Sam Sinyangwe ‘fireside chat’


We’re really excited that Sam Sinyangwe is here today to talk with you I was at his last night and it was
incredibly enlightening and inspiring so I’m really glad that we have this
opportunity to hear a little more from you. Professor Hoffman is here to lead
the conversation and I’m just gonna turn it over
thank you and thanks everyone for coming. The way this is gonna work I’m gonna go
ahead and say just very briefly a little bit about Tim’s work and then but mostly
leave it to him I will ask the first couple of questions to kind of shape the
conversation and then we will open it up and well it will be kind of a moderated
QA that’s yeah I don’t see any strong objections to that format
great uh … so Sam Sinyangwe is a policy analyst activist data
scientist who who leverages the the power of data for for racial justice
and social justice broadly as well and it sounds easy
and cliche to say like leverages the power of data to do X because one of the
things that became very clear last night is that the the kind of work that you’ve
done as the founder of projects like mapping police violence and campaign
zero has been overcoming a lot of the limitations of what data is available
how its how its formatted how it speaks to each other or doesn’t speak to each
other and and so before I get into some of that I would love to hear a little
bit about sort of your journey and how you you came to engage in this sort of
data focused a the intensive policy work yeah so I study political science I
really wanted to be a political science professor that was like what I wanted to
be I after graduating in 2012 I joined a
policy think tank called policy Lincoln Oakland it’s a research institution
focused on issues of racial and socio-economic inequity and so that’s
really where I learned how to use data out of the context of like an academic
setting and more in the context of using a neck in an applied way to inform
program and policy interventions with their social justice focus so my job was
to provide technical assistance to 61 federally funded communities as that
were part of the Obama administration’s Promise Neighborhoods program this was
operated the Department of Education it was an attempt to to scale the Harlem
Children’s Zone model to communities of concentrated poverty across the country
what made that program unique was that it was incredibly data-driven so in
order to qualify for a 30 million dollar grant over five years a neighborhood so
30 million dollars an individual neighborhood is like a lot of money the
neighborhood had had to prove that it was making progress on the same ten
results and 15 indicators that track from the time a child was born to Tom
entered college and career across that whole lifecycle they had to prove that
they were making progress on an indicator at the population level so
that meant everything was it was building a data system my job was to
help them build a data system that could track the provision of services all the
way down to the classroom level all the way up to every every single child in
the neighborhood level that is no small feat it’s something that was massive
because in many ways you’re talking about schools nonprofit organizations
all kinds of youth development programs daycares all of these different you know
health providers all of these different agencies and organizations all have to
begin tracking and using data in a coordinated way across all
these different indicators right so things from making sure every
five-year-old was entering kindergarten court ready to learn according to
developmental assessments that every child from K through 12 is performing
one academic assessment that we were addressing high school dropout rates
that we were improving college access and completion rates and that that was
happening at the population level so that was the the initiative that I was
supporting and working to help communities use data to track all of
that to evaluate to backtrack from the top level indicators to understand what
were the inputs here what were the approaches that worked in this classroom
it didn’t work in that classroom what were the interventions that worked in
this program that didn’t work in that program with the same population that we
can really use to inform what’s working so we can scale it over time and
increase how we’re doing with all these indicators then in 2014 Mike Brown was
shot and killed in Ferguson and sparked his Ferguson uprising national
conversation shifted very much to focus on policing and police violence and
racism white supremacy in society and in their early days and weeks and months of
that conversation there was a huge question about data in particular as
community members were taking to the streets demanding justice and
accountability they were met with resistance from a lot of folks from
policymakers from academics from folks in the media who would push back and say
well how do we know that this is really a systemic issue how do you know that
this is really racism where is the data to show that this is a national crisis
and the reason the data was not available readily available was because
the federal government even to this day has not collect comprehensive data on
people killed on please so they can tell you how much my father was and rural
Missouri going like 100 years cannot tell you how many people woke up other
police last year or the year before and so that really was the beginning of my
journey trying to figure out how to apply a data-driven results based
approach that I had been using in the context of education and dressing
issues concentrated poverty to the issue of police violence where we didn’t even
have initially the data on the most fundamental indicator which was how many
people are being killed by the police and where so that was the work was
building the database that didn’t exist right putting to figure out where where
was the data turns out it was out there but it hadn’t been compiled into one
place it hadn’t been analyzed hadn’t been unpacked to add value to that
conversation so the work became bringing out where the data was so in many places
it was in local media reports whose information in police records that were
that you could obtain in public records requests as information in obituaries
information available online on social media information about that in
coroner’s reports oftentimes information available through criminal records
databases so figure out where and other crowdsource databases that it exists
existed it was putting all that together merging it together and then figuring
out how to bring it to life how to tell the story behind the data in ways that
could have how our activists across the country to use that data and their
advocacy to directly push for change so that’s been the work yeah this is this
is this is fascinating there’s a in the context of the Academy there there is a
kind of a I don’t know a stereotype or a joke about you know people engage in in
certain kinds of data intensive research going for the convenience like whatever
data sets are available that’s what we end up doing research on we do more
research on Twitter than we do on Facebook because it’s easier to get
access to Twitter data than it is to get access to Facebook data and just some
things that let that stereotype is it bears out in Santa think it’s some
domains and and not on others but the the work of compiling and bringing the
these disparate forms of data together it’s just just a tremendous undertaking
for the project and I’m wondering if you could tell us a little bit about how you
got that done yeah I mean really this was so I was
still working about a minute this time this was actually like kind of
interesting because what they wanted me to do the job that I was there to
actually do and I was like gotta build his database so I show up to work and
the CAC my time I would just be like plugging away at a spreadsheet the wait
that got it done was first of all looking at what was out there in terms
of existing crowd-sourced databases there were two databases one was fatal
encounters and the other was killed by police net these were operated by two
different people who essentially for the past several years before that would
wake up every single day and they had a system of Google Alerts where whenever a
article that had keywords and that like officer-involved shooting police
shooting in custody death whenever there were articles like that they would just
find the article get the link put it in a spreadsheet post spreadsheet online
and this is where so this information existed there was an article that came
out from fivethirtyeight.com that in those early weeks of the movement was
like the federal government doesn’t have good data on Authority a lot of police
but there are these two crowdsource databases that existed and they audited
those databases they took a random sample it was the 200 or 400 of those
cases and they checked them onto it these were 100% accurate and by the way
between the two databases there were records on about between 1,100 and 1,200
people killed by the police at per year so at that same time I was working with
the human rights data analysis group but collaborating on another project this is
a group that does they are the are the group that uses statistical techniques
to estimate the number of people who are killed in conflicts where there is no
reporting right where the media doesn’t have an accurate estimate where the
government’s estimate can’t be trusted in conflict areas around the world and
their technique was to apply multiple systems analysis to the data the two
official databases that existed so this was the CDC database in the FBI’s
database and what they found by performing this analysis essentially by
calculating the level of overlap between the two databases that did exist
to estimate the total universe of cases that existed including those that
weren’t captured by the other database their estimate was that they were about
to a lot of people kill about police each year so looking at those those are
like the two pieces of information that we had early on that were like wait the
data does exist it’s most almost all of it is here between these two
spreadsheets but there were these huge gaps so there were gaps around race
about 40% of the records in the database we’re not coded by race so then it was
how can we figure out how to code these right so then you’re looking at
information online you’re seeing if you can find a picture of the person who’s
kebap police a picture of their family or victori information you’re looking at
information in criminal records databases which do code by race in many
cases you so it’s figuring out what where can we get the gaps where can we
find the gaps there was information that was missing none of those databases
included when the person was armed or not at the time they were killed so that
was from the media reports we could glean for the media reports we could
find information where we could from police records to some extent although
we tried we tended to not try to prioritize like just the police
narrative information that was obtained from videos that were shared online or
information that was posted online about what happened and put all that together
just to help code that column and so from there it was just expanding right
so we have our set where we merge these two databases we filled in the gaps then
it was how do we bring this to life and how do we keep it updated over time
fortunately about two months or three months into this after we released the
database The Washington Post and The Guardian essentially lifted our
methodology and created their databases which like is whatever because they had
a lot more resources so they could like maintain this process over time so then
we were able to integrate some of the work that they had done after the fact
to integrate it to the database where they it doesn’t follow up work getting
information that we might so now we have access to a range of
different databases that were able to keep updated over time to merge together
to fill in the gaps over time and that has been the data set for fatal use of
force that we’ve maintained now have over six years of data that’s incredible
it’s mind-blowing how much yeah how much work went into this and but given some
of the some of the limitations that you’ve noted some of the the gaps that
you you sought to fill in and various sources some of which are more reliable
than others how how have you been able to or how do you talk about the work to
people who whether they’re in policy or in academia or elsewhere who might want
to like hand-wave away because of the some of the limitations or because some
of the messiness of the work what is the what is the response yeah well part of
this this is again why it’s been helpful to have media organizations also decide
to get into this space because you know if you’re quoting like if I’m gonna give
you a statistic and you dispute my statistic I mean the Washington Post in
the guardians database is almost identical the only difference really is
that the Washington Post database is only failed police shootings and so if
you are killed by Lou’s chokehold or some other former force they don’t
capture that the guardians database wasn’t updated past I believe 2017 but
for a long time I mean it was like you know we had helped shape and establish a
methodology that the field began using so you didn’t have to depend on us for
the data you could get the data from a variety of sources but it would say the
same thing and so you know that was sort of a way that we could overcome some of
that so you didn’t have to trust my word for it you had to look at the
information that we had posted online and if you didn’t trust that information
then the information that’s post online and this other site the other piece of
this was just being radically transparent about the work and the data
so it wasn’t that I’m just like saying a statistic out of nowhere and you just
have to trust that I know what I’m talking about all the data is posted
a full spreadsheet you can download it conduct your own analyses and so that
was part of it as well making sure that we were being as transparent as possible
so that if people did dispute the data they could conduct their own analyses
and come back so those are like two two ingredients that helped yeah that’s
great yeah now I want to get into this transparency piece for a second so there
is kind of there there isn’t some spaces what you might look one might even kind
of like call like a cult of openness where there’s there’s an idea that that
merely by you know posting the information merely by making it
available as like a spreadsheet online or something is serving the interests of
like the democratization of information but the projects that you’re working
with sort of go beyond that so when you talk about radical transparency it means
more than posting a link online and and I would love to I would love to hear a
little bit more about that yeah so it means not only doing the research doing
the analysis and having the source data somewhere and github or whatever it
means actually bringing the data to life in ways that are accessible to as many
people as possible so if you conduct a really complex analysis and the only
thing that you do is produce a really dense academic paper that maybe only
three people on the planet can read like have you really added value to the
conversation or have you done an interesting project so part of this is
how are we communicating the work with all of its complexity with all of its
challenges and in ways that that regular people can understand that regular
people can use in their work right to inform their work to shape their
behavior to shape agencies and institutions behavior so that’s been
that’s when the work is not only publishing the data but visualizing it
in ways that are accessible in ways that are localized so you can see the data
for your particular city you can see which policies your Police Department
has in place which is which policies it should have in place instead you can
read the language of those policies you can see the link to the policy documents
if you want to the source document you know you can see for it for our
database you can see each incident right you can click on the incident you can
see the person who was killed by the police you can see this being narrative
about what happened you can see the link to the source article so it’s it’s just
making the unpacking the data in ways that are accessible to people I think is
critical is a critical skill so you’re not just talking to yourself right and I
think particularly for social change particularly for changing policy it
requires being able to change public opinion requires me to change the
opinions of legislators often of often whom are not very data literate right
they demand data they demand research but they really don’t understand data or
research and so you have to unpack that in a really simple way for people to
understand and that’s a skill like that you have to know how to do that if you
want to change the policies that need to be changed to address the issue that you
care about so it is a mandatory requirement so so that’s been sort of
the work also using social media so at Twitter to unpack the data as well so it
and that means that you have to also be responsive to the national conversation
it’s not on your timetable it is on the nation’s timetable and that means that
on Twitter people are talking about police violence in the in fits and
spurts right there talk about police violence when a video comes out showing
a high profile end up with a police violence that everybody is talking about
and in those moments if you don’t shift the conversation in
those moments when the news breaks you’ve lost a lot of people a lot of
people tune out after a certain amount of time if they didn’t get that
information that they needed to understand what policy and practice
interventions what the data show are the most effective interventions for them to
take what are the advocacy opportunities that they should take who should they
call who should they should which City Council meeting should they show up – if
you’re not packaging that information for people on the platforms where they
are when they’re talking about it we’ve lost 99% of your audience already and
that’s just not OK we just can’t have that right and on
these issues that would care about so so that’s all part of democratizing the
process democratizing the work so that more and more people can use the data in
their own advocacy yeah I really appreciate that sentiment uh that yeah
that there’s a certain level of engagement involvement that is that is
necessitated by this idea of democratizing information and not just
the information wants to be free or it’s just gonna be you know we just put it
out there it’s it’s gonna work before it turn it over to the folks to have some
questions I would love to since we’ve talked a bit about the process and the
projects but we haven’t talked about any of the results what it what are what’s
something that what was a finding or two findings that really surprise to you and
then also what has been what would you consider to be some of the most powerful
sort of changes it’s effective so in terms of findings I would say one thing
that and I think I hadn’t realized getting into this work was the
importance of police union contracts to the structure of police accountability
it’s like this strange like obscure concept that seems like really in the
weeds but it turns out that your Police Department is negotiating or your city
is negotiating it’s police accountability system every four to six
years with the police union those negotiations happen behind closed doors
the City Council has to vote to approve whatever contract they come up with and
many City Council members turns out don’t even read those contracts where
they approve it and once they approve it the contract has a greenlight provision
which means it remains in effect until the until both parties agree to change
it including the police union is one of those parties so essentially what that
means is a number of cities including Seattle are locked in to an environment
where they’ve already negotiated away so many of the powers that could have had
to hold police accountable and so you have cities like San Antonio
70% of all officers who hired are rehired because of one line in the
contract or in Chicago where after every five years all records of police
misconduct are mandated to be destroyed and I just negotiated that in the
contract so that was that’s something I just hadn’t really thought about like I
thought we would be talking about issue I think in the early days and weeks and
months it was a big conversation about with body cameras and training inclusive
by explicit bias and like you know more diverse police departments
demilitarization but in looking at the data a lot of those initial things just
haven’t been substantiated as effective remedies for the situation where some of
the things that nobody was talking about and a few people still talk about are
actually really big and so after we analyzed the police union contracts of
the hundred largest cities in the country and published all that data
online researchers were able to take that information and show that not only
were the cities that had worst police union contracts more likely or less
likely to uphold a complaint of misconduct but they also had higher
rates of police violence right particularly against unarmed people and
so I think that’s been something that has been surprising just like what are
the things that actually seem to be the most important every it’s amazing the
contracts is what was it yeah and so with what is a sort of an outcome of
this project something that has been encouraging or particularly like yes
for you yeah I would say use of force policies have been an area where we’ve
seen a lot of traction in part because of the research and the data that we put
out there so each Police Department they’re 18,000
across the country 18,000 police departments they should have their own
use of force policies these are policies that determine how and when police can
use force how much force that can use in a particular situation our project
looked at the hundred largest ease in the country
reviewed each of their use of force policies establish a methodology for
determining how restrictive or permissive those policies work so based
on whether they did things like require offices teams de-escalation require them
to give verbal warnings before using deadly force ban chokehold it’s like a
variety of different things and it turns out we were able to show that Police
Department with the most restrictive use of force policies had the lowest rates
in police violence and that going we identified eight types of restrictions
and after performing statistical analysis we showed that going from zero
restrictions to all eight was associated with a 72% reduction in killings by
police controlling four demographics and community arrests Russ rates and other
things so that research has directly informed real policy changes right
things that you can point to so like Baton Rouge Police Department
after meeting with their mayor you know she actually adopted five of the eight
recommendations in their policy California just passed 83 92 which
showed just in deadly force law that law was it’s the language of law was
inspired by the research around use of force the peace act that’s been
introduced in Congress also incorporates that language and would require any
Police Department that receives federal funding to adopt those words or four
standards and now even after meeting with many of the presidential candidates
and their teams like that that has become a standard in the Democratic
field now if you look in their platforms whether it’s mayor Pete or Elizabeth
Warren or Bernie Sanders or Joe Biden you know changing police use of force
standards even incorporating the specific types of changes identified in
that study are mentioned in there right and these are the people who
hopefully will win and after power and to enact those standards through
administrator back to their the new legislation to actually get that done
tha so so I think that has been one of those examples where like just this one
study that we did with no resources really volunteers and students across
the country on helping to compile these documents and evaluate them training
people how to review their policies for their cities release contribute to
substantial change and we’re seeing in places now that have changed the use of
force policies we’re seeing a real reduction in actually making about thank
you uh now I’m gonna go ahead and open it up for folks who who might have
questions we have about we’ve about 30 minutes left we our hearts stopped at
4:30 so if you have a question you keep it succinct very awesome thank you yeah
please introduce yourself yes there has been so since we released
the project there were some initial movement in San Antonio San Antonio
became one of the first cities that we found where like this became a legit
like the accountability measures in the contract became a legitimate policy
issue that was being discussed usually the collective bargaining is all about
wages benefits and the accountability issues are just sort of slipped in there
as a concession by the cities but they’re not they’re not there’s not
reported on they’re not the subject of like debate this is one of the first
time on the City Council did not have a unanimous vote to approve the cop the
contract was like a 6-3 vote they did ultimately approve the contract they had
I believe only Doncaster actually spoke out about the contract which is like or
the first times like a major figure nationally they spoke out about it that
was like the first movement that we saw then we saw in Austin where after we
published the study again because we broke it down by City right we didn’t
just do like a research paper that was like 50% of contracts have these
provisions and not tell people which city’s had which provisions in the
contract and so in Austin organizers after we published a study saw that saw
that Austin had one of the worst contracts in the country and then they
organized a whole local campaign around it they turned out about 200 people to
city council meeting where they were voting on the contract they were able to
get the city to allow them to sit in on the contract negotiations they invited
us and so we met with us council members all the City
Council members mayor’s office and they did change the contract the city voted
unanimously to reject the existing contract and to adopt a different
contract didn’t get everything that we wanted but did change a number of the
accountability issues that we flagged select for example we flagged an issue
where a hundred and there was one hundred and eighty day rule so if you
submit a complaint about police misconduct more than 180 days after an
incident happens they in Austin they’re partners throughout the country and so
there was a very high-profile incident where the a young black woman was beat
up by the police she said she did not see make a play at the time but later on
video came out that that they think it was courted on video the video came out
about 180 days after the fact once the video came out that the police
department could do their own internal investigation if it would happen they
decided that they couldn’t discipline the offices because 180 days had elapsed
so that provision has actually been dramatically change from the contract
and there are a range of other things that also happen the contract that have
reinvested millions of dollars into community-based approaches rather than
policing approaches to a range of different things so so that has been one
of the examples where I think the city has really stood up and decided to hold
the line on the contract we’re seeing in Portland right now where there
to do the same thing and Seattle next year your contract is up for negotiation
so this could be an opportunity for Seattle to be an accident question you mentioned that a number of
media organizations adopted here methodology to develop their databases
and last night you mentioned that the FBI also in 2016 announced that that
they would start keeping better track of police killings and that hasn’t happened
yet I realize but I’m just wondering if any of these organizations have actually
reached out to coordinate with you given anything back what if those
relationships been like especially with government agencies yeah
no not like we haven’t coordinated the FBI on a data collection but they I
would say they recognized that there was the problem in probably 2015 2016 under
the Obama administration probably from a lot of pressure coming from the top
their data collection system is so this year is the first year that they’ve
announced they’re collecting data the process by which they’re collecting it
remains vlog so if they still don’t have any enforcement power if a department
doesn’t report the data and what they’re trying to do is collect data that goes
beyond killings by police to also include all shootings by police
whether fatal or non-fatal and all other deaths are serious injuries caused by
police so this would be a pretty big expansion of the data the problem with
that is the methodology that we’ve used which which depends to a large extent on
local media reports we couldn’t really apply to those other cases because in
many cases when something seriously injured but not killed by the police
it’s not reported again when there’s a police shooting that doesn’t hit anybody
but the police shot that’s not reported in the media so oftentimes so so there
are limitations to what we can collect using our existing methodology where the
government should step in and play and play a role in collecting it but they
really aren’t gonna be able to do it without any enforcement power and
they’re probably not gonna really want to use any enforcement power that
they’re granted under this administration so I think there are a
lot of political challenges to it the other challenge is data is power control
of the data is power the way that things are coded and defined there are
ambiguities in the data right with any type of data and an organization like
the FBI might think about defining things differently with an AI not to
accountability but rather towards defending the police so for example in
Dallas the I’ll never forget this this is a police data initiative summit at
the White House under the Obama administration and the Dallas police
chief gets up and he’s announcing all the work that they want to talk about
that making data available and accessible on policing so he says we’ve
published all of our data data on police shootings online you can go to our
website and see it and this really allows us when he’s talking the other
police shoes in the room he’s like this is something that you all should do
because this really allows us to tell our story so I was like okay interesting
and then it turns out we can’t go into the database the way that they coded the
data is almost just intentionally probably intentionally designed to
exonerate the police so the way that they coded unarmed there is an unarmed
category of the persons unarmed but there’s also a category called hands and
feet yeah hands and feet so if the person is unarmed but I guess the police
say that they use their hands and feet to resist arrests or whatever that’s not
coded as an arms could his hands and feet so when they’re talking about the
percentage of people that they shot who are unarmed they are grossly under
estimating data and so there are just ways like that that we see
police agencies law enforcement agencies code data in ways that reinforces the
existing narratives the other one is when they evaluate rates of police
shootings so when we evaluate police shootings we do it either on a per
population basis or a per hour rest basis the way that some people in the
conservator side of things have tried to do it is doing it on a very violent
crime basis so in other words they’re saying we’re on your rate of police
shootings is the number of police shootings divided by your violent crime
rate in a city or but the population level violent crime rates or like
violent crime rate by design created by race what that does is it grossly and
artificially inflates the rate for white people and deflates the rate for black
people so that it shows that whites are more likely to be shot by the police
when you use that methodology now why do they use violent crime well because of a
theory and an assumption that the reason police are using deadly force is because
they’re encountering violent crime well it turns out that when you evaluate that
assumption it really doesn’t hold water so only
about 5% of arrests nationwide are for violent crime and a violent crime
arrests police are only about one point three percent one point three times more
likely to use force during the violent crime arrests compared to a non violent
crime arrests so really I mean if we were thinking about a benchmark it would
either be population level which is the absolute risk of being exposed to police
violence or a per arrest level if you wanted to figure out what’s your risk of
being exposed once you’ve been arrested but perv on a crime literally is just
likely this an exercise in scientific racism like it’s an exercise in
inventing a methodology that artificially inflates the artificially
changes the dynamics of who’s being impacted by police thoughts in your data
so so things like that I’m really wary of and frankly why I think we haven’t
been as active as other organizations federal government to collect the data
and report it because we know that oftentimes they play these trips in fact
even homicide data a lot of so the federal government right now has a
category called justifiable homicide which is where a lot of police killings
are categorized and then they have other homicides right so homicide justifiable
homicide if a police so in the federal database they only have about 460
records per a year people killed by the police we know that they’re about 1200
what happens to a lot of the rest of those records well they get classified
in and grouped in the broader homicides category so that means when you’re
looking at homicide rates in US cities and your sink and you’re seeing you’re
looking at these statistics where they put up statistics on all the homicides
in Chicago for example right they use all these statistics in ways to
reinforce racist narratives about who’s committing crime and who the police need
to be policing a lot of those statistics artificially inflated because they’ve
added in police shootings to the homicides data so you have places like
Oklahoma where a city where one in six homicides are committed by police or
Phoenix last year one in five homicides were committed by police and they just
grouping all that in the homicides data and then the police are saying we’ve had
an increase in homicides this year so we’re gonna need more policing so
literally like they’re manipulating the data to serve their purposes so for
those reasons I think I’m not excited about I was curious part to the broader effort
how are you great question to the first question not well I think yeah we could
use help with that if you have a methodology that we can do that quickly
right let’s do it with collaborate preserving the documents I was always
thinking we could use the Wayback Machine
you know maybe find these I hope that works I don’t know maybe probably able
to tell me better than that I know if that approach will work if you know
three years four years down the road if the article is no longer available it’s
the second question about sustainability it’s difficult there’s not a lot of
funding for work focused on ending police violence there’s some funding for
criminal justice reform work it tends to be focused on issues like bail reform
electing different prosecutors and district attorneys some work on broader
issues of mass incarceration but not a lot of funding about how do you stop the
police from killing people like not a lot of funding in that I think there was
initially in 2014 and 15 there was some money made available by major
foundations that money I think is all gone I don’t know where it went like
honestly it went to groups that I hope used it in good ways but there’s like no
accountability for it and so now I think it we’re in a circumstance where the
only money about 90% of the funding in the space is going to very few
organizations mainly the center for policing equity and the problem with
that is one of the reasons they get a lot of funding is they’re able to say we
work very closely with police departments and that sort of in their
pitch they work with police departments that allows them to get access to all of
the police data then the police may hesitate to give us you know legally
they’re required to so they charge us a lot of money for it
so they get that Center for policing everybody can get that data the problem
is in order to get it they have to sign a contract with the police department
that stipulates that they can’t show that data publicly that they actually
can only in the end of all of this produce an internal report for the
police department and the police department doesn’t listen to the
conclusions that’s it it’s not helpful to organizers at all
so so that’s we’re not gonna sell out right like we’re not gonna figure out
we’re not gonna try to create an environment where the only way we can
say something is if the police allow us to say it or the only way we can release
data on an issue of critical Public Interest it’s only if the police
greenlight it that’s just not the work the way we do the work I think it’s
important to have people who are like insiders who can help the police build
out the systems I can actually track things that they aren’t tracking so that
we can then request that data later but that’s just not us in the space and as a
consequence it is harder to raise money and so a lot of it so the majority of
our money is actually from individual donations that come in online people go
to our websites and just like make small dollar donations
which abs and flows right that depends on public interest in the issue I think
particularly since now that wonder the Trump administration a lot of the oxygen
has just been sucked out of the room for a variety of causes like the only thing
that’s in the news right now is the next scandal that the Trump administration
has engaged in and that’s when the attention and the resources go so so
taking just you sustainability in the sector at large we’re no different than
that I do think I’ve been fortunate to get like a fellowship right so the
leading edge fellowship from 25th 2015 through 18 was able to like provide me
with like no strings attached like essentially a salary no strings attached
to do whatever work to sort of set up the vision but now that’s over right so
now it’s like figuring out how do we sustain that one strategy has been just
volunteers so like working with volunteers across the country who care
about this issue who have some time to spare and figuring out how to plug plug
them into the work so that it’s not all on you know a small number of people
have been to manage all of this data collection analysis but you know scaling
is very tough for the survey under those circumstances thank you was there anything that you
saw on the data where your intuition was like totally from I always find that
really interesting okay so I think one thing that has been interesting is this
conversation around body cameras so I think early on there was a big push
around body cameras getting all these departments of body cameras surveyed bit
ambivalent about it I was like maybe this helps maybe it doesn’t like I don’t
think that research is really there yet to figure out if this helps and in the
in the meantime we should be investing in what works there are also concerns
about mass surveillance and from what what I’ve seen I think the most
comprehensive study on body camera swing that was done on a DC Police Department
they studied thousands of officers I was a randomized control trial and in the
end they essentially found there was no impact that it wasn’t impact there was
no impact economic for positive or negative like there wasn’t the police
weren’t arresting more people not that all this video of people that they were
encountering so there wasn’t like I know a lot of the negative impacts of people
thought there weren’t any positive impacts either like there wasn’t a
change in use of force the only thing that body cameras really have added
value for in the space in general have been two things one for researchers
there are new research applications because it’s all this video available
that can be requested and obtained which means if you want to know how often the
police don’t report force after when they use force well get all the videos
and see you know how many times do they actually use force and not that doesn’t
end up in the use of force reports so like Oakland inspector general actually
did a whole study where they found that the police were misreporting when they
pointed a firearm at somebody which they were supposed to be reporting
but in a large number of cases they weren’t reporting and had the body
camera video to figure that out similarly like things like researchers
have in Stanford have found that in Oakland using the body camera video
police had an elevated tone when encountering a black person compared to
a white person so they were escalating the situation it was just another data
point that you wouldn’t be able to get otherwise
so for researchers it might be useful in some cases the video evidence has been
helpful to poor accountability in the sense of being able to uphold a civilian
complaint of misconduct when they have the video it’s more likely to happen but
other than that it’s been pretty much a wash and it doesn’t really justify all
the investment and focus on it in my opinion unless you’re axon yeah yes
they’re big on it yeah so axon the company that produces the bio memorise
another fun fact they allow the police departments to manipulate the footage so
police departments have the power to edit the footage and the power to reduce
the quality of the video and they have the power to highlight certain aspects
of the via the video and blur out other aspects of the video so again like if I
were thinking about how to structure this program it would definitely not be
giving the police department full control over the video it would be
housing that video that independent third-party organization that was
community led and that could hold the police accountable with that video but
that’s just not what we’ve seen I’m gonna actually look this oh right
down there okay one question I just I’d be curious to hear more about the
private sector yeah so the privatization of aspects of policing has been a huge
issue so when we talk about use of force policies and how more restrictive use of
force policies are associated with lower rates of police violence well it turns
out that there is a single company called LexA poll that writes 95 percent
of all the use of force policies in California and a large proportion of use
of force policies across the country and they’re run by former law enforcement
it’s a private corporation they get paid by cities to write their policies for
them and that means that we’ve outsourced the creation of policies that
could be saving lives but instead without sort stood to a company that is
not at all interested in that in fact the way that they sell this they pitch
to police departments is saying we’re actually gonna write policies as vague
as possible so we can protect you from lawsuits that’s like what they do and so
the privatization has been a huge aspect of this issue just like we’re seeing
with private prisons and privatization privatization of public prisons which is
another issue in policing is also a huge issue axon is another one predictive
policing technologies are another piece of this so that’s all becoming a part of
it and many of these corporations are not subject to public records laws and
so we can’t get access to what they’re doing it’s
you know intellectual property the algorithms they’re using so it’s very
very difficult like a black box to figure out what’s happening but it tends
to be not good yeah so you know our methodology for
doing this has been to train people across the country in how to collect the
data how to evaluate interpret the data how to review your Police Department use
of force policy or policing in contract what information you need walking into a
city council meeting to be able to provide the data that they might need in
order to push them and take a given action and we will train anybody on how
to do that and part a lot of that happens digitally so it happens at SLAC
so anybody can join the slack group and we’ll put you in a channel with people
in your state and with issues that the type of work that you want to do so if
it’s data work policy work elections work and then design and development but
we need help on all of those things so it’s sort of like very open we want to
make us as Democratic as possible we don’t have a lot of resources so we need
like as much help as we can get and in addition to that you know we’d be happy
to Train folks in person as well all right like going to a school and I’ve
done this before helping them use data in a different way and it doesn’t have
to be you know the people killed by a police data there’s a whole bunch of
data on school to Prison Pipeline and school disciplinary issues that we could
use that might be even more relevant to students right and so yeah five minutes
left I want to close on this we had a chance to talk over lunch and there was
a question I asked you and so I know the answer but I really I think it’s
important to talk about especially in the context that we’re at at the
University of Washington surrounded by researchers inspiring researchers so
there’s been some academic research that has drawn on the efforts of these
projects that has you know shown some things that are that are interesting and
genuinely useful additions to the conversation but I’m wondering in what
ways have do you see you know do do the academic community does the academic
community engages with you they give back in certain ways and they’ve
supported the project financially whether it’s donating back giving a
portion of grant funding to how has that worked
yes so they support in some ways but not with funding which nice you know so a
couple things one there has been because we make all this information public
there have been academics who have essentially plagiarized the work and
published papers that just like lifted our work and put it in an academic paper
so that is happened that happened with a Duke Law Review so we did this
policewomen contracts analysis somebody at the Duke Law Review a professor
published basically that analysis with some academic writing you know to
contextualize it it was the exact same analysis it was the exact same
visualization of the analysis but they never credited us for it so that became
a big fight and but what they did do when we held him accountable for in my
way basically went to their superior was that they share it with us they had
record we’re just sitting on these records of 656 police union contracts
that they just never published that they still haven’t publishes publicly they’re
just sitting on all of this these contracts that like we could have really
used right and so we got access to it through that process because they needed
to do something it’s like a concession they also changed the paper to cite us
and all of that and so I think part of this is as researchers and we want you
to use the data in your work but we want you the reason we made of public is to
use it like you know credit to like created the data in the study and then
also I think just as a just for the sector in general don’t sit on research
don’t sit on data that other people can use just because it didn’t make it into
your paper doesn’t mean it’s not important like first of all put it in
the paper if you can if it’s not in the paper like make it available there is so
much research and so much data on policing in particular that people have
just obtained not only researchers and academics but journalists right all
these big in-depth investigations where they never publish the actual source
data that they’re talking about right for us as organizers to collect that
data would be it would cost so much money if we can obtain it at all because
maybe the police know who we are they don’t give us the
data so publish it like make it public put it available online the source data
the policy documents that you received the the use of force data that you might
have received anything in your space in your field it’s like make that public
and then you know with funding I mean yes we would love funding figure out how
to break us off some money that’d be great but in general like if you have
time if you can direct your project towards helping us answer questions that
we want to answer but we don’t have the resources to answer we’d love to
collaborate with you on how to do that there’s so much reading this is an
emergent field right we’re talking about police violence this is something where
for decades you had billions and billions and billions of dollars being
spent on the field of criminology you had all kinds of money being spent on
academics that were using those resources to produce research solely
with the goal of figuring out how to use police as a strategy to address crime
that was like the premise of the work they had really asked or spent a lot of
time thinking about whether police were the most effective approach to crime in
the first place so they like skipped ahead a step and then they weren’t
really focused on how to prevent crimes committed by the police so it was a huge
issue right and it’s how often white supremacy operates in privilege acertain
research questions for all de-emphasizing other research questions
that could be actually answered to save lives but instead we’re focusing on the
97th study i’m talking about how do we stop crime in Chicago instead of like
the number two study that could establish this is the type of policy
that should be implemented across the board to address issues of police
violence across the country so so I think in your field where are the gaps
right where are the gaps where we urgently need your skills to answer
fundamental questions that can help save lives that’s the that’s what you should
hold yourself accountable to and not just doing like the 50,000 study on the
same issue that’s like personally interesting but may not
a use case but think about like what are the fundamental challenges that you want
to address it you just go seduce it I think that’s look great no to conclude
on so thank you I me in thanking me I pent in a row today so so there was
really like a feat of strength and we’re very very grateful so thank you for your
time awesome thanks all for coming there’s

Leave a Reply

Your email address will not be published. Required fields are marked *