Tuesday, October 20, 2009

Child abuse report: The children who don’t matter to “Every Child Matters”


(A large part of the post below involves countering bad numbers with good numbers. Unless otherwise noted, citations for the data in the post below can be found in NCCPR "Other Issue Paper 2," and Paper 3. These Issue Papers originally were written to refute misinformation commonly spread in the 1980s and 1990s, but largely discredited by 2000, so they are not as prominent on our website as the Issue Papers on family preservation, nor were they updated as often – until now. Sadly, Every Child Matters has chosen to revive these misleading claims.)
A group calling itself Every Child Matters (ECM) has scheduled what it calls a Rally for Change on the West Lawn of the U.S. Capitol today (Oct. 21, 2009). But it's really just a Rally for the Same Lousy System Only Bigger.
Like most people in child welfare, those at ECM have good intentions. But the people at ECM are serial statistics abusers whose rights to their pocket calculators should be terminated. Their stock-in-trade is hype and hysteria, and many of their recommendations would only endanger the children they want to protect.

The problem of child abuse is serious and real, but ECM's solutions are phony, and they're using bad data to promote those phony solutions.

ECM proposes taking $3 billion to $5 billion that could be used for real prevention and family preservation and using it for a huge expansion of the agencies that investigate families and take away their children. Not only will this traumatize even more children who never were abused, it will further divert workers' attention from finding the children in real danger who really must be taken from their homes.

At a time when 77 percent of child abuse allegations already turn out to be false, and many more are cases in which poverty is confused with "neglect," ECM wants to burden child welfare agencies with more false leads by starting still another campaign to urge every American to turn in her or his neighbor based on the slightest suspicion of maltreatment. In at least one American city half of all children have endured the trauma of a child abuse investigation by age 10. Does ECM really think that percentage is too low?

● Though it is not a formal recommendation, the report also appears to call for ending all screening of calls to child abuse hotlines – meaning that even the most absurd, frivolous or vindictive allegations would have to be investigated.

To justify this take-the-child-and-run agenda ECM grossly misuses data, and sometimes uses data that are incomplete and out-of-date even when better data are available. ECM also attempting to compare fatality rates among the states, even as ECM itself effectively acknowledges there is no reliable way to do so.

And ECM shows a sickening willingness to exploit tragedy. They include among their list of children who died of abuse at least one child who actually was taken from a loving birth mother only to be killed by her foster mother – a little detail ECM neglects to mention to readers.

Every Child Matters sent an endless stream of e-mails urging people to come to the rally and support their agenda – while keeping the actual agenda, and a report that goes with it, secret until less than 48 hours before the rally. That kind of arrogance has been par-for-the-course for America's "child savers" since they proudly gave themselves that name in the 19th Century.

And the people at ECM are right out of that 19th Century tradition. They are good people who really want to help children. But along with that comes an ends-justify-the-means mentality, best seen in a penchant for "advocacy numbers" – misleading statistics that don't stand up to serious scrutiny. (After all, what's a little exaggeration if it "raises awareness," right?)

The group is run by Michael Petit, a former commissioner of the Maine Department of Human Services. Petit founded ECM right after leaving the Child Welfare League of America, the trade association for public and private child welfare agencies, including the ones that couldn't survive without a steady supply of foster children. CWLA is one of the biggest obstacles to real reform in American child welfare, and Petit brought the CLWA mindset with him to ECM.


Instead of offering positive, constructive 21st Century solutions to the serious and real problem of child abuse – the kinds of solutions NCCPR offers in its publications Twelve Ways to do Child Welfare Right and Civil Liberties Without Exception, ECM calls for a $3 billion to $5 billion expansion of the agencies that tear children from their parents.

Of course they also mumble the usual platitudes about "prevention" – but mostly prevention that, while sometimes genuinely useful, such as voluntary home visiting, does nothing to curb coercive intervention into families. There is not a word about family preservation which both keeps families together and targets families about to lose their children to foster care. And they make no specific recommendation to spend a particular dollar amount on prevention. Similarly they acknowledge that poverty is bad and say we should do something about it. But they propose no dollar figure. In fact, their proposal amounts to stealing $3 billion to $5 billion that might go to prevention, family preservation and help to ameliorate poverty and spending it instead on investigating families and taking away children.

In its promotional material, ECM says their report shows "that the current child protection network is stretched too thin … the sheer magnitude of child maltreatment exceeds the capacity of the system to respond to every child needing help." Translation: spend a lot more money doing just what we do now.

Actually, it is the hype and hysteria spread for decades by groups like ECM that have stretched the system too thin – diverting it from its mission of finding children in real danger into wasting its resources on huge numbers of false allegations, trivial cases and cases in which family poverty is confused with "neglect."

Many states do indeed need to spend more on child welfare. But all states need to spend smarter. ECM calls for expanding what amounts to dumb spending. And ECM is on record as opposing allowing states to spend smarter. Thus, ECM opposes letting every state get the kind of waiver from federal funding restrictions that is allowing Florida to transform its child welfare system.

The net of voluntary help to families should be cast wide. But both because is traumatizes children who never were harmed and diverts resources from finding children in real danger, the net of coercive intervention into families should be narrow.


Here are some facts you won't hear at ECM's rally or read in ECM's report:

ECM says the "real" number of child abuse deaths may be 50 percent higher than the official figure. But even if you go much further and triple the official estimate, in any given year, 99.993 percent of American children will not die of child abuse.*

In any given year, 98.9 percent of American children will not be abused or neglected in any way – and that's true even when one counts all those cases in which what child protective services agencies call "neglect" really is poverty.**

The reason you almost never see those numbers is because groups like ECM have successfully intimidated critics away from mentioning them, using a kind of "bait and switch" technique. The bait: First they use inflated, phony numbers to lure us into their tent to sell us snake oil solutions. Then, if anyone tries to put the numbers into context comes the switch, as they say, in effect, "How dare you quibble about numbers when children are dying? If even one child dies of abuse it's one too many."

In one sense they are right, in another, tragically wrong. They are right in the sense that the problem of child abuse is not minor. The United States is a very big place, even a small percentage is a big number. And yes, even one child's life lost to the sadism or brutality of a parent is one too many. The percentage of children thrown into foster care also is low when compared to the general population (it can be a lot higher, however, in poor minority communities) – but that hasn't stopped NCCPR from working to make it lower.

But they are wrong in dismissing the importance of getting the numbers right. The numbers are significant not for what they say about the importance of the problem, but for what they say about how to solve it. The problem of child abuse is serious and real. It's ECM's solutions that are phony. Using bad numbers to promote phony solutions only makes it more likely that the real numbers, whatever they are, will get worse.

The fact that the percentage of children who face child maltreatment is, in fact, quite low, and the horror story cases that make headlines – or are ripped from the headlines for Law & Order scripts - is tinier still, has profound implications for how we try to reduce the number still further.

The horror story cases are needles in a haystack. Real solutions require finding more precise ways to detect the needless. Instead, we keep trying to vacuum up the entire haystack. The net of coercive intervention is made ever wider, with new categories of mandated reporters, broader definitions of maltreatment and constant exhortations to turn in our neighbors at the slightest suspicion of maltreatment.

As a result, we get that other big-number-small-percentage: about 300,000 children are torn from their families every year, and a lot of those removals are unnecessary. And we get a much larger number: Every year, at least 2.4 million children, are traumatized by investigations of false allegations of maltreatment.


The people at ECM are masters of weasel-wording, and that's what they use to get around those 2.4 million inconvenient facts. So in its report, ECM says that "many" reports initially labeled false will turn out to be true when the same family is reported again. But "many" can mean anything – or nothing. And ECM offers no support for that claim – the endnote leads only to this government statistics home page, with no indication of how ECM came up with this claim, or even where to look. Furthermore, when multiple reports do lead to substantiation, that may be only because there were multiple reports - so CPS workers may assume they must be true.

The one serious study I know of to examine this issue found that caseworkers are two to six times more likely to wrongly substantiate a case than to wrongly declare one to be unfounded. So if anything, the official number of false reports understates the problem.

And, because of the racial bias that permeates child welfare, unlike actual child abuse, which is no greater in minority communities, the chances of being traumatized by a false allegation or needless foster care are much higher if you are a minority, particularly if you are Black or Native American. Indeed, the Cleveland Plain Dealer reported in 2005 on a study revealing that in Cleveland, nearly half of all children were forced to endure a child abuse investigation by the time they were ten years old.
Does ECM really think that's not enough?

ECM's statistics abuse doesn't stop there. In addition to overstating the percentage of cases that workers substantiate (it's under 23 percent, not 30 percent as the report claims) ECM calls these cases "confirmed." Not true. No judge or jury reviews such decisions; it's just a caseworker checking a box on a form. And in many states the worker is supposed to "substantiate" the case even when there is more evidence of innocence.

But ECM's mastery of evasion is most apparent when they say "there is no evidence which suggests that intentionally false reports alleging maltreatment are a serious issue." The weasel word is "intentionally." Whether a child is subjected to a traumatic investigation, a stripsearch and separation from everyone she knows and loves because of a malicious report or because of a well-meaning error by someone who listened to one of those endless exhortations to report anything and everything isn't likely to matter much to that child. (And of course, even if the number of intentionally false reports is low, why don't the people at ECM believe even one is one too many in this case as well? Why do they simply dismiss the enormous suffering of children victimized by repeated investigations of malicious false allegations as not "a serious issue." You may be sure it is serious to the children affected.

Bottom line: No matter how much ECM tries to obscure the issue, the fact remains that, malicious or not, at least 77 percent of reports are false.

Furthermore, of the "substantiated" reports, the overwhelming majority are neglect – which often means poverty, since typical state laws define neglect as lack of adequate food, clothing shelter and supervision – a perfect definition of poverty. (For details and full citations, See NCCPR Issue Papers 5, 6 and 7.)

Indeed, out of every 100 children investigated as possible victims of abuse, at least 77 simply weren't - the report was false. Fourteen were "substantiated" victims of neglect, and 7 were victims of either sexual abuse or any form of physical abuse, from the most minor to the most severe. (Three of those eight may have been victims of both and/or other forms of maltreatment as well.) One was a victim of psychological maltreatment. One more falls into a category listed as "other."

All this is a huge tragedy, and not just for the children subjected to needless investigations and foster care. The time wasted on such cases only diverts more and more resources from finding children in real danger. All that time trying to vacuum up the haystack only leads to overlooking more needles.

And year after year, groups like ECM keep making it worse. Now, they're actually proposing what amounts to still another campaign to get everyone to turn in their neighbor based on any vague suspicion of maltreatment. (If you want to see how vague, just check out the lists of "symptoms" and "signs" of maltreatment that many organizations publish and that routinely turn up as "info boxes" in news stories. Based on these lists, there is hardly a child in America that couldn't be suspected of being maltreated at some point in her or his life.)


But wait, there's more.

In its new report, ECM purports to compare rates of child abuse deaths among the states – implying that the states it likes, big-spending and usually northern, are safer than the states it doesn't like, generally smaller-spending and southern. (This claim is made right after admitting that "… a clear correlation has not been established on how much states spend on child protective services and their child abuse and neglect death rates…")

But even were such a correlation to be found, it is, in fact, impossible to compare child abuse fatalities among states because of what has aptly been described as "differing definitions of child abuse and neglect among the states, varying data collection methodologies and inconsistent record-keeping across the country due to a lack of enforceable standards."

And who described it that way? ECM itself in a letter promoting today's event.

Of course, they noted these flaws only to suggest that fatalities allegedly are being undercounted. ECM never explains how, despite these very same failings, it's supposedly possible to compare states accurately. And in at least one case, the same failings have led to overcounting. ECM complains that deaths are undercounted by states that are prone to rule a child's drowning as an accident. But ECM never mentions that Florida's supposed high rate of fatalities only occurred after enormous pressure by the chair of that state's child death review team to take an extremist view and label almost every drowning as neglect. (For details, see a previous post to this Blog, "Drowning in Misinformation.")

At the very start of the new report, ECM uses the same trick used by "child savers" since the 19th Century – they start with horror stories about child abuse fatalities and immediately jump to a classic scare number – the fact that there are millions of "reports" of child abuse every year. We are meant to be left with the gut feeling that all those reports are just like the horror-stories. But, as noted above, the overwhelming majority of those reports are false and the overwhelming majority of the rest are nothing like the horror story cases.

This misrepresentation is reinforced with photos of one dead child for every state – but even there, ECM misleads. For Maine they use a photo of Logan Marr – without bothering to mention that Logan Marr was taken from a loving mother because family poverty was confused with neglect. She was placed in the foster home of a caseworker for the state child welfare agency. The caseworker/foster parent killed Logan. Since then, the Maine system has dramatically improved – by moving in the opposite direction from that pushed by some of its prior leaders.

Indeed, one of those prior leaders, Michael Petit, seems to have a lot of trouble getting things right about Maine.


Like other ECM reports, this one also suggests that if your state spends a lot on its child welfare system it must be a good system, and if it doesn't, the state, and by implication, the people of that state, don't care if children are abused. The report compares what is says are the five highest-spending and five lowest spending states and implies that the high-spending states must be investing in training and services.

Now, I'm a tax-and-spend liberal and proud of it. I believe strongly in spending more on child welfare. But you also have to spend smarter. ECM's comparison draws no distinction between states that spend a lot wisely and states that spend a lot stupidly. And that makes it a stupid comparison. Worse, ECM used data that are older and less complete than the best data available.

So of the ten states listed, it is one of the supposedly low-spending states, Maine, that has one of the nation's best systems. Indeed, this year Maine's child welfare transformation was a finalist for a prestigious Innovations in American Government award from Harvard (because the current leadership is doing a far better job running the system than Michael Petit ever did.) In contrast, none of the high-spending states is exceptionally good – and a couple of them, Alaska and Rhode Island, are real stinkers, as are the systems in seventh-ranked Iowa, 12th ranked Colorado and 13th ranked Nebraska.

That's because of the great paradox of child welfare: The worse the intervention is for the child, the more it costs. So of course, states like Alaska, Rhode Island, Iowa, Colorado and Nebraska, which tear apart families at among the highest rates in the nation, may spend more than states like Maine that may spend a little less, but get better results by wisely investing limited resources in safe, proven programs to keep families together. Indeed, Maine has had great success reducing the most expensive, least effective form of "care" of all – residential treatment.

And in any event, ECM got the figures for Maine all wrong. That's because they used data from 2004 and used a secondary source – the Child Welfare League of America – to obtain it. In fact, data are available, from Child Trends, for 2006. Those data show more than three times as much spending in Maine. (NCCPR can supply these data and full sourcing on request.) Of course, spending in Maine didn't really triple; rather, the older data were incomplete – but because ECM didn't track its 2004 data to their original source, pages 47 and 48 of this report from the Urban Institute, ECM never figured that out. So they the wound up offering up inaccurate, out-of-date data to make an ideological point.

And finally, though I am a lifelong liberal, I find it offensive to see that, in another ECM publication, Geography Matters, ECM, which says it is non-partisan, repeatedly presents what it considers to be the "good" states in blue, and the "bad" states in red.

Even the one recommendation in today's report that might do some good, a call for more transparency in child death cases, does not go far enough. We need a rebuttable presumption that all court hearings and nearly all records in all child maltreatment cases are open. Otherwise you get a look at the system that is like a reflection in a funhouse mirror. If the only time there is any openness is when a child dies, people think that's the only error child welfare agencies make. All those cases of wrongful removal, all those children's lives destroyed by needless foster care, remain secret. And apparently, that's the way ECM likes it.

In short, ECM isn't living up to its name.

--The children traumatized by false allegations of maltreatment and forced to endure stripsearches and other indignities don't matter to Every Child Matters.

--The children torn from everyone loving and familiar and thrown needlessly into foster care don't matter to Every Child Matters.

--The children moved from foster home to foster home emerging years later unable to love or trust anyone don't matter to Every Child Matters.

ECM should stick to what it knows: issues like education and health insurance and after school care, and stop promoting a child welfare agenda that will leave us with nothing but the same lousy system only bigger.

*This figure is calculated by taking ECM's figure for child abuse fatalities in 2007, tripling it, and then dividing it by a Census Bureau estimate of the total population of Americans under age 18
* This figure is calculated by taking the total estimated number of children for whom allegations of maltreatment were "substantiated" as published in note 8 on this page of this federal report, and dividing by the total population of Americans under age 18.