, 300 miles southeast of Portland where the trial took place. Five co-defendants also were tried one or both of the charges. . . . 
The armed occupiers were allowed to come and go for several weeks as authorities tried to avoid bloodshed seen in past standoffs. . . . The group began occupying the bird sanctuary in remote southeastern Oregon on Jan. 2. They objected to prison sentences handed down to Dwight and Steven Hammond, two local ranchers convicted of setting fires. They demanded the government free the father and son and relinquish control of public lands to local officials. Ammon Bundy gave frequent news conferences and the group used social media in a mostly unsuccessful effort to get others to join them. The Bundys and other key figures were arrested in a Jan. 26 traffic stop outside the refuge that ended with police fatally shooting Robert “LaVoy” Finicum, an occupation spokesman. Most occupiers left after his death, but four holdouts remained until Feb. 11, when they surrendered after a lengthy negotiation. 
At trial, the case was seemingly open-and-shut. There was no dispute the group seized the refuge, established armed patrols and vetted those who visited. “Ladies and gentlemen, this case is not a whodunit,” Assistant U.S. Attorney Ethan Knight said in his closing argument, arguing that the group decided to take over a federal workplace that didn’t belong to them. 
On technical grounds, the defendants said they never discussed stopping individual workers from accessing their offices but merely wanted the land and the buildings. On emotional grounds, Ammon Bundy and other defendants argued that the takeover was an act of civil disobedience against an out-of-control federal government that has crippled the rural West. Federal prosecutors took two weeks to present their case, finishing with a display of more than 30 guns seized after the standoff. An FBI agent testified that 16,636 live rounds and nearly 1,700 spent casings were found. 
Bundy testified in his defense, spending three days amplifying his belief that government overreach is destroying Western communities that rely on the land. He said the plan was to take ownership of the refuge by occupying it for a period of time and then turn it over to local officials to use as they saw fit. Bundy also testified that the occupiers carried guns because they would have been arrested immediately otherwise and to protect themselves against possible government attack. Ryan Bundy, who acted his own attorney, did not testify.
Authorities had charged 26 occupiers with conspiracy. Eleven pleaded guilty, and another had the charge dropped. Seven defendants chose not to be tried at this time. Their trial is scheduled to begin Feb. 14.
Despite the fact that the trial was held in Portland, the jury pool drew from all over the state and not just this coastal urban center.

27 October 2016

Denver Is A Popular Destination

About 30 percent of Denver's renters are coming to the city from out of town, which is helping drive rent prices upward. . . . Only Nashville, Tennessee (42.2 percent), Raleigh, North Carolina (38.1 percent), San Jose, California (34 percent), and San Francisco (30.9 percent) have higher percentages of renters coming from other cities to rent.
Zillow estimates the Denver rents have risen 3.3 percent in the past year and estimates the median monthly rent in Denver is $2,008.
From the Denver Business Journal.

Some of Denver's attractiveness is driven by the $2.4 billion that legalizing marijuana has added to the Colorado economy.

23 October 2016

Thirteen General Features Of Revolutions

1. Revolutions almost always involve a regime whose legitimacy has been undermined in some way.

There are lots of reasons to be unhappy with a regime, but only undermined legitimacy will provoke a revolution.

Fights for legitimacy capture ancient instinct driven responses in people that go back all of the way to our hunter and gatherer days. We are hard wired to respond to leadership disputes in very particular ways.

Even very, very bad regimes (e.g. North Korea during its worst famines, Albania) won't face revolutions if the regime's legitimacy and authority is not seriously subject to question.

There is a myth that revolutions are caused by mismanagement of government, but, in general, that is not true. It takes not just disagreements on matters of policy and/or bad results, but a perception that the current regime is insane or controlled by outside conspirators or has engaged in impermissible corruption against its purported values, to make a regime or leader ripe to be replaced.

Regimes are more vulnerable to revolutions when the regime is not well established yet, are more vulnerable where they is a succession dispute or disputed election, are more vulnerable to revolution when ideas like "democracy" delegitimize the hereditary or other basis of the existing regime, are more vulnerable when regime leaders appear to defy or disregard the basis (religious, legal, or otherwise) that justifies their authority.

2. Revolutions are normally led by secondary elites in the existing regime.

To present a viable alternative to the status quo there need to be people who are suitable to serve as leaders of a new regime, particularly if they have been somehow slighted relative to their perceived entitlement. Even if the mass of the revolution is run by the masses, the revolution will usually be lead by young military officers, university students, union leaders, religious leaders, or disaffected aristocrats who won't be getting inheritances because they are younger children.

For example, when new political parties that turn out to be viable are formed and replace some existing major political party, the leadership of the new political party almost always includes experienced politicians of repute who have defected from their prior political party affiliations.

3. Revolutions need to create a sense of identity in the revolting population.

An ethnic or "national" identity can be particularly effective in this regard but isn't the sole available identification. Revolutions basically harness tribal instincts and are most effective when a "tribe" feels like it doesn't have any meaningful way to receive a fair shake through the status quo. One of the reasons that ISIS was able to take so much land so quickly in Iraq and Syria was that neither the Iraqi government, nor the Syrian government, both of which were dominated by non-Sunni Arab factions, seemed to provide a meaningful voice for Sunni Arabs.

It doesn't really matter if the policy differences for the "oppressed" tribe are really all that substantive, indeed symbolic differences are probably more powerful. Instead, what matters is an appearance of having no advocates for your "tribe" in the current regime.

Social class, if clearly enough delineated and felt by people can form a basis for a revolution, but if it is divorced from an ethnic identity, it is less likely to. For example, even though the economic interests of whites in Appalachia and blacks in the United States have lots of similarities, the lack of a shared ethnic identity has prevented them from forming viable coalitions with each other in the past few decades.

Even an oppressive caste or slavery system can be quite stable, so long as the legitimacy of the system is not cast into doubt, the oppressed "tribe" has no one to lead them from the elite, or it is hard for the oppressed population to see itself as not part of the same social and political system as their oppressors.

Religious affiliations can be a powerful way to create a sense of "tribal" identity and a source of crystalized sense of membership in an ethnicity which often tracks religious affiliation anyway, although these are hardly the exclusive means by which this can be accomplished.

4. Revolutions need a spark.

It can be almost anything - a woman refusing to take her designated place on a city bus, a single incident of police brutality, or an incident that victimizes a member of the revolting tribe that is viscerally humiliating or disgusting in a way the causes other to feel empathy.

But, without a triggering spark that creates a unified emotional reaction, the masses that need to rise up in a coordinated matter won't be able to mobilize. Contrawise, a good spark can trigger the early phase of a revolution even without much organization on the part of would be leaders, although it will only keep burning instead of puttering out if the other elements of a successful revolution are present.

5. Revolutions need to be mishandled by those in power.

If the existing regime can find a way to promptly defuse the revolutionary movement, it will die. Crushing early revolutionaries without mercy can also be effective if this successfully creates the appearance among the revolting populace that they are hopelessly outmatched. Thus, the regime must either demonstrate that it doesn't deserve to be the subject of a revolution (or that there are viable alternatives like political campaigns through ordinary channels to a revolution), or convince potential revolutionaries that it is hopeless to try, or both.

6. Revolutions need resources.

Successful revolutions, unlike unsuccessful revolutions that start but don't follow through, need a logistical basis of funds and personnel. It can be foreign or domestic.

Economics can drive this by creating a class of overqualified and underemployed people with the time necessary to commit to the cause and little to lose. A failure to co-opt almost everyone who is competent to lead a revolution or even simply to work in it as a follower, is a common failing of a regime that creates conditions ripe for revolution. For example, a lot of Islamic revolutionary action is driven by huge numbers of young unemployed men with theology degrees in Saudi Arabia in large families who have no way to prove their own worth and lots of time on their hands, freedom and a sense of privilege.

The labor movement thrived when competent people were kept out of management by accident of birth, and deflated when more meritocratic grounds for advancement plucked the most qualified potential union leaders into the ranks of management, both directly by depriving unions of resources and indirectly by creating a perception of opportunity that those who advanced from lesser conditions by virtue of merit created when they filled leadership positions.

A subset of this is religious backing. Religions thrive when they provide a cultural center for a threatened or oppressed culture, and wither organizationally when their creeds merely replicate the status quo establishment position. People back religions actively and with resources because they think this support is necessary for their culture to survive.

7. Revolutions need to inspire self-sacrifice.

Some of those personnel need to perceive the cost as just enough for it to be heroic to die for it, or at least to take great risks of ruin for it beyond personal advantage. Unless some key figures in the revolutionary movement see their efforts as advancing the greater good of a worthy cause, the revolution will fail.

Religiously based revolutions are often good at inspiring self-sacrifice, but religious revolutions have no monopoly on this and fear of a theocracy from a regime can be just as effective in many cases at inspiring this. Still, there needs to be something that some true believers can sincerely hold onto as being worthy of self-sacrifice without question.

8. Revolutions can easily fall into stalemate mode.

The more time passes, the less likely the insurgency is to be resolved finally for either party. Many revolutions and coups are swiftly successful or fail just as swiftly; but very long standoffs where the revolutionaries don't have the capacity to replace the regime in the near future, but the regime doesn't have the capacity to end the insurgency are common and can last for decades. Even 5% or 10% of the population that is absolutely convinced that the current regime is illegitimate is enough to keep a revolution going indefinitely, but that is not enough support to bring about a change in control. Regimes don't need mass agreement with their policies to succeed, but they need nearly uniform (certainly more than 99%) belief in their legitimacy to avoid a prolonged and difficult to manage insurgency.

9. Revolutions need to convince people that their conditional support will pay off.

While every revolution has die hard supporters on each side, lots of people are contingent and conditional supporters of whichever side that they perceive will prevail who are trying to act in their own best interests in a confusing time.

The appearance of a likelihood of success can dramatically chance the tides of an uprising, as the recent attempted coup in Turkey illustrated. The President's ability to access mass media and convince people that he was alive and in control caused lots of people to mobilize in support of his regime and to remain silent as he ruthlessly consolidated control.

The perception that you can win is as importance as the reality. This is why control of communication channels is so important during revolutions - it affects people's perceptions.

10. A revolution's leaders need to have enough of a vision to act decisively until things settle down.

A revolution, should it achieve temporary control has a brief honeymoon period, like any new regime, during which it has to show that it is in control and bringing about sweeping change that benefits the winners of the revolution and justified the cause. An appearance of incompetence, corruption, or lack of direction and vision and change, can open the door to a swift and successful counter-revolution by old regime elements who can show that they know what they are doing and what needs to be done, while disavowing the most abhorrent acts or individuals of the old regime.

11. Revolutions can happen at any scale.

A revolution can involve changes in group leadership at any level that is or can be made to become sovereign (i.e. accountable to no one above them in all material things in the near term) from a group of cast aways on an isolate or group of POWs in a camp, all the way up to whole nations or even groups of nations.

But, revolutions do not generally happen at any level that is less than sovereign, because subordinate units in a hierarchy that don't claim to be sovereign can almost always have any legitimacy or succession dispute resolved by those who are higher up in the hierarchy that it is a part of.

12. Revolutions tend to come in waves.

A strategy sufficient to undermine the legitimacy of one regime will often have a contagion effect and spread to other similar regimes (a la Arab Spring, the Revolutions of 1848, the Revolutions of the late 1800s that removed monarchies, created national identities and drove Latin American independence movements, the Communist wave of revolutions, the post-colonial independence movements).

These larger waves provide intellectual fuel to undermine legitimacy, economic and personnel support and training, a model for creating tribal identity, and by the success of the early versions of the attempt give hope for those following them that it is viable.

This hope is often misfounded as the homegrown version of the revolution is often more solidly organized, well planned, and tuned to local conditions to an extent often not recognized (unless the general approach is revised substantially for local conditions as Mao did for Stalinism). But, it still creates hope and people often don't look very critically at the prior revolutions about which they are often ill informed anyway in deciding to make conditional and contingent commitments to the revolution that are critical to its success.

13. Revolutionary Leaders Often Lose Control

This is point thirteen for a reason. Revolutions rarely go as planned. Revolutions can get ugly, and revolutions can be hijacked by people who hold different ideologies or by pure opportunists. The initial revolutionary regime is often not the last. Instead, cycles of unstable revolutionary governments until one sticks are common.

For example, the Russian and French revolutions cycled rapidly over periods of days and weeks. The U.S. soon replaced its confederation with the 1789 constitution. The Glorious Revolution was short. China's Republic lasted only a generation. Rome cycled rapidly through political upheavals.

Most 20th century revolutionary regimes cycled quickly through more than one regime before stabilizing.

21 October 2016

Democrats Pumped, Republicans Not

Democrats may not be uniformly enthusiastic about voting for Hillary Clinton in particular, but they are enthusiastic about voting his year in Colorado in general.
Longtime Colorado political journalist Ernest Luning has been keeping track of ballot return figures via his Twitter account. This is not good for Republicans:
Ernest Lee Luning @eluning
Dems keep lead in CO statewide ballot returns thru 10/20: D-18,867; R-12,570; U-10,203  1/5https://twitter.com/eluning/status/789144937300267008 
9:26 AM - 21 Oct 2016 The last time that Democrats were this far ahead of Republicans in Colorado in early voting returns was…well, maybe never.
From here.

A solid Democratic lead in early voting is a sure sign that Democrats are excited by the election this year in Colorado and that Republicans are not.

20 October 2016

What Predicts Law School Grades? UPDATED.

A new, well done study analyzes the quantitative impact of LSAT scores, college grades, college major, college quality, work experience and other facts on law school grades.  (Substantially updated and expanded on October 21, 2016). Finer details are provided below the break.

The Main Relationship Discerned

The study makes a good case that if maximizing the law school grades of admitted students is the goal that undergraduate grades should get more emphasis relative to LSAT scores in the law admissions process than they do today. Typical admissions indexes don't give undergraduate GPA enough weight. In their regression model:
Each 0.06 rise in undergraduate GPA is akin to one extra LSAT point, but above 3.4, the effect doubles, so each 0.03 rise in undergraduate GPA is akin to one extra LSAT point. Thus, the difference between average and weak undergraduate GPA is material (e.g., 3.0 vs. 3.3 is akin to five LSAT points), but not as powerful as the difference between good and elite undergraduate GPA (e.g., 3.5 vs. 3.8 is akin to 10 LSAT points).
The strong admissions bias for students with high LSAT scores and low undergraduate GPAs over comparable students with low LSAT scores and high undergraduate GPAs (the former are roughly twice as likely to be admitted as the latter) has no support in subsequent law school performance. But, these biases may be a necessary response to U.S. News and World Reports law school ranking incentives.

STEM and to a slightly lesser extent economics, finance and accounting majors (3.5 to 4 LSAT point equivalent effect), as well as applicants with 4 to 9 years of non-military work experience (a 6.5 LSAT point effect) should be preferred in the admissions process, while criminal or disciplinary history should be a more strong negative in the process than it already is now (a 7 LSAT point effect). An adjustment based upon the selectivity of the applicant's undergraduate school, which is particularly strong for less selective schools is also appropriate.

Troublingly, after controlling for all other factors in the study it found that "any self-identification as a person of color—African American, Latino/a, Asian American, or Native American—was a statistically significant negative predictor of both LGPA and 1L GPA . . . each category of person of color self-identification was akin to 9 to 9.5 LSAT points." This is equivalent to a law school GPA difference of about 0.15 grade points and affects 18% of incoming students. 

This result does not appear to involve any affirmative action effect (except that it is significantly stronger for academically weak African Americans who are in the bottom quarter of admitted law students measured by LSAT scores and undergraduate GPA relative to African Americans who are in the top 75% academically, who actually do better than other non-white, non-Hispanic students after controlling for all other factors), mostly disfavoring the "mismatch hypothesis."

The study is also quite unusual for statistics of this kind for Asian Americans to cluster with other non-white students in terms of academic underperformance (although despite below average law school grades Asian Americans are significantly less likely to be in the bottom 25% of their class, something not true for any other group of minority students holding all other factors equal).

In contrast, there were no statistically significant effects or only marginal ones, associated with gender, other amounts of work experience (except for a slight negative effect associated with prior work in the tech jobs), other college majors, graduate school attendance, military veteran status, or college leadership roles.

The 2% of students who were military veterans, and 5% who had significant criminal or disciplinary records were the most heterogeneous groups in the study.

The Limits Of Explanatory Power

The tools available to admissions officers at law schools are still not wonderful at explaining variance in law school grades.

The full model in the study predicts 26% of the variance in law school GPA and 28% of the variance in first year law school grades, and would predict slightly less if statistically insignificant variables were removed. A regression model with only LSAT scores and undergraduate GPA to first year law school grades explained 22% of the variance in a large LSDAS study and 15% in this study.
LSAT is stronger at predicting first-year grades (the correlation between 1L GPA and LSAT, and 1L GPA and UGPA, are 0.36 and 0.27, respectively); UGPA is slightly better at predicting cumulative grades (the correlation between LGPA and LSAT, and LGPA and UGPA, are 0.28 and 0.29, respectively)." Note that the amount of variance explains is the square of correlation, not the correlation itself. "LSAT and UGPA had a positive and modest correlation of 0.187. . . . in a study of 152 law schools between 2011 and 2012, finding the average correlations between LSAT and UGPA are close to zero and range from -0.45 to 0.24[.]
The best predictor of law school grades is the not entirely independent factor of first year law school grades. First year law school grades explain 68.8% of variance in overall law school grades.

If this discrepancy between the predictive power of admissions data and the predictive power of first year grades is typical, perhaps a sink or swim approach, while wasteful, is worth serious consideration.

19 October 2016

Mosaicism And Brain Disorders

Genetic epilepsy and developmental disorders in non-affected parents are sometimes de novo disorders, but are also sometimes caused by mosaicism in a parent, which means that the parent has one kind of DNA in some cells and a different kind in other cells. As a result, the risk factor genes may evade detection by normal DNA tests that only look at one, easy to access site. These variants can make their way a child's DNA in vitro after initial fertilization of the egg giving rise to a zygote.

23 and Me Releases Nail Biting Paper

We describe the discovery of genetic and phenotypic associations with “nail biting,” technically known as onychophagia. Over 180,000 participants who consented to research in the 23andMe customer base responded to the question “How often do you bite your nails”; 37% reported biting their nails and 7% said they bite very often. Consistent with the literature, “nail biting” was correlated with “conscientiousness” and “neuroticism” of our five dimensional personality questionnaires. Individuals who become nervous easily or are moody report a higher frequency of nail biting. 
Our genome-wide scan identified 21 significant associations (p < 5e-8) with nail biting. We identified a loss of function variant (rs117612447, p=4.6e-22) in KRT31, a keratin gene involved in hair and nail formation, and a variant (rs10876505, p=5.5e-9) near HOXC13, a gene linked to nail and hair developmental disorders. Six of the identified loci (rs713843, p=4.2e-26; rs35754740, p=4.8e-11; rs4776970, p=7.4e-11; rs4775313, p=8.4e-11; rs62264775, p=9.4e-9; rs149994299, p=2e- 8) were also associated with BMI in the same direction. Five of the identified loci (rs1442883, p=3.8e-19; rs8095324, p=1.7e-13; rs7837754, p=3.3e-12; rs7411445 [NEGR1], p=8.6e-10; rs2977694 [CSMD1], p=7.2e-8) were also associated with “sweet tooth," but in the different directions. The NEGR1 and CSMD1 regions that have been previously implicated in psychiatric disorders. We also identified variants near GRIN2A in 16p13.2 (rs2014151, p=6e-19) and near NRG1 (rs13255543, p=5.7e-13). Mutations in these two regions have previously been linked to diseases such as autism, schizophrenia, and bipolar disorder. 
We estimated a positive genetic correlation between nail biting and BMI (LD score rg=0.17, p=1.46e-14). We found a near-zero genetic correlation between nail biting and sweet tooth. Although they shared many associations, the effects from those pleiotropic loci are not in the same direction. Overall, our findings revealed genetic contributions to nail biting. They also point to a possible connection between nail biting, BMI, and taste perception, which is interesting in light of prior findings that BMI GWASes implicate neural regulations; personality factors such as anxiety and the ability to cope with stress have been discovered to change hormones and act on taste. Our study may provide molecular evidence for neural mechanisms underlying personality and taste.
C. Tian, J. Tung, and D. Hinds of 23andme "Genome- and phenome-wide study of “nail biting”: Not just a habit." ASHG Conference Presentation 276 (October 2016)/

Your Genes Are More Important Than Environment In Determining If You Get Fat

Your genes have an impact about seven times as great as your socio-economic status on weight gain among people in the United Kingdom. But, note that the U.K. may lack the extreme deprivations found in other places and thus have less variation due to socio-economic status than the U.S. or less developed countries. 

There are 69 known genetic risk factors for obesity that have been reduced to a genetic risk score. On average, each allele in the score that is present increases the weight of a 5'8" person by 0.737 pounds (in theory an average difference of 50.853 pounds between with no risk alleles, and someone with all of the  risk alleles for a 5'8" person). The weight gain associated with the genetic risk is 7% higher than average for someone in the bottom half of the socio-economic scale, and is 7% lower than average in someone in the top half of the socio-economic scale. (The combined 14% difference from top to bottom is about 1/7th of the total.)
Statement of Purpose: Susceptibility to obesity in today’s environment has a strong genetic component. However, little is known about how genetic susceptibility interacts with modern environments and behaviours to predispose some individuals to obesity whilst others remain slim. Social deprivation is associated with a higher risk of obesity but it is not known if it accentuates genetic susceptibility to obesity. Previous gene-obesogenic environment studies have been limited by the need to perform meta-analyses of many heterogeneous studies and studies have not necessarily corrected for statistical artefacts such as different variances between groups (heteroscedasticity). We aimed to use 120,000 individuals from the UK Biobank study to test the hypothesis that objective measures of relative deprivation in the UK accentuate genetic susceptibility to obesity. 
Methods: We used the Townsend deprivation index (TDI) as a measure of deprivation and a 69-variant genetic risk score (GRS) as a measure of genetic susceptibility to obesity. We tested the association of the genetic risk score with BMI in high and low socioeconomic groups and tested for interactions (using the continuous TDI as an exposure measure). To test the specificity of any apparent interactions we repeated analyses using a simulated environment (that was correlated with BMI in the same way as TDI) as an interaction term and using randomly selected groups of individuals of different BMIs.  
Results: We found evidence of gene-environment interactions with TDI (Pinteraction=3x10-10). Within the 50% of most deprived individuals, carrying 10 additional BMI-raising alleles was associated with approximately 3.8 kg extra weight in someone 1.73m tall. In contrast, within the 50% of least deprived individuals carrying 10 additional BMI-raising alleles was associated with approximately 2.9 kg extra weight. When we used a simulated environment or randomly selected groups of individuals to be of different BMIs, we observed only nominal evidence of apparent interaction, (simulated environment Pinteraction = 0.04; randomly selected groups: Pinteraction=9x10-4) suggesting the interaction was specific to TDI.  
Conclusions: Our findings provide evidence that social deprivation accentuates the genetic predisposition to obesity.

There Are Two Distinct Forms Of Autism Syndrome Disorders

Autism Spectrum Disorder (ASD) cases can be clustered into two types based upon the risk factor genes involved. This division coincides with a distinctions between severe symptoms and less severe symptoms. The two clusters appear to have distinct causes. Basically, it appears that there are two different conditions that happen to have symptoms that resemble each other. 

Knowing what causes a particular individual's ASD could be critical in figuring out what kind of therapies or symptom management strategies are likely to work best for a particular individual.
Autism Spectrum Disorder (ASD) is a complex neurodevelopmental disease primarily characterized by deficits in verbal communication, impaired social interaction and repetitive behaviors. It exemplifies profound clinical heterogeneity, which poses challenges in diagnosis and treatment. Genetic studies have pointed to hundreds of presumptive causative or susceptibility genes in ASD, making it difficult to find common underlying pathogenic mechanisms and suggesting that multiple different genetic etiologies for ASDs influence a continuum of traits. 
Deep phenotyping analysis allowed for re-categorization of genetic variants. Our previous analysis suggested the existence of two significant subgroups within the existing ASD classification. To investigate this hypothesis in greater detail we have performed in-depth analysis using phenotypic and genetic data from Autism Genetic Resource Exchange (AGRE) and Autism Genome Project (AGP). Our initial findings on both phenotypic and genetic data (1,262 cases and 2,521 controls using familial transmission disequilibrium test) suggest existence of two groups that range in severity. Findings were replicated in a validation dataset. Genetic risk scores (GRS) were used to sum up the total effect of several single-nucleotide polymorphisms characteristic of the two clusters. The high discriminatory ability of the genetic risk score to define cluster 1 from cluster 2 case group at different combinations of sensitivity and specificity was assessed and clearly demonstrates strong signal with AUC being 0.74. There is a significant signal differentiating the 2 clusters relying on non-genetic risk factors and even greater signal when using both non-genetic risk factors and GRS. The detection and validation of the two groups allowed us focus on convergence of findings at the pathway level. ASD heterogeneity was leveraged via large scale pathway analysis within those two categories, which led to identification of a driver gene set across significant pathways. The significant pathways in cluster 1 (severe, affected = 300) include autoimmune disease, vitamin B6 metabolism, whereas in cluster 2 (non-severe, affected = 921) included oxytocin signaling pathway, WNT signaling pathway and glutamatergic synapses (all at P < 0.001). We envision that systematic study of all genomic pathways obtained given a set of redefined categories will yield profound findings for ASD even in the absence of strong individual variant information.
S. Smieszek and J.L. Haines., "Autism redefined: Genomic pathway approach to autism spectrum disorder." ASHG Conference Presentation 33 (October 2016).

18 October 2016

I Voted!!!!!!!!

My Colorado mail in ballot arrived in the mail this afternoon after being sent out yesterday. I completed it and turned it in at the drop box at the Washington Park Recreation Center this evening.

It is three weeks before the ordinary election day of old.

When Life Gets Weird, The News Gets Weirder

A man suspected of leaving a backpack bomb outside a police station in a small Colorado mountain town tried several times to remotely detonate the homemade device using a cellphone but failed, according to court documents. 
David Michael Ansberry, 64, of San Rafael, California, was arrested this weekend in Chicago after surveillance video captured him at the stores where he bought the cellphones that he expected to trigger the explosive in the town of Nederland, investigators said. 
He was easily recognizable because he is 3 feet 6 inches tall and 100 pounds and wore a ponytail, a ball cap and using crutches. The Nederland police chief told investigators he spotted a man matching Ansberry’s description leaving a hotel as it was being evacuated during the Oct. 11 bomb scare.
From here.

I don't have any idea what all of this is about, but I've been at pretty much the exact spot where the bomb was left within the past year or so.

17 October 2016

Colorado Ballots In The Mail

This is the first year that Colorado will have an all mail-in ballot election. Ballots for every active voter in the state were mailed today and should arrive at voter's mailboxes in the next few days.

If you don't receive a ballot by the end of the week, you should investigate the situation to see if your voter registration has lapsed.  A lapsed voter registration or lack of voter registration may in principle be corrected anytime through election day, but it is best to deal with it as soon as you learn that there is a problem.

Incidentally, Colorado is currently projected to once again (as it has been in the previous two Presidential elections) be the marginal state that pushes the winning Presidential candidate over the 270 electoral vote mark (sorting state or separate Congressional district that can award electoral voters results by the percentage of the popular vote for the winning candidate).

Any swing in the mood of the voters necessary to change the outcome of the election has to be big enough to swing Colorado.

Frivolous Bits

* Over the weekend, my children finally dragged me into the future and got the entire family on Spotify Premium (previously one child had a paid account and the other had a free one). The amount of technology and legal coordination that goes into that service is astounding and I still have no idea precisely how options like the ability to download content onto a device work in terms of licensing arrangements and technological standards.

It does take more awareness of the music you like than a Pandora account, but a decade of attention to a Pandora station (which keeps meticulous records of your preferences) made that easier than it might have been.

* I also discovered an interesting new webcomic that is not connected to any of the usual web comic aggregator sites that I could discern featured on the Hiveworks webcomic aggregator site. It is called "Cut Time", is well drawn with a nice plot, and available here. It features the adventures of a magical blind girl who can see through the eyes of a familiar bird, a witch doctor, and a misfit heir of a noble house with no interest in the niceties that this position in society requires. Also, somebody is trying to kill the heir to the noble house.

The Year That Republicans Alienated College Educated Whites

One of the oft repeated fact of the 2016 election cycle is that Trump does well among working class whites, while Clinton does well among college educated whites.

This is true, but not because working class whites have suddenly started trending Republican. That transition took place in the 2000 election between Bush and Gore and has stayed that way ever since then (arguably based upon the particular candidates Democrats have advanced to represent them; Bill Clinton knew how to appeal to working class white voters in a way that subsequent candidates have not).

But, Republicans have had an edge with college educated white voters in every recent election before this one. Prior to the 2000 election, they did better with college educated whites than with non-college educated whites. For the last sixteen years, however, they done better with non-college educated whites than with college educated whites.

Trump is actually performing more poorly among whites both with and without a college degree compared to Romney in 2012. But, Trump has lost just 3 percentage points of an edge among white voters without a college degree, while he has lost 29 percentage points with college educated whites.

It is entirely plausible that this stunning turnabout will be a one time blip, rather than a sustained trend, particularly because this shift is disproportionately among white women with college degrees in a race between an unapologetically misogynist serial rapist Republican nominee against the first woman to be a major party nominee for President in U.S. history. White female college educated voters are unlikely ever again to face such clear gender driven signals in a Presidential race.

Republican voters as a whole are still more affluent than Democratic voters, and Trump voters are as affluent or more so than his GOP competitors in the primary election were. But, what has really distinguished Trump supporters from voters who have abandoned ship to support Clinton, tell pollsters that they are undecided, or support a third party candidate, is racial resentment. This is so pervasive among non-college educated white voters that it has swamped gender effects; while it is far less common among college educated white voters. As Dylan Matthews explains at this link:
[S]upport for Trump in the primaries strongly correlated with respondents' racial resentment, as measured by survey data. Similarly, Republican voters with the lowest opinions of Muslims were the most likely to vote for Trump, and voters who strongly support mass deportation of undocumented immigrants were likelier to support him in the primaries too. . . . 
Even in the general election, while support for Trump is correlated most strongly with party ID, the second biggest factor, per the analysis of Hamilton College political scientist Philip Klinkner, was racial resentment. Economic pessimism and income level were statistically insignificant. 
The message this research sends is very, very clear. There is a segment of the Republican Party that is opposed to racial equality. It has increased in numbers in reaction to the election of a black president. The result was that an anti–racial equality candidate won the Republican nomination.
This year's electorate is split by race and social class, not be income per se.

Education has always been as much of a means by which to socialize students into the norms of a society or social class, as it has been to provide practical skills or increased intelligence (particularly in the liberal arts and many business majors). And, in this case, higher education has successfully, over two generations, socialized the college educated population of the United States into anti-racist norms. It has also taught students to reject conspiracy theories and accept credible information from the mass media to usually have at least some grain of truth to it. 

But, this process did not socialize affluent individual who achieved their economic success without a college education. Among these individuals anti-racist norms run only skin deep when they are present at all.

Also, religious distrust of science has driven many who want to adhere to their religion's doctrines to adopt a conspiratorial assumption that science, higher education and the media are controlled in a way that subverts religious truth. In that context, other ridiculous lies and conspiracy theories easily piggy back on a pre-existing structure that neutralizes reality based checks on alleged religious truths. Despite the fact that Trump is not the human being that many Evangelical Christians would like, the fact that he shares their conspiratorial worldview makes him more credible in their eyes than a candidate like Hillary Clinton who embraces truths about the world (which the college educated call "reality") and a worldview in which it inevitably follows that many of their literalist religious beliefs are false. That is a bitter pill indeed for many of these believers to swallow.

Thus, in a very real sense, the answer to Trump's success is in Genesis.

Colorado Supreme Court Ignores Big Picture In Sentencing Ruling

Schneider sought review of the court of appeals’ judgment affirming his convictions and consecutive sentences for two counts of sexual assault. The jury returned guilty verdicts on one count of sexual assault of a physically helpless victim and another count of sexual assault by causing submission of a victim by means of sufficient consequence reasonably calculated to cause submission against the victim’s will, based on evidence of a single, continuous penetration of the same victim; and the trial court imposed mandatory consecutive sentences for conviction of separate crimes of violence arising out of the same incident.
From here (emphasis added; quoting the official syllabus of the case, which is Schneider v. People, 2016 CO 70).

Despite the fact that the Defendant raped a single individual in a single incident, both the Colorado Court of Appeals and the Colorado Supreme Court affirm two consecutive sentences for the offense, basically doubling the jail time involved because the defendant was prosecuted on two theories and the jury agreed with both of them. This is an obviously unjust result that the courts managed to talk themselves out of caring about. The fact that the decision is unanimous in the face of such a clear case is particularly troubling.

Also, keep in mind that the more serious of the two offenses is a class 3 felony (one notch below second degree murder) and is subject to an indeterminate sentences with a minimum set by the judge and a maximum life sentence. It is not as if concurrent, rather than consecutive sentencing would have resulted in a lenient sentence for an admittedly serious crime.

This kind of ruling is particularly troubling because a great many particularly long sentences occur because legislatures set sentences largely based on what they deem appropriate when there is a single offense and don't contemplate carefully enough how consecutive sentences can result in a punishment that is more severe than is appropriate for the course of conduct giving rise to the punishment.

For example, I suspect that many legislators would be stunned to learn that consecutive sentences were possible in the fact pattern presented by this case.

14 October 2016

The U.S. Does Not Have A Significant Child Marriage Problem

Campaigns to end child marriage in the United States are grossly exaggerating the magnitude of the problem, are misleading the public regarding the circumstances of the typical child marriage, and misleadingly imply that girls who get married before turning age eighteen are not acting in their own best interests or in a way contrary to how they would act if they had free will. 

In the U.S., child marriage is not the legitimate and serious problem that it is in many poor countries in Africa and Asia.  In large part, this is because the institution of marriage in the U.S. gives husbands and parents far less coercive power than in many other countries. The people leading those campaigns may be sincere and misinformed, but as far as the U.S. is concerned they are wrong.

Teen pregnancies have become dramatically less common in the last twenty-five years. Teens who marry before turning eighteen are overwhelmingly either pregnant or are already mothers and are marrying the father of at least one of their children. Teen marriage is become more rare as teen pregnancies become much less common. Marriages of girls aged 13 to 15 which require judicial consent in New Jersey are particularly rare in New Jersey (even relative to the number of teens that age who have children). There is no evidence whatsoever that any significant number of girls in the U.S. are being forced into child marriages that are not in their best interests with partners that they have not chosen themselves. If there is a child marriage problem in the U.S., it is pregnant teens and teen mothers in New Jersey are not getting married as often as their own best interests would suggest, rather than that they are getting married too often.

The main problem in the U.S. is that unmarried late adolescent are getting pregnant too often, not that they are getting married too often or that they are being pressured by their parents to have sexual relationships with older men. But, incredible progress has been made in reducing the number of teen pregnancies in the U.S. in the last twenty-five years (particular in the more recent years), resulting in far fewer teen mothers and far few abortions by pregnant teens. This is mostly due to greater contraception use by teens.

The Facts

There were roughly 90,000 girls in New Jersey who were aged 13-15 years old and 60,000 girls who were 16 or 17 years old, when the last census was taken in the year 2010. According to an opinion piece in the New York Times (corrected from the original erroneous version of the article):
3,481 children were married in New Jersey between 1995 and 2012. Most were age 16 or 17 and married with parental consent, but 163 were between ages 13 and 15, meaning a judge approved their marriages. . . .  91 percent of the children were married to adults, often at ages or with age differences that could have triggered statutory-rape charges, not a marriage license.
This is about 193 children including 9 between ages 13 and 15, in an average year.

Overwhelmingly,  people who marry before age 18 are girls rather than boys, and this is even more strongly the case when people marry between ages 13 and 15.  So, the annual marriage rate for girls between 13 and 15 in New Jersey in the relevant time period is approximately 1 in 10,000 (i.e. 0.01%). In truth, there are probably significantly more 15 year olds than 14 years olds, and significantly more 14 year olds than 13 year olds in that sample.

The odds that a New Jersey girl will marry any time in her life before age 16 is about 0.03%.

The annual marriage rate for girls aged 16 and 17 in New Jersey in the relevant time period is about  1 in 326 (i.e. 0.3%).

The odds that a New Jersey girl will marry before turning 18 in New Jersey is about 0.63% (i.e. about 1 in 159).  The other 99.37% of New Jersey girls will not marry before age 18, although many who do not marry will have children.

There were about 1,030 children born to mothers aged 15 to 17 in New Jersey in 2014 (a rate of 5.8 per 1,000 girls aged 15-17 down 15% from last year and down 78% from a peak in 1991), and about 37 children born to mothers under age 15 in New Jersey in 2014. Thus, there were 1,067 children born to mothers under age 18 in New Jersey in 2014 and roughly 193 marriages of girls under the age of eighteen. 

Unsurprisingly, black and Hispanic girls are roughly six times as likely to be mothers under age eighteen as white girls in New Jersey.

There are about 5.5 teen births by mothers under the age of 18 in New Jersey for every marriage by a girl under the age of 18 in New Jersey (8% of the births by teens under the age of eighteen are second or later births). About 94% of births to mothers aged 15 to 19 in New Jersey are non-marital. So, there were fewer than 64 births each year in New Jersey to married mothers under the age of eighteen in 2014. Indeed, the number is probably significantly less because if 15-17 year old mothers were equally likely to be married as 18-19 year old mothers in New Jersey, the number of married teen 15-17 year old mothers would be 18 in 2014, and in fact, it is almost surely the case that 18-19 year old mothers in New Jersey are at least somewhat more likely to be married than 15-17 year old mothers in New Jersey.

The rate at which teens under the age of eighteen become pregnant in New Jersey is roughly 8 times the number of births to teens under the age of eighteen in New Jersey. Some of that is the statistical quirk the arises because more than three-quarters of seventeen year olds who get pregnant ultimately give birth when they are eighteen. But, probably something like a third of those pregnancies of girls under eighteen years of age end in miscarriage or stillbirth, and roughly half of those pregnancies are terminated with abortions.

The facts suggest a couple of things.

First, the nearly 80% decline in births to mothers under eighteen since 1991 in New Jersey has probably been accompanied by a similarly great decline in the number of marriages by women under the age of eighteen in New Jersey since then. So, there were probably far fewer than 193 teens under the age of eighteen and were probably fewer than 9 teens under the age of sixteen married in 2014 in New Jersey.

Second, it suggests that most teenaged girls under eighteen who marry with judicial or parental consent in New Jersey, rather than being pregnant, are already mother's of the husband's children by the time that they marry.

No doubt some girls under the age of eighteen in New Jersey who marry are not pregnant or mothers when they marry, but I suspect that this is a small minority of the total and consists almost entirely of sixteen and seventeen year olds who require parental approval but not judicial approval to marry.

Is This Outrageous?

The author of the opinion piece is outraged by these facts and appalled that a judge would every approve a marriage in these circumstances. 

I am not nearly as concerned, even though I am quite sympathetic to the concerns raised by her about child marriage in foreign countries.

Both spouses consent is required in a request to seek judicial approval for a marriage. Because such cases are so rare, judges generally pay close attention to these requests. It would be highly unusual for a judge to grant a request for a girl between the ages of 13 and 15 to marry without hearing live testimony from the prospective wife, the prospective husband, a parent or guardian of the minor, and at least one other witness such as a social worker, a friend of the bride, or a sibling of the bride. Usually, a court would appoint an attorney and/or a guardian ad litem, both of whom would be officers of the court and known to be trustworthy by the judge, to represent the girl in the case. 

There is no judicial assembly line out there blindly marrying off unwilling 13 years olds to adult men so that they can get the girl pregnant legally in the first place. I would be surprised if more than 2 or 3 cases in the entire 18 year time period studied involve that set of facts.

Generally, in cases where judicial approval is sought for the marriage of a teenaged girl between age 13 and 15, the girl is already pregnant, does not want an abortion, does not want to put her child up for adoption, and seeks marriage to the father who is an adult capable of providing for her. Indeed, it seems likely that in most of these cases, the wife has already had a child with the husband.

While the statute may not require that she be pregnant or a mother already, I have never heard a single case in a U.S. state in modern times authorizing the marriage of a boy between the ages of 13 and 15 to an adult woman, or authorizing the marriage of a girl between the ages of 13 and 15 who is not pregnant or already a mother. I would be quite surprised to see a judge authorize the marriage of someone aged 13 and 15 who was not pregnant or already a mother in the State of New Jersey in the time frame in question, although it isn't, strictly speaking, impossible that this could have happened some of the time.

Often there are pressures on the girl from the father, her family, her friends, her faith, and society at large for her to have or to not have an abortion. Myriad legal struggles have gone into determining the circumstances under which this decision is made by a teenaged girl. More often than not, this issue was already decided by the time a case seeking permission to marry reaches a judge, the pregnancy is quite advanced, and only a very late term abortion would be possible if it is possible and legal at all. So, once a judge faces a young teenaged girl's marriage request, the abortion question is over and done with and the girl has made a decision, albeit not final and binding, not to put the child up for adoption.

Generally, the girl's parents are not pleased by the fact that she got pregnant as a young teenager and did not encourage her to get into that situation, and often the father's family isn't pleased either.

Generally, the choices available to the girl and the authorities once she has rejected having an abortion or having the child and putting it up for adoption, who often already has and is raising a child, are for her to: (1) have a child and be a single mother while prosecuting the father for statutory rape (which would usually involve relatively a short prison sentence) with or without a parallel civil action to terminate his parental rights which might or might not succeed, (2) have a child and coparent that child with the father outside of marriage, or (3) have a child and marry the father.

Often, a teenaged girl and an adult man who gets her pregnant in a case where a judge is asked to approve their marriage have feelings for each other and are reasonably compatible with each other.

In the usual case, where a judge is asked to approve the marriage of a thirteen to fifteen year old, if the judge says no and the parents of the child support the marriage, the child can marry with parental permission one to three years later without judicial approval in any case if the child, the other spouse and the parents of the child still approve of the marriage at that time.  

In the meantime, the judge cannot generally prevent the couple from cohabiting, is generally bound by law to set the terms of their relationship as coparents if either prospective spouse requests this, and has no power to compel a prosecution of an adult spouse of a minor for statutory rape without the assent of the District Attorney in the jurisdiction.

A teenaged girl who is not married to the father of the child generally has significantly fewer rights vis-a-vis the father of the child than a child who is married to the father of the child. 

For example, a married girl would generally be entitled not just to chid support, but also to alimony and an equitable share of marital property owned solely in the name of the father if their relationship ended. She would also have rights to pursue a wrongful death claim and to claim Social Security and insurance benefits from him if he died in her own name. The parental responsibilities and child support obligations of the teenaged girl and the father in the event that they break up would ordinarily be identical.

So, it would rarely be the case that a pregnant teenaged girl who is having a baby with an adult man is going to be better off as a single mother who cohabits or coparents with him, than as his wife.

Of course, if they do not get along, or if she feels that he is economically worthless, she doesn't have to seek to be married.  

Another factor that makes a child marriage far less coercive in the United States than it might be in some other countries. "No fault" divorce is widely available in the United States, without parental permission as the marriage emancipates her, so the wife can legally, unilaterally leave the marriage at any time, while and enjoy any and all marital rights that she may have as a result of the marriage.

Of all the cases that come before judges in the United States, a request of a pregnant thirteen to fifteen year old girl to marry the usually adult father of her child is rarely a difficult one and approving the request is rarely a bad decision.

Sixteen and Seventeen Year Old Spouses

The case of 16 or 17 year olds (again overwhelmingly girls marrying older men, but not quite as uniformly pregnant at the time) who marry with parental consent, do involve less scrutiny. But, these young women are generally a year or two away from having the right to marry unilaterally.

Indeed, at age 16 or 17, the public is at least assured that a parent consents to the marriage and that the spouses, however grudgingly, have consented enough to say so and sign the marriage certificate.

Eighteen years old may be a widely recognized point to draw the line regarding adulthood, but ultimately it is an arbitrary one that has varied somewhat in time and place over the ages. Often adulthood is as old as twenty-one, and for many practical purposes we treat sixteen year olds as adults.

It has been perfectly normal for much of U.S. history for sixteen or seventeen year old women to marry older men, and it is a developmental fact that a sixteen year old woman, on average, is more mature than a sixteen year old boy and is probably, on average, at least as mature as an eighteen year old boy.

These couples are never subject to statutory rape charges, because the age of consent in New Jersey is sixteen years old.

If they marry, these women retain the privileges of "no fault divorce" (without parental permission as their marriage emancipates them) and will almost alway have more rights that protect them as spouses than as cohabiting girlfriends of their spouse's. Parents may have some legal authority over their sixteen and seventeen year old children, but their practical authority over them should not be overstated. Sixteen year olds can drive, can be legally hired to do most kinds of jobs that they are qualified in any way to do, and can usually rent an apartment if a co-signer can be obtained or if a landlord isn't very particular. As a result, if they do choose to divorce, they are likely to be able to manage as well as most other adults do at that point, and if they married at sixteen or seventeen years of age, they are likely to be quite close to being eighteen years old adults in any case, once they are actually divorced.

It is also worth keeping in mind in all of these "child marriage" cases that marriage grants a husband much less power of his wife in the United States than it does in many other counties. It does not give the husband the legal right to rape his wife. It does not give the husband any immunity of any kind from criminal or civil liability for assault, kidnapping, theft or other crimes or torts. It does not give the husband legal control over property titled in the wife's name (other than an inchoate interest in marital property upon divorce) and does not give him the right to claim her wages. It does not give the husband the right to insist that she cohabit with him or to prevent her from separating from him at any time.

The rights of a husband and wife with regard to their children are the same as the rights of an unmarried pair of parents for whom there is acknowledged paternity.

Simply put, there is little reason to think that the requirement of parental consent to marry when a young woman is sixteen or seventeen years old facilitates a significant number of involuntary weddings. And, even if legal marriage were prohibited in the cases where parents have such power over their late teenaged daughters that they can force them to marry in the United States, it is likely that the young women would have been forced into religious marriages without any of a legal benefits of a civil marriage, in any case.

Bottom Line

There are certainly many places in the world in Asia and Africa where girls and young women are forced by their parents into marriages to which they do not meaningfully consent, and those women then go on to have negative experiences as a result, ranging from marital rape and domestic violence to childbearing before the mothers are developmentally ready to have children, to losses of economic and educational opportunities, to loss of autonomy and happiness in life.

This is not the norm in the United States at this time, and it is understandable that someone might b concerned that these undesirable practices could be exported to the United States by immigrant populations or by extremist (by U.S. standards) religious groups.

But, because of the way that the institution of marriage has evolved in recent U.S. history, and because of the nature of the legal safeguards in place in cases of marriages by minors, these relatively rare marriages are not something about which great public concern is appropriate. 

If anything, the bar to marriage by minors where it is appropriate is set too high and pregnant teens are more often harmed by a failure to marry than by being given permission from a judge or a parent to marry. The fact that significantly less than 6% of mothers in New Jersey who have children under the age of eighteen are married, when the economic benefit of doing so to the mother are very often positive, suggests that child marriage is too hard to get, rather than too common, in New Jersey.