Monday, January 24, 2022

The Los Angeles Times is getting child welfare wrong again – and that’s bad news for Los Angeles children

Garrett Therolf was among those speaking at a Los Angeles Times
"Ask a Reporter" event on January 20. Questions were screened in advance.

As usual it's because of Garrett Therolf. The former Times reporter and still their go-to guy for child welfare stories continues to soft-peddle racial bias.  Now he’s campaigning for the use of computerized racial profiling – a.k.a. predictive analytics.  He’s even effectively blaming readers for his own failings as a journalist. 

First of two parts Read part two here.

Remember Garrett Therolf?  He’s a former Los Angeles Times reporter and current Times contributor via his current job with the University of California, Berkeley Investigative Reporting Program.  While writing story after story attacking efforts to keep families together he was accused – by other journalists - of repeated misrepresentations and, on one occasion, accused of making up a quote

 ●LA Weekly noted how the Times quietly backed off claims in a Therolf story about child abuse deaths. 

● The respected online news site WitnessLA reported: 

[A]ccording to a number of sources and experts, many of Therolf’s supporting facts range from very fuzzy to downright inaccurate—and seem to be immune to correction … 

Much is at stake when the state’s largest newspaper starts using some high profile tragedies to push—without adequate factual basis and no visible thoughtful consideration—for a change in policy that would result in more kids snatched unnecessarily from non-abusive families into the trauma that is foster care. 

● The founder of the child welfare trade journal The Imprint, Daniel Heimpel was more blunt, blasting “The Times’ myopic, misleading and reckless reporting …” 

Therolf’s specialty is taking child abuse deaths, tragedies that are as rare as they are horrific, and doing two things: Covering them almost to the exclusion of everything else, leaving the impression that the only error the system makes is leaving children in dangerous homes, and drawing broad, sweeping conclusions from these tragedies.  That contributes to foster-care panics, sharp, sudden increases in children needlessly torn from their homes. 

That not only hurts the children needlessly removed, it also overloads the system, making the next tragedy that much more likely.  Of course this doesn't mean Therolf wants his reporting to do harm; I'm sure he wants his reporting to save lives.  But to do this reporting he must immerse himself in the details of some of the most horrific tragedies imaginable.  As he has acknowledged, that can take a personal toll.  And, as with other reporters who have done the same kinds of stories, it seems to have blinded Therolf to the real-world consequences of an approach that omits crucial context.


When Therolf’s reporting came under increasing criticism, the Times issued a response that boiled down to: If you challenge our coverage in any way you want children left in dangerous homes! (See also WitnessLA’s response to that here and here.

Garrett Therolf’s journalism certainly isn’t the only reason Los Angeles tears apart families at the second-highest rate among America’s largest cities and their surrounding counties, I think it plays an important part. 

Therolf has a particular problem with the idea that there is systemic racial bias in child welfare – boosting the most extreme white scholars who lead child welfare’s de facto “caucus of denial” while ignoring the credentials and demeaning the work of Black scholars.  As we noted in this earlier post about one such story, in Garrett Therolf’s world white people “marshal data.”  Black people just want to rely on anecdote and “folkways.” 

Hyping “predictive analytics” 

So it should be no surprise that Therolf is pushing that dangerous fad in child welfare predictive analytics, in which an algorithm dredges up data on families who were the subjects of child abuse reports and tells the family policing agency if the case is “high risk.”  Therolf regurgitates all the favorite talking points of Emily Putnam-Hornstein, America’s foremost evangelist for what is actually computerized racial profiling – and someone whose own extremism and penchant for deriding the work of Black people is documented here. 

Perhaps Therolf fears that people are finally taking seriously the existence of systemic racial bias in child welfare.  That would explain two odd events at the Times on January 20: A so-called “news analysis” story that pushes predictive analytics as the silver bullet for child welfare – the only idea greeted with approval in a story headlined “How to fix L.A. County DCFS” -- and an “Ask the Reporters” video presentation in which almost no one outside the Times itself got to ask anything of Therolf, his current reporting partner, Matt Hamilton, and Holly Mitchell, who chairs the Los Angeles County Board of Supervisors. 

Normally, there is a news peg for stuff like this – some big event in the news prompting a newspaper to offer this kind of additional “context.”  But the only thing new in Los Angeles child welfare is the departure of the head of the County’s family policing agency, the Department of Children and Family Services – a thoroughly routine event that happens on average roughly once every two or three years – and the emergence of a powerful new force for change, the Reimagine Child Safety Coalition.  They’ve issued a series of demands, directed at the Board of Supervisors, which would, at long last, bring a measure of racial justice to Los Angeles child welfare.  In this, they march alongside DCFS-Give Us Back Our Children, which has been active for well over a decade.  I am proud to work with both groups. 

In part two of this post, tomorrow, I’ll go paragraph-by-paragraph through the “news analysis” story, since it is a splendid example of Therolf in full spin mode.  But for now, let’s go to the videotape.  

During the carefully stage-managed event, almost all the questions came from Times columnist  Anita Chabria. Audience questions had to be submitted well in advance for screening.  One of those questioners asked: 

Are you concerned that reforms are driven by the worst failures of child protective services, and in focusing on children who died, officials and the public largely forget the trauma and harm experienced by a huge number of children removed from their homes and then removed from repeated subsequent placements? 

Hamilton replied: 

 I don’t think it’s an if/or proposition. You can focus on these complex cases that have a lot of very serious allegations but still also look at the broad number of cases that come in, I don’t think reform is being only driven by these type of cases. 

The problem with that is two-fold: First, for decades what passes for “reform” has been
driven by the horror story cases and almost nothing else – that’s why it’s failed. The one reform that tried to curb needless removal, a waiver to allow more flexible child welfare funding, was a prime subject of misleading attacks by Therolf. 

And second, while one can look at cases that illustrate needless removal, with only token exceptions the Los Angeles Times hasn’t actually done that, something discussed in more detail below. 

The most misleading statistic in child welfare 

Then it was Therolf’s turn.  The first thing he did was remind us of the horror stories – in a common, but misleading way. In fact, he used what may be the most misleading statistic in child welfare.  He said that four to eight children die of abuse in America every day and half are in some way known-to-the-system. 

That is not wrong, but it lacks crucial context.  First, “known” can mean anything from one call to a hotline years before, to a casefile with more “red flags” than a Soviet May Day parade; which are, of course, the ones on which Therolf focuses. 

But more important, the phrasing – half of children who die were known-to-the-system leaves a visceral impression of something very different: That half of children known-to-the-system die  -- and therefore agencies must be bending over backwards to leave children in dangerous homes. 

To really understand the meaning of half of child abuse deaths involve children known-to-the system, you have to understand how many children are “known to the system.”  Any story that uses the statistic should also include the context. (Shall we take bets on whether someone at the L.A. Times or some other journalist-on-the-defensive will misrepresent the above and say: “Oh, you don’t want us to report how many deaths were known to the system”?)  So here’s the context: 

Nationwide, more than seven million children become “known to the system” every year.  Using Therolf’s high-end estimate, 1,460 of them die.  That means every year 99.998% of children “known to the system” do not die of child abuse and neglect. 

It looks a bit like this:

 


As soon as you point that out, however, it plays into the hands of those, like Therolf and others at the Times, who say this means we must be minimizing the problem and don’t really care if children are hurt. 

And sure enough, that’s what Therolf did, declaring: 

Sometimes reporters are criticized for covering child abuse fatalities: Oh, you’re covering an unusual event.  And while these events are rare, they are happening every day with regularity, and it’s a place where our democracy has an opportunity to make an improvement in an area everybody agrees has not fulfilled its promise. 

No one says journalists shouldn’t cover child abuse fatalities – it’s just Therolf’s way of evading responsibility for the failings in his journalism.  Rather, we’ve said Therolf, and others, should do it better – so their coverage might bring us closer to the only acceptable goal for child abuse fatalities: zero, rather than simply repeating the very cycle of failure Therolf acknowledges. And we’ve said they should cover all the other system failings, not instead of the horror stories, but in addition. 


Therolf’s critics are simply doing his job for him.  We’re explaining the problem – and why foster-care panics make it worse -- and suggesting real solutions.  Child abuse deaths are needles in a haystack, and the foster-care panic typically engendered by journalism, Therolf-style, only makes the haystack bigger. 

And then, incredibly, Therolf claims he already covers wrongful removal – and suggests that if readers haven’t noticed it must be their fault.  He gives a lovely little speech that bears no resemblance to his actual reporting. It begins like this: 

On the other end of the spectrum, we also know that far too many families are caught up in child welfare systems that don’t belong there.  

He goes on to label as “horrifying” the fact that “half of all Black children are investigated over the course of their childhood as possible victims of abuse and neglect.” (By the way, in Los Angeles it’s probably 58%.

He talks about the documentary King Richard, in which, he says, Venus and Serena Williams describe the trauma of a false allegation of child abuse against their father, and how that was “one of the most searing experiences of their lives.”   And he adds: “It happens with extraordinary regularity to families that have done nothing at all to warrant that type of intervention.” 

But then he has the gall to suggest, without quite saying it, that such stories are prominent in the Times and a regular part of his reporting. 

We’ve certainly covered it at the Los Angeles Times at various periods of my career. …  I did a story about a Black mother named Monique Baker who struggled for over a year to regain her children on the basis of allegations that proved to be false.  We have done those stories. 

Yeah, once.  Oh, wait, that’s unfair.  There was also that great story he did covering the issue when it came up at a Board of Supervisors meeting – in 2002.  Odds are, over the course of more than two decades, there have been one or two more. 

But, on those extremely rare occasions when the cries of children wrongfully removed make
it into the Los Angeles Times, those cries are drowned out by the tsunami of coverage of the horror stories – coverage that almost always implicitly, and sometimes explicitly, blames efforts to keep families together.  Times subscribers can easily confirm this with a check of the newspaper’s online archive. 

Or consider the promotion for the online event itself.  The headline reads “Children keep dying. Why can’t L.A. County protect them?” If Therolf’s words on the video were reflected in his work at the Times the headline would have been more like “Families torn apart but children still die. Why can’t L.A. County stop the tragedies?”  The promotional material includes a link to that one story Therolf cited about wrongful removal – and at least 27 stories about the deaths of children “known to the system.” 

Flunking the Bagdikian test 

Therolf’s claims in the video are, in fact, the classic approach news organizations take when criticized for ignoring perspectives they don’t like.  Former Washington Post editor Ben Bagdikian explained it decades ago in his seminal work of media criticism The Media Monopoly.   

One of Bagdikian's important observations is how newspaper editors who have strong feelings about an issue deal with the side of that story they don't like: 

They cover it. Prominently. Once. 

That's just enough for the editors to say to critics "Oh, we covered that. We even put it on the front page" – when they know full well that anything covered only once has almost no impact. It's what's repeated over and over again that leaves an impression. 

Columbia Journalism Review made much the same point last week, in a column about coverage of the humanitarian crisis in Afghanistan.  What makes the difference, CJR argued is 

[W]hich stories are not just worthy of an article, but of real focus and consistent amplification. [Emphasis in original.] 

Even that one prominent story Therolf points to with pride stumbled when dealing with the issue of racial bias.  In fact, that part of the story made Therolf so uncomfortable that he accompanied the story with a sidebar  – the story mentioned at the beginning of this post, in which he stacks the deck to support the claim that any disparity in entries into foster care is due to “past” discrimination making Black people bad parents. 

And then comes the coup de grace: Therolf says to readers, in effect: If you haven’t noticed the side of the story I almost never cover, it must be your fault!  Or as Therolf puts it: 

Unfortunately, [stories about wrongful removal] don’t stick in folks’ memory quite as much as the fatalities do, but they are there for the folks who are interested in looking at them. 

Good luck finding them. 

Read part two: Garrett Therolf and The Magic Algorithm!