Tuesday, May 31, 2016

New longitudinal study finds epidemic of fear, not an epidemic of child abuse

When a longitudinal study of children in California found that they are more likely to be reported as allegedly being abused or neglected over five years than over just one, this led to some remarkable leaps of logic. It was suggested that the study shows “child maltreatment is a far larger societal epidemic than we’ve previously thought” and “a reckoning is coming in child protection.”

I’ve previously noted that it isn’t exactly a surprise to find more of something when you follow a population for five years than when you follow it for one.

But the other problem is assuming that a whole lot of people picking up the phone to report their slightest suspicion of any form of “maltreatment” means that there is actually an epidemic of child abuse.
A new longitudinal study has been released in Britain that has some results similar to the California study. While there are some differences both in terminology and in the process for handling reports about children deemed “at risk,” this study found that 22.5 percent of English children – about 150,000 – are reported to child welfare authorities by the age of five. And 75 percent of those 150,000 children will be subject to some sort of government intrusion into their family lives.

Since well over half of all such reports, known in England as “referrals,” occur between ages 5 and 18, it’s easy to see that the proportion of families subject to this intrusion over the course of the child-rearing years is staggering. And this is an average figure.  One can only imagine how often this happens to families in impoverished communities.
One could, of course, interpret this as meaning England is a cesspool of depravity where a “societal epidemic” of child abuse is so huge that the majority of parents abuse their children.
 But the authors, Professors Andy Bilson and Katie Martin of the University of Central Lancashire School of Social Work, offer a more sensible interpretation: It’s an epidemic of fear, not an epidemic of child abuse.
The number of referrals, investigations and children taken from their homes and placed in foster care in England all have shot up since 2008. There is no evidence that child abuse skyrocketed in this time. On the contrary, sexual abuse remained largely unchanged and physical abuse actually decreased.
The increase in the English equivalent of “substantiation” was almost entirely in the areas of “neglect” and “emotional abuse” – a particular paradox since a child abuse investigation often is, in itself, emotionally abusive for a child.
And while reports we would call “substantiated” increased by 40.4 percent, reports we could label “unfounded” more than doubled.

The real reason for the increases in reports and in foster care is largely a result of a nationwide foster-care panic resulting from a notorious child abuse death in London in 2007, known as the “Baby P case,” and other cases in Britain.

As Bilson told the Times of London: “The tragic deaths of children . . . and desperation not to be the one who misses the early signs next time have led to a climate of suspicion with child protection investigations spiraling.”
Or, as one newspaper headline more succinctly put it: “150,000 Kids on Baby P. Fear List.”
In addition to the harm to families needlessly investigated, Bilson said that the high number of reports makes it harder for caseworkers to identify the small group of children in genuine danger. And, he says, there is no evidence that skyrocketing numbers of reports and the increase in foster care have done anything to make children safer. 

Indeed, for at least four years after the Baby P. case made headlines, child abuse deaths increased.
The study concludes:
These policies bring high levels of suspicion, fear and shame on a considerable proportion of families in the most deprived areas where this activity is concentrated.
 This is done without evidence that the individualised, investigative approach is effective in preventing further harm. Alternatives include a more humane developmental social work orientation and approaches that promote cohesion in neighbourhoods and reduce deprivation and poverty. But, to achieve this, we need to step away from our current preoccupation with the search for ever more parents to investigate and blame.
That’s hard to do in the midst of an epidemic of fear.

Tuesday, May 24, 2016

Indian Child Welfare Act: The real tragedy is that it’s not enforced

Before ICWA: The Carlisle Indian Industrial School, 1900
This is an expanded version of a column that appeared originally in the Chronicle of Social Change

Casey Jo Caswell of Lansing, Mich. made a terrible mistake. Homeless and jobless, she turned to Michigan’s child welfare agency for help raising her son, Ricky. But the agency offered no help with housing, no help with a job, and no help with education.  They told her to surrender the child to “temporary” foster care, and then rushed to terminate her parental rights.

Ricky was placed with middle-class foster parents in a nice, big home, first as a foster child, then
it became his adoptive home.

Once, during a counseling session, the Detroit News reported, the boy was playing with two plastic horses, when he said: “This little horse is going to die if he can’t be with his mother.” That proved prophetic.  Ricky Holland’s white adoptive parents murdered him. They stuffed his body in a trash bag and left it by the side of a road.

Yet in all the years since, no one has suggested that, because of this horror story, the so-called Adoption and Safe Families Act (ASFA), the law that spurred the rush to place Ricky in the home where he was adopted to death, should be repealed or curbed. That’s understandable. There’s an excellent case for curbing ASFA, but it shouldn’t be built on horror stories that unfairly stigmatize entire groups.

Yet Marie Cohen offers a similar horror story involving a Native American home (recycled from the far right Goldwater Institute ) as the only evidence in support of her claim in a column for the Chronicle of Social Change, that the Indian Child Welfare Act (ICWA) should be curbed – or maybe never should have been passed at all; she’s not clear about that. The Goldwater Institute has filed a class-action suit against ICWA.

Cohen dismisses in a single sentence the horrors inflicted on Native American children that led to passage of ICWA. In fact, from the 19th Century through the 1960s, American child welfare agencies tried to effectively eradicate Indian culture and, indeed, Indian tribes, through the expedient of taking away children. First, they were warehoused in hideous orphanages. Later, there was a campaign of mass adoptions. Melissa Harris Perry called the orphanages an “explicit cultural extermination mission.” The Lakota People’s Law Project is calling for a Truth and Reconciliation Commission.

By the mid-20th Century people stopped actually saying “kill the Indian, save the child” but it took ICWA to change practice – though it hasn’t changed nearly enough.

Opponents of ICWA respond exactly as the Supreme Court majority dealt with the Voting Rights Act: Well, yes, racism used to be a problem, but not anymore!

There are two problems with this:

First, if you curb a bad practice by passing a law and then you eviscerate the law, it’s not hard to figure out what will happen next – witness the wave of voter suppression laws that followed the Supreme Court voting rights decision.

Second, contrary to what John Roberts and Marie Cohen seem to think, racism isn’t dead.

In 2003, Dewey Sloan, chief juvenile prosecutor in an Iowa county where Native American children were in foster care at a rate seven times higher than the rate for white children told the Des Moines Register, "I don't think there's anything in any of these cases that points to something positive about Indian culture, except the culture of drugs and the culture of poverty and the culture of abuse."

Or consider what happens to Native American children in South Dakota. That state tears apart families at a rate 80 percent above the national average.  And while Native Americans are 15 percent of the population, they are more than half of all foster children. The horrors inflicted on Native American children by the state were documented in a heart-rending series by NPR – in 2011.

NPR’s findings were confirmed by a federal judge last November. And, in another lawsuit, the federal government is charging the South Dakota Department of Social Services with discriminating against Native American job applicants.

So yes, there’s a problem with ICWA alright – the problem is it’s not being enforced.

And the right-wing propaganda about ICWA has been remarkably effective – even to the point of getting some on the left to help do their dirty work.

In Santa Clarita, California, Rusty and Summer Page took in Lexi, a child who is part Native American, knowing at the outset that the plan was to return her to her own father or, if that wasn’t possible, to extended family. But when the child’s tribe moved to do just that, the Pages went to court. When they finally ran out of legal options, they created a media circus. Though it was the Pages who stalled and stalled the return of the child, that didn’t stop them from claiming in a petition  -- on Change.org – that theirs was “the only family that she has ever known.” The claim was echoed in one news account after another.

The claim is false. Lexi was in continual contact with her extended family – and they already had taken in Lexi’s sister.

Almost every news account claims the child was sent to relatives, in the words of one story, “all because of her Native American heritage” and ICWA.  But federal law also calls for giving preference to relatives, and keeping siblings together – precisely because both practices are best for children.

Indeed, in a series of tweets, Nicole Cliffe, editor of The Toast explained the entire case better than most of the mainstream media.

 Cohen bemoans the fact that under ICWA, child welfare agencies are supposed to make “active efforts” to keep families together, theoretically a higher standard than the “reasonable efforts” required for non-Native children.

But the reasonable efforts requirement was never enforced and was rendered effectively meaningless by ASFA.  In Michigan, where Ricky Holland died in his adoptive home, 40 percent of judges surveyed admitted that they lie and certify that “reasonable efforts” were made to keep families together even when they don’t really believe it. 

And the fact that four separate studies since 1996 found that 30 percent of foster children could be home right now if the families simply had decent housing further illustrates that in practice “reasonable efforts” often means zero effort. So in real life, “active efforts” means ever-so- slightly more than zero. 

When it comes to the treatment of Native American children, that’s the real horror.

Monday, May 23, 2016

New Zealand analysis should (but probably won’t) burst the predictive analytics bubble


“New Zealand Crunches Big Data to Prevent Child Abuse,” declared a Chronicle of Social Change headline on a 2015 story about The Chronicle’s favorite child welfare fad, predictive analytics, or as it should properly be called, data-nuking poor families.

The story quotes Paula Bennett, New Zealand’s former minister of social development, declaring at a conference: “We now have a golden opportunity in the social sector to use data analytics to transform the lives of vulnerable children.”
If implemented, the story enthuses, it would be a “world first.”
All this apparently was based on two studies that, it now turns out, used methodology so flawed that it’s depressing to think that things ever got this far.  That’s revealed in a detailed analysis by Professor Emily Keddell of the University ofOtago.

The studies that supposedly proved the value of predictive analytics attempted to predict which children would turn out to be the subjects of “substantiated” reports of child maltreatment.
Among the children identified by the software as being at the very highest risk, between 32 and 48 percent were, in fact, “substantiated” victims of child abuse. But that means more than half to more than two-thirds were false positives.
Think about that for a moment. A computer tells a caseworker that he or she is about to investigate a case in which the children are at the very highest level of risk.  What caseworker is going to defy the computer and leave these children in their homes, even though the computer is wrong more than half the time?
But there’s an even bigger problem. Keddell concludes that “child abuse” is so ill-defined and so subjective, and caseworker decisions are so subject to bias, that “substantiation” is an unreliable measure of the predictive power of an algorithm. She writes:
How accurately the substantiation decision represents true incidence is … crucial to the effectiveness of the model. If substantiation is not consistent, or does not represent incidence, then identifying an algorithm to predict it will produce a skewed vision …


Turns out, it is not consistent, it does not represent incidence, and the vision is skewed. Keddell writes:
Substantiation data as a reflection of incidence have long been criticized by researchers in the child protection field … The primary problem is that many cases go [unreported], while some populations are subject to hypersurveillance, so that even minor incidents of abuse are identified and reported in some groups.

That problem may be compounded, Keddell says, by racial and class bias, whether a poor neighborhood is surrounded by wealthier neighborhoods (substantiation is more likely in such neighborhoods), and even the culture in a given child protective services office.

Predictive Analytics Becomes Self-fulfilling Prophecy

Algorithms don’t counter these biases, they magnify it.
Having a previous report of maltreatment typically increases the risk score. If it’s “substantiated,” the risk score is likely to be even higher. So then, when another report comes in, the caseworker, not about to overrule the computer, substantiates it again, making this family an even higher “risk” the next time. At that point, it doesn’t take a computer to tell you the children are almost certainly headed to foster care.
Writes Keddell:
“prior substantiation may also make practitioners more risk averse, as it is likely to heighten perceptions of future risk to the child, as well as of the practitioner’s own liability, and lead to a substantiation decision being made.”

So the predictive analytics become a self-fulfilling prophecy.
Keddell also highlights the problems when even accurate data are misused by fallible human beings:
Several researchers note the tendency for individualised risk scores to be utilised in negative ways in practice, where actuarial approaches are prioritized over professional judgement. While statistical modellers may understand the tentative nature of statistical prediction or correlation … practitioners tend to treat statistical data, especially when stripped of its explanatory variables, as solid knowledge, which can function as a received truth.

But it turns out there may be one area where predictive analytics can be helpful. Keddell cites two studies in which variations on analytics were used to detect caseworker bias. In one, the researchers could predict which workers were more likely to recommend removing children based on questionnaires assessing the caseworkers’ personal values.
In another, the decisions could be predicted by which income level was described in hypothetical scenarios. A study using similar methodology uncoveredracial bias.

So how about channeling all that energy now going into new ways to data-nuke the poor into something much more useful: algorithms to detect the racial and class biases among child welfare staff? Then we can protect children from biased decisions by “high risk” caseworkers.

Monday, May 9, 2016

Hey, look everybody! The magic orphanage is back!

Every few years media and politicians fall in love with another Magic Orphanage. News stories assure us that the Magic Orphanage is nothing like those mean old Victorian places. Now “house parents” supposedly work miracles with happy children rescued from their horrible parents.

60 Minutes spun that story about Maryville Academy near Chicago. Florida reporters flocked to SOS Children’s Village. And of course former House Speaker Newt Gingrich cited Boys Town when pushing the idea of orphanages for the children of parents whose only crime was poverty. (Give Gingrich credit for this much: He was honest about the purpose of orphanages.)

And now, behold! Marie Cohen offers a new candidate for Magic Orphanage: The Crossnore School. Apparently based on that most objective of sources, the school’s website, Cohen gushes about the place. Nineteen – count ‘em, 19! – different forms of therapy! An adventure playground! Almost all the kids graduate from high school! What could be better?

A family, that’s what. A “house parent” is not a parent. (For starters, they typically quit every year or two.) And “homelike,” a word that appears all over the website, is not home.

When people with no ax to grind look at orphanages, they come to very different conclusions. It takes three single-spaced pages just to list the studies describing the harm of institutionalization. Institutions that rebrand as “residential treatment centers” do no better. The younger the child, the worse the harm. Crossnore institutionalizes children as young as one year old.

As for all that therapy: Now that we’ve invented the automobile, it’s possible to bring children to therapists and therapists to children without actually making children live in an institution. It works for playgrounds, too.
The graduation rate probably is helped by the fact that Crossnore decides whom to take and whom to exclude. Crossnore also takes private placements, so not all of those graduates are foster children.
But the big advantage Crossnore has is, of course, money. Crossnore spends $66,265.06 per year per child – and that’s just the operating budget.
A Better Way to Spend $66,265

Now, let’s consider a case from another Cohen column: a 19-year-old who came forward to adopt his 3-year-old son. Cohen bemoans the fact that a judge placed the child with his father – who had never abused or neglected him – instead of letting foster parents adopt him. She notes that the father was unemployed and had not finished high school.

But what would happen if we provided the father with a new house, and $66,265 per year to help him finish high school, go to college, get a job and provide whatever other help he and his son needed?
Given what a mere $4,000 can do, that kind of help really could work magic.

But Cohen is horrified at the thought of this child being raised by his own father, and joyous at the prospect of children like him, and even younger, being raised in institutions.
The people working real magic are those who have realized institutions are a failure and are working to change their own institutions – including some that used to get the Magic Orphanage treatment in the media.

There’s one other problem with Magic Orphanages. Sometimes, things are pretty ugly behind the curtain.
§  Six years after 60 Minutes visited Maryville Academy, it was revealed to be a place of terror for many of the children confined there. The Chicago Sun-Times reported that “the place is often up for grabs, with staff struggling to handle suicide attempts, sex abuse, drug use, fights and vandalism…” In 2001, police were called to Maryville 909 times.

By 2004, Illinois had pulled all 270 state wards out of Maryville – something it could do because it had done such a good job of reducing needless foster care.
§  At SOS Children’s Village in Florida, between 1999 and 2001, 33 reports were filed with Florida’s child abuse hotline alleging abuse of children at the 50-bed facility; 21 were “substantiated” or “indicated.” During the same time period, 13 “house parents” and 14 “parent assistants” quit or were fired.

§  And in 2010, the state of Nebraska suspended admissions to two programs run by Boys Town amid allegations of misuse and overuse of “restraints” and medication.

Those are just the magic orphanages. When it comes to typical institutions, we’re not talking rotten apples, we’re talking rotten barrels.


Get the children who don’t need to be in any form of substitute care back into their own homes, and there will be plenty of room in good, safe, foster homes for the children who really need them. And no one will even think of warehousing more children in orphanages.

Sunday, May 8, 2016

Child abuse in Texas: Things you should not do in an emergency

Photo by James Cridland
The Dallas Morning News reported last week that: 
Tens of thousands of infants and children believed to be in imminent danger of abuse or neglect, even death, are not being seen promptly by state child abuse investigators — and thousands of them haven’t been checked on at all. …
Across Texas, on an average day, nearly 700 unseen children are classified as extreme cases — “Priority 1” in the agency’s terms …
But the most staggering numbers come from Harris County, where about 1,300 children who are supposed have already had a face-to-face contact with a caseworker haven’t been seen. The unvisited children account for about 22 percent of all cases referred to that office in March and April, The News’ analysis found. 
The reason all these children aren’t being seen?  Caseworkers are overloaded.

A Morning News editorial correctly calls this “an emergency.”

Here are some things you should not do in an emergency:

● Texas should not be investigating and taking away children from a Harris County family like this one.

● Texas should not be investigating and taking away children from a Harris County family like this one, either.

● Texas should not be taking away children at a rate we now know is far above the national average, even when rates of child poverty are factored in.

And Texas journalists might want to spend some time watching out to be sure that the new spending required to meet the terms of a federal court decision isn’t stolen from the few things being done in that state that actually work – like this, in Bexar County.

Then, maybe, caseworkers actually would have time to get to all those “Priority 1” children they’re not seeing now.

Wednesday, May 4, 2016

Child abuse: The not-really-all-that-shocking truth

Imagine you read an article that began like this:
“A reckoning is coming in gerontology. New studies show that when a group of 95-year-olds is followed for five years, a greater proportion die than when they are followed for only one.”
You’d probably think it was a story from The Onion. You’d probably think the same if you read a story saying that people followed for five years, versus those followed for only one, are more likely to catch cold, fall in love, get in an auto accident, take a dream vacation, or discover the presence of gambling at a certain café in Casablanca.
And yet we are supposed to be shocked – shocked! – when a study that follows children for five years finds more of what authorities define as abuse and neglect than studies focusing on a single year.
Aside from the reference to gerontology, I didn’t make up that line about “A reckoning is coming…”  It’s straight from the first words in Chronicle story about a California study released by the Children’s Data Network (CDN), which matched birth records to child abuse reports over a five-year period.
That story also quotes this from the CDN website:
“[One-year] estimates give the impression that only a small share of children are maltreated or placed in foster care, whereas cumulative estimates demonstrate the true severity of the risks and the resulting public health burden.”
But no one has ever suggested that one-year studies are meant to capture the full extent of the problem. Rather, looking at a single year allows us to return year after year to determine a trend.

It’s how we know that child abuse almost certainly is decreasing.

Nevertheless, the problem of child abuse is serious and real. And I agree with my fellow tax-and-spend liberals that we don’t spend enough to prevent it.
Unfortunately, the misuse of studies like this one to fuel the child abuse hype machine is likely to prompt two outcomes:
Funding diverted from the best options for curbing maltreatment, such as concrete help with housing and day care, and evidence-basedintensive family preservationservices.

Money pouring into bad options, such as investigating even more families – as the so-called Commission to Eliminate Child Abuse and Neglect Fatalities wants to do – and more foster care and institutions.


The definitional quagmire


CDN’s five-year study suffers from the same flaw as the one-year studies: a reliance on official definitions of maltreatment, definitions that make it easy to confuse poverty with “neglect.” And “neglect” constitutes the overwhelming majority of maltreatment reports.
California’s definition states that a child has suffered neglect if the child has suffered or is at risk of suffering harm or illness as a result of, among other things: “the failure or inability of the parent or guardian to adequately supervise or protect the child.”

Translation: Mom has to work and leaves the child home alone.

Or: “The willful or negligent failure of the parent or guardian to provide the child with adequate food, clothing, shelter, or medical treatment.”
Translation: The parents are poor.

That’s how you can wind up with cases like these from Los Angeles:
Gilbreania Wallace was taken from her grandmother when the pipes in their rented house burst, flooding the basement and making the home a health hazard. Instead of helping them find another place to live, the Department of Children and Family Services placed Gilbreania in foster care. She died there, allegedly killed by her foster mother. (DCFS, which would spend nothing to move the family, offered $5,000 for the funeral.)

Dakota Prince, age 5, and Nehemiah Prince, age 3, were taken from their mother because of what the DCFS deputy director at the time called “just an inability to provide adequate care.”  They were placed with a foster mother who could afford the best – including a Cadillac Escalade.
One day she forgot that she’d left the children in the Escalade in 100-degree heat where, a deputy district attorney said, “they cooked inside the car and died.” The foster mother was convicted of involuntary manslaughter.

Of course, a neglect definition this broad also can encompass sadistic acts of extreme and even fatal neglect. But since fatalities are, fortunately, a tiny percentage of all cases, the oft repeated fear mongering about how “more children die of neglect than abuse” tells us exactly nothing about typical neglect cases.

And the CDN study doesn’t just count cases where the allegation was true. When this study finds that 15 percent of children born in 2006 were reported for child maltreatment over five years, it means only that 15 percent of children were subject to someone – anyone – picking up a phone and expressing a suspicion that a child just might be maltreated under definitions as broad and vague as those above.
This also is why there is so much less than meets the eye to other study findings:
§  The not-so-shocking finding that a lot of children are re-reported. If a family is reported for being poor when the child is age one, there’s a good chance the family will still be poor when the child is four, and it should come as no surprise if the family is reported for being poor once again.
§  The not-so-shocking finding that very few children subject to reports actually received any “services.” This is partially because we do not, in fact, spend enough on helping poor families. It’s partly because the limited available funding tends to be scarfed up by a foster-care industrial complex that feeds off a steady supply of children placed in expensive foster homes, more expensive group homes and ridiculously expensive institutions.

And it’s partly because so many of those initial reports were, in fact, false reports. (For reasons discussed above, the fact that many children are reported again does not mean the initial report was valid.)

But there is one finding in CDN’s study that truly has not gotten enough attention. By the time a California child is just five years old, odds already are one in ten that he’s been subjected to the trauma of a child abuse investigation (based on the figures in the CDN study, minus the average percentage screened out). In poorer neighborhoods in Los Angeles, it’s above 15 percent. Now imagine what the chances are throughout his entire childhood.
In short, what this study really tells us is the extent to which child protective services has become a terrifying – or to use Baltimore City Department of Social Services Director Molly McGrath Tierney’s apt word – oppressive force in impoverished communities, especially communities of color – harming the overwhelming majority of the children subjected to their scrutiny.


The CalYOUTH study


My fellow columnist in this series, Sean Hughesalso feels the world should pay more attention to Chapin Hall’s CalYOUTH study. I’m glad to oblige.

Some people view this study as a child saver’s white middle-class rescue fantasy come true: evidence that children are taken only from horrific circumstances and consider themselves “lucky” to be in foster care.
About one-quarter of these foster youth clearly did suffer serious physical abuse at the hands of a caretaker. But other categories combine maltreatment that is clearly serious with conduct that may or may not be serious. Still other categories, such as “Youth missed school to care for family member or do chores” can once again be an indicator of poverty, not maltreatment.
And in reporting rates of sexual abuse, the study combines alleged abuse by a caretaker with abuse by anyone.
The study also depends on the recollections of the young people, which can cause errors in both directions. Some may not know how bad things really were at home when they were taken, sometimes a decade or more earlier. But others may be basing their statements on what they’ve been told by a succession of caseworkers.
Another column in The Chronicle of Social Change argues that because 57 percent of those surveyed said they were “lucky” to be in foster care, that means a majority of removals in California are justified.

There are several problems with this assumption:
§  Ten percent of the original random sample could not be interviewed because they had run away, were in jail or had returned home. I think they would be less likely to view placement as “lucky.”
§  The remaining young people were interviewed while still in foster care. I’m sure they were promised their individual answers would be confidential. Nevertheless, after the interview, they had to return to the foster home or group home and face the foster parents or staff. That’s an inherently intimidating prospect.
One need only contrast the CalYOUTH findings to a major study of young people questioned after they’dleft foster careand two studieofactual outcomes to see the problem.
§  Most important: If children really are removed from their homes only when it is absolutely, positively essential, shouldn’t the proportion considering themselves lucky be closer to 100 percent? If we’re really going to use perception as a sole guide to reality, then by that same logic this figure “proves” that 43 percent of California foster children never should have been taken away.

When it comes to the prevalence of child abuse and whether we’re spending enough to deal with it, the real question is one that Hughes and I were not asked, but I’ll answer anyway.
Question: To solve the serious and real problem of child abuse, do we need to spend more or do we need to spend smarter?


Answer: Yes.