What Happens When Predictive Analytics Enters the World of Child Protection?….How Do You Define a Gang Member?……The LAPD & the Guardian’s CountJune 2nd, 2015 by Celeste Fremon
Much has rightly been made of the unbearably tragic child deaths in Los Angeles and elsewhere in the state, at the hands of those who should have kept them safe, deaths like that of 8-year old Gabriel Fernandez. To refresh your memory, when paramedics showed up at Gabriel’s mother’s home in May 2013, they found the little boy with a fractured skull, three broken ribs, bruises and burns in too many places to count, and his mouth absent two of his teeth. BB pellets were embedded in his lungs and his groin.
Both LA County’s Department of Children Services and the LA County Sheriff’s Department had received complaints that Gabriel was being abused. But somehow nobody acted. And the two-agency non-action resulted in the torture and violent death of an eight-year-old.
Yet, there are other documented cases where DCFS seems to act too quickly, yanking kids out of less-than-ideal but non-dangerous homes and putting them through encounters with the foster care system that were, at best, traumatic and, at worst, deeply damaging.
So how does one tell the difference? Certainly, in some cases, it seems that a modicum of caring attention and common sense would have helped. But in others, the lines may not be so clearly drawn.
Some counties and states around the nation think they might have found at least part of the answer in the realm of what numbers geeks call predictive analytics.
Take for example, the case of Florida’s Department of Children & Families, which had nine child deaths in the state’s Hillsborough County area between 2009 and 2012. All of the kids were under three years old, and all but one were killed by either a parent or paramour.
At the time, the region’s child protective services were contracted out, at a cost of $65.5 million a year, to private youth services agency called Hillsborough Kids.
Florida dumped Hillsborough Kids, bumped up the budget for social workers and, perhaps most significantly, Florida officials contracted to use a new decision-making tool to help the agency prioritize calls of suspected child abuse. It is called Rapid Safety Feedback.
Darian Woods, writing for the Chronicle of Social Change, takes a look at where predictive analytics has entered the world of child protection, who is involved, and what that entry could mean in terms of the future safety of kids.
Here’s a clip:
So in 2012, the department made changes. It commissioned a comprehensive analysis of the data behind the child deaths that were concentrated in Hillsborough County. Hillsborough Kids lost out on the $65.5 million contract and went into liquidation. A private youth services agency, Eckerd Youth Alternatives, was selected by the department to take care of approximately 2,900 abused children in Hillsborough County. The next year, Florida Governor Rick Scott boosted funding for new social workers. Perhaps most radically, a new decision-making tool called Rapid Safety Feedback was introduced in the county.
Rapid Safety Feedback uses — in the parlance of big data crunchers and, increasingly, social scientists — predictive analytics to prioritize calls of suspected child abuse.
Predictive analytics in child protective services means assigning suspected abuse cases to different risk levels based on characteristics that have been found to be linked with child abuse. These risk levels can automatically revise as administrative data is updated. Administrative data may be as simple as school reports or could delve deeper into other information that the state holds: the parents’ welfare checks, new criminal offenses or changing marital status.
Combining predictive analytics with more investigators seems to be producing results in Hillsborough County. According to Eckerd, who also holds contracts in Pasco and Pinellas counties, since it took over the contract in 2012, the quality of reviews has improved 30 percent. There is a significant increase in completed documentation by caseworkers. There have also been zero child homicides in the county since the handover.
LA County is one of the counties that is looking hard at the use of predictive analytics, but they are less positive that big data can solve the problem.
HUMAN JUDGEMENT VERSUS THE MACHINE: CAN SAVVY PEOPLE KEEP KIDS SAFER THAN PREDICTIVE ANALYTICS? OR IS BIG DATA THE ANSWER?
Holden Slattery, also writing for the Chronicle of Social Change, looks further into what LA County is doing as it “struggles to strike the right balance between human judgement and increasingly sophisticated predictive tools when determining the risk that a child will be abused.”
Here’s how Slattery’s story opens:
On weekdays, calls to Los Angeles County’s child abuse hotline reach their peak between 2 p.m. and 6 p.m.—right after school. On average, 70 to 80 calls about child maltreatment in Los Angeles County reach the hotline per hour during that span, according to the Department of Children and Family Services (DCFS), the agency charged with responding to alleged abuse.
There are about 85 social workers manning the phones at any given time. They ask callers to explain how child abuse or neglect took place.
The number of calls made to the largest child welfare system in the United States creeps up each year, said Carlos Torres, an assistant regional manager for the DCFS hotline. In 2014, the hotline received 220,000 calls, he said.
After listening and marking down answers on a computer program, the social workers decide whether a situation meets the criteria for an in-person response. They also decide whether DCFS should respond by the end of their current shift, within 24 hours, or within five days, Torres said.
These decisions, based on small bits of information shared by a caller, determine where DCFS directs its limited human resources. DCFS responds with an in-person investigation to 35 percent of the calls, Torres said. In these cases, a social worker drives to the home, interviews the family, gathers information, and enters his or her findings into a web-based decision-making tool, which, like a questionnaire that an insurance company gives to prospective clients, estimates risk; in this case, risk that a child will be abused.
When everything goes right, DCFS can save a child from harm. When something goes wrong, the result can be heartbreaking. A 2011 report on recurring systemic issues that led to child deaths in Los Angeles County put the onus largely on flawed investigations and problems with the decision-making tool employed. In the search for solutions, public officials have looked toward new technologies, such as analytics software used primarily by private companies, to see if that can keep more children out of harm’s way. As public officials make these kinds of inquiries, in Los Angeles County and across the globe, they confront the conundrum of human judgement versus machine. Some say technological advances hold the answers, while others say that only savvy people are up to the task.
Slattery notes that a number of experts cite research that suggests all this predictive analytics isn’t particularly effective when it comes to assessing if a kid is safe or not.
In any case, read on.
IS IT TIME TO REFORM CALIFORNIA’S “STEP ACT?”
One night in January 1988, rival gang members were shooting each other on the streets of Westwood and mistakenly hit and killed a young woman named Karen Toshiba.
The murder of Karen Toshiba became a flashpoint, as such tragic deaths often do, and 1988 became the year the so-called war on gangs was declared in Los Angeles and, in Sacramento, the state legislature passed the Street Terrorism Enforcement and Protection Act (STEP Act), Statute 186.22 of the penal code.
Among its other functions, the the STEP Act imposed greater punishment for crimes committed “for the benefit” of a criminal street gang. In the beginning, the sentencing “enhancements” were no more than a few years. But it 2000, crimes that were “serious” or “violent,” as defined by the California Penal Code, could be enhanced by five or ten or, in certain cases, a life sentence.
The STEP Act can be brought to bear even when a young man or woman is at the periphery of a gang, with a relationship that has more to do with where he or she lives, than any kind of actively committed or formalized association.
It has resulted in multi-decade sentences for juveniles tried as adults as a consequence of their proximity to violent acts in which they did not participate, even in cases when no one was injured.
If a so-called gang expert can successfully label a defendant as a gang member, even if he or she is not, then the enhancement can kick in, and conviction is also much more likely.
In a story by Daniel Alarcón in this week’s New York Times Magazine called “How Do You Define a Gang Member?” Alarcón
describes a case that shows the STEP Act in action.
The story has to do with a case in Modesto, California, where the primary gangs are variation on the theme of Norteño, or northerners, or Sureños—southerners.
Here’s a clip:
On a rainy day last December, in a courtroom in downtown Modesto, Calif., a 24-year-old white man named Jesse Sebourn, along with five co-defendants, sat accused of second-degree murder. The victim, Erick Gomez, was only 20 when he was shot to death. He was a reputed Norteño gang member who had lived just a few minutes’ drive from the working-class Modesto neighborhood where Sebourn was raised. The police estimate that there are as many as 10,000 gang members in Stanislaus County, where Modesto is, most either Norteños and Sureños, two of California’s most notorious Latino street gangs. The feud between them often turns deadly, and according to Thomas Brennan, the district attorney, this was one such instance: Sebourn and his co-defendants were Sureño gang members hunting for rivals on Valentine’s Day in 2013, when they found Gomez, out on a walk with his girlfriend.
Brennan was not saying that Sebourn had fired the gun; in fact, the accused shooter, Giovanni Barocio, had evaded arrest and is believed to be in Mexico, while witnesses and time-stamped 911 calls made it difficult to believe Sebourn had even been present at the scene when Gomez was killed. But according to the prosecution, Sebourn had set the entire chain of events in motion a few hours before the shooting, when he and two of his co-defendants tagged a mural eulogizing dead Norteños in an alley behind the building where Gomez lived. Sebourn and the others were caught in the act and beaten by Norteños, though they got away with little more than scrapes and bruises. But the prosecution argued that spray-painting over a rival’s mural was an aggressive act intended to incite violence — the equivalent of firing a shot. By this interpretation of events, the afternoon scuffle led directly to that evening’s murder: tagging, fisticuffs and finally, hours later, homicidal retaliation, each escalation following logically and inevitably from the previous. “Ask yourself,” Brennan said to the jury in his opening statement, “what are the natural and probable consequences of a gang fight?”
But this time the defense has a gang expert of its own, a former gang member turned PhD named Jesse De La Cruz…
In any case, read on.
THE LAPD HAS THE MOST POLICE KILLINGS IN 2015 OF ANY LAW ENFORCEMENT AGENCY IN THE NATION, SAYS THE GUARDIAN, WHICH HAS DECIDED TO COUNT
The Guardian newspaper has launched a project it is calling The Counted, the purpose of which is to count people killed by police in the U.S. in 2015.
It’s an interactive project, which you can find here.
Over at KPCC, Aaron Mendelson writes that, according to the Guardian’s database, the Los Angeles Police Department has killed more people (10), than any other law enforcement agency in the United States this year, that’s twice as many as the four law enforcement agencies, one of which is the LASD, that are in second place.