service that joanne hayes white has served this great city. we are looking forward to working together to accomplish what we need to do. we are the fire department, we save lives, we respond to emergencies. we ensure that the buildings in san francisco are safe, and we make sure that our duties and staff are well-kept. thank you very much, mayor breed, for this. congratulations, and as we say, let's get working. thank you. [applause] >> all right. that concludes our press conference. there will be a swearing in at a later date. you all will be invited. i'm really excited about this, along with so many other incredible things happening in san francisco. thank you all for taking your morning to be here. and, again, congratulations to our new fire chief. and we will be happy to take questions on the side from the press. that concludes the press

conference today. thank you. >> ladies and gentleman, the chair has called the meeting to order. turn off your electronic devices. can you please rise for the pledge of allegiance. >> i pledge allegiance to the flag of the united states of america and to republic for which it stands. >> good evening, everybody, this is the march 202,019th meeting of the san francisco police commission. we have a heavy closed session so i'll limit public comment to two minutes and we are ready for the first item. >> commissioner, i would like to call role.

>> pair operato(roll call). also present is chief william scott of the san francisco police department and standing in for paul henderson is sandra hawkins from the department of police accountability. >> good evening, everybody. and we are ready now for the first item on our agenda. line 1a, reports to the commission, discussion. chief's report. weekly crime trends, provide an overview of offenses occurring in san francisco. significant incidents, chief's report limited to a brief description of the specific incidents. commission discussion will be limited to determining whether the calendar any commission.

provide a summary of planned activity and events occurring since the previous meeting. this will include a brief overview of any unplanned events or activities occurring in san francisco having an impact on public safety. commission discussion on unplanned events and activities, the chief describes women be wie limited tor a future meeting and presentation of early intervention system, eis, fourth quarter, 2018 report. >> good evening, chief. >> good evening, president and commission and chief of staff. sarah hawkins, sorry. good evening, everybody. i'll start off today with the crime statistics' update. please report again overall violent crimes down 17% and total violent crimes -- i'm sorry part one violate crimes down 17% and total violent

crimes down 19% and homicides are down and we had eight this time last year as opposed to five last year to date. our last homicide reported over a month ago. four out of five homicide cases have been cleared. and respect to gun violence, we have had a 25% reduction in gun violence, so shooting incidents, anyway, 2018 over 2017. our property crime is also down. we're 17% down and property crime and this includes burglaries, motor vehicle thefts, including auto burglaries and arsons. auto burglaries are down 22% in 202018 compared to 2017 and we'e focusing on that issue in terms of focusing on the hot spots around the city where auto burglaries occur accordingly.

overall, the crime picture is looking very pleasing and we will keep our eye on it and keep the strategies going that we deploy. i have one significant case, a fraud case to report to the commission and public. there may be other victims out there, but this is a case of what is known as a pigeon drop scam in which the suspects usually offer some type of financial incentive for the victim to mingle the suspect's money with the victim's money. under the guise of they're able to bank it or unable to the money in the bank. usually that's the fraud and basically -- usua usually they y on elderly person and once the victims the money in the bag, there's an exchange of the bag

and the victim usually finds out that there's a shredded paper or the like in the bag so we've had 11 of those crimes recently and we were able to make an arrest on the suspects. multiple counts were filed on this case, so if there are any other victims that fall prey to this scam, please give us a call and we do believe that they're particularly are other victims out there and again, people looking to scam vulnerable population including the elderly. so that investigation has not concluded yet. although we have enough to file on some of the cases that we made arrests on. we had a busy week in terms of traffic. i know i appealed to the public last week to please pay attention to your driving, your walking, your bicycling activities. we had six major injuries, traffic collisions over the past week and the most serious of which was a accident or

collision at the 500 block of john mere drive and skyline where a mother and her 14-year-old daughter were both struck by a vehicle. the driver stayed at the scene, was just a really unfortunate incident and the 14-year-old young lady is in critical condition with life-threatening injuries. so she's still in the hospital and i don't believe at this point she's regained consciousness. again, appeal to the public to please pay attention to the rules of road, particularly if you're bicycling, walking, and driving. it's really important that everybody shares a road and we've had a busy couple of weeks with traffic collisions so we are out enforcing the traffic codes of the city and the state and we will be out there. so we also will be out there to educate. and to make sure people understand what infractions tend to get people in harm's way in

tems of traffic violations. >> i'm sorry, how old was the person, the hit-and-run? >> which one? the person driving the car? >> no, the person who is unconscious. >> 14 years old. so it's a really unfortunate situation. we have a developers conference and the estimated is 12,000 and we're deployed to the area there will b.there will be a rally byr bernie sanders and we'll be deployed to assist the federal park police on that. don't have an idea right now of how many are attending but it's getting a lot of activity on social media so probably a pretty good attendance for that

event. and those are the highlights, commissioners, for this week. i believe the next item -- if you have any questions before i move on? >> any questions from the commissioners? i don't see any. >> i will present the next item on the chief's report and that's ththe early intervention proceeding. >> good evening. commissioners, chief scott and mr. henderson, i'm the sergeant over the eis unit and i'll be doing an overview of the fourth quarter eis system. i believe you should have it in front of you. is that correct?

>> yes. >> it's on page 2 and you'll see, basically, an overview of the ten indicators that we do track. >> can you use that microphone? thank you. >> better? the ten indicators we use for thest is systethe eis system ant activates an eis aflir alert fre system. on page 3, you'll see a flow chart of how the eis system works. it's activated in the system and reviewed by the eis unit. we look forward to see if it's a valid alert. if it is a valid alert, it's sent to the stations to be administered to the officer. they determine if there's a pattern of at-risk behaviour. if they determine there isn't one, they go over the alert with the officer themselves going over each item, each indicater and they send it back with a recommendation to close it. my office goes true the eis alerts to see if we agree with

it. if we agree we'll close the alert and if not, we'll send it back and we ask that the captain of the station go through the alert themselves and open up an intervention on that officer. to the right, you'll see the associated factors. they are not tract and aim. it's something the sergeants administering the alerts to look through while going through alerts. on page 4, there's a break down of the types of alerts. so fourth quarter 2018, alerts by type, we had 95 alerts triggered from three uses of force and dpa in six months and 34 with five indicators of six or more and two are four dpe in a 12-month period and six for a total of 193. members receiving alerts, we had 137 members that received at

least one alert, ten members received three, 36 received two and 91 received one. page 5 we'll talk about the interventions. so in 2015 we opened nine interventions and in 2014, we opened four and 2017, three and in 2018, we opened five currently that are still active. the five that are open currently, two are for failure to appear in court and three of them are for tactile communication. for the court ones, we do a lot of mentoring where the supervisors are keeping track of the court dates for the officers. they make sure that the calendars are filled out, both for the supervisor and for the officer to make sure they do attend court. for the technical communication, we do utilize a post certified class which is an online class on tactile communication as well as we send them to the academy to work with the academy staff

on lose force. on useless force. page 6, we're looking at the alert which is the actual paper that goes out with all ten indicators on it. so this one here is a breakdown where you can compare fourth quarter 2017 to fourth quarter 2018. and in 2018, we had 193. so the alerts did go down from last year. on page 7, you have fourth quarter alerts by station, mission station receiving 40 and park station received three and graphed there for you on the right. page 8, fourth quarter alerts by unit and it's fun to remember that alerts go with the officer and don't go true the station and if anyone officer get injured to the medical liaison

unit it shows they have on this paper they have three alerts and that's because officers that are transferred, the alerts follow them. page 9, these are the alerts by station trailing for the last 12 months and you have an overview of every month for 2018. we had a total of 791 alerts for the ten district stations. and on page 10, we are now switching to indicators and now you're looking at the indicators which are the ten uses of force, civil suits, ia complaints, you're looking at 2016, at the top there, compared to 2017 and 2018 broken down by quarter. below that we have broken down how many officers, how many sworn officers we have. so a you can see 2018 further quarter, 2,330 officers total. and then in 2018, a total of 3,582 indicators. and an indicater such as a use

of force, even though it looks like a high number, that indicater is a use of force that could involve multiple officers at the same time. so a use of force incident has multiple officers on scene. so that's multiple indicators. so that's why that number seems inflated. >> what page are you on? ten in. >> just went to page 11. we've got questions. where the numbers you were talking about on ten. >> you were saying some of the incidents, multiple officers and it's reflected with multiple reports and what exactly on the power point were you referring to. >> on page 10? so at the top on the left, you have 2016, 2017, 2018. so if you look at fourth quarter 2018, 742 indicators, that 742 indicators is not a scene or an incident. that is the number of indicators

that were sent out. mission station had a big demonstration and force used by ten officers. that's one incident but that would be ten indicaterrers. indicaters. so that's why you're seeing ten indicators from one incident. it could be an inflated number. >> let's just finish the report and we'll take all the questions. >> indicators over the three years for each indicater so you can see ois, o i.d. and all ten broken down for the last ten years and these are indicators, not incidents. so when you see 21 for 2018 and that's the number of officers involved and we had five ois in 2018 with 21 members involved in those five incidents. page 12, fourth quarter 20 indicators by station.

>> you'll see central had 91 uses of forces, one iad complaint for a total of 110 and then it's broken down by station underneath. page 13, indicators by station trailer forethe last 12 months. so this is the full month of 2018 for indicators by station. mission station having 475, park 122 and a total of 2,857. page 14, tout breaking down a little bit of the use of force. first quarter -- so at the top on the left, you have the use of force incidents for all four quarters. fourth quarter 2018, 301 incidents involving 505 members with the count of 354 subjects involved in the 301 incidents, for a total of 635 applications

of force. and down below that, you're going to see it broken down with fourth quarter for incidents, members, subjects and the applications of force and then we've compared it by total uses of force and then without pointing of a firearm. and then it's graphe graphed oue right for you. page 15 is applications of force for 2018. broken down by types of force, point of a firearm being 55% of the uses of force used at 1,489 spike strips at 13. and page 16 is a comparison by quarter. so if you look at pointing of a firearm for first quarter, there is 183 and the second quarter had 163 which is an 11% decrease

and you look at the third quarter which is 144 which is a 12% decrease to the second quarter. and fourth quarter had 150, and that's a 4% increase from the third quarter. that's how to read that slide. >> thank you. i have a couple of questions before i turn it over to other commissioners. i looked at this earlier today and there are numbers that just jump out at me and i'm wondering how the department deals with these numbers if you do it all. if you don't, i'd like to know why. page 6, we see that use of force is about 50% of all of the alerts. and on page 9, we see that the missing station has by far the most alerts for a station in a trailing 12-month period. i'm just wo wondering, what does the department do with numbers

like this because they jump out at us. >> so mission station is usually the one that's always number one because of their call lion. they have the most calls and based on the number of calls they're responding to equates over to the number of indicators that that station receives. >> so are the calls there, were they, for example, be twice bayview or twice the tenderloin to match pretty much the alerts? >> it varies allot. a lot. we did run the numbers comparing the stations as far as indicators this week and you just see a little bit of a scattered pattern no matter which station you're at. sometimes they're high, sometimes low and evens out to where this is your average. >> any way to analyze whether this is a cultural issue or are they inclined to use more force than one of the other stations?

>> commissioner, so once we get the data, then it's what we do with the data, so yes, there are things we look at in terms of the issues. the system won't tell us that, but you start to look at things like supervision and leadership and what the patterned an pattes are. there are multiple incidents where you have multiple officers which tends to spike the numbers a little bit. those are issues that we can draw down on and with d we do lt those things. it won't necessarily be something that will be apparent based on our system. but we can and we do look at those things. >> i would just think that any is report would trigger inquiry because it does raise numbers that you can't analyze or understand just on their face. >> right, exactly.

you are absolutely right. and it's is incumbent upon us to take a look at what those numbers mean in that context so there are cultural issues or maybe training issues that might impact these numbers one way or another in terms of doing business a different way or better way or more efficiently. that's up to the management or leadership to take this data in and take it to that next step. and i will add, too, that the commissioner, we are looking at other systems that give -- there are systems out there that gives additional capabilities to do a little bit more of that from a technical point of view than our system is capable of. so you have to human eyes on it but there are other systems out there we're considering and we just had a presentation today and we're going to follow up on that research to help us get better in this area.

>> ok. vice president taylor. >> hi. i've got three questions. first on slide ten, i'm looking at the able to the right. indicators per member. what is you does your zero reprt there? >> the number of officers that did not receive any type of indicators. so for instance, 2018, you're looking at fourth quarter, the bottom row. there were not 1,814 members out of the 2,330 that received no indicators at all. >> got it. >> then you have 370 that received one. >> ok, understood. now looking at slide 11, the next slide, there seems to be a pretty staggering increase in officer-involved shootings over the last three years. am i reading that correctly? so going from five in 2016 to --

or indicators. >> these are indicators. >> yes, i apologize been 21 in 2018. to what do you attribute that? >> so these are just the officers involved at the scene is what we're looking at here. so it's not a reflection of how many ois there were. >> ok. >> so for instance, there were five in 2018 and of those five, there were 21 officers involved in the five. >> where does it say there were five in 2018? it does not. i just know the numbers. so this is not an indication of the number of shootings. >> they are indicators for the purpose of eis and sending out alerts. >> do you know to fill in the blanks on in. >> you just know 2018?

>> i know 2018. >> there were seven in 2017, commissioner, seven. >> are you able to identify patterns with officers? if you have the same officers involved year after year, are you able to track that? is that part of the work being done? >> we are able to track that, yes. every time an alert goes out, the system tracks. so it's there permanently. why it was triggered for that particular summer. officer. if they tri trigger again, we kw to look back to see what the patterns are and a lot of times they're just for that quarter or particular period so the supervisors that are called back and ask questions, we're able to relay, they're prior alerts were for failure to appear in court. while this next one was because

of use of force so that way they get a broader picture ba the br. >> are you responding that's consistent throughout the department? so if it's the same officer year after year involved in incidents like these, is the supervisor's response in the mission consistent with the response in park? can you give us a sense of what happens when you have the same officers given these alerts, giving off these alerts. >> are you asking do officers get alerts, alert and alert each time? >> i'm making the assumption some officers have gotten alerts more than once and i want to know what the response is and if that response is consistent across the department. >> they look at a full year. so we look at a 365 calendar year and myself, i get an alert because i have three uses of force in three months in march.

so i'm going to get an alet. my captain will get an alert back to march of 2018. so they'll be looking at march 2018 to march 2019. and all of my indicators are listed on there. whether it's my use -- all ten indicators. so they're looking at the entire year of what my work has been. that they're supposed to research everything on that year. they look at all of the uses of force, pull the use of force reports and review them and look at the body camera of the incident. we send them the torque claims and civil suits and also the dpa complaints so they have a whole package of something to go through. they go through each alert and each indicater and make a decision on that. they're not investigating but looking at a picture and looking for a pattern of that risk behaviour with what's on that alert. >> i guess what i'm getting at, is that decision of what to do, kind of dependent on the individual supervisor or the individual station? or are there guidelines or some

consistency across the department as to how you respond when you have officers having multiple alerts? >> there are consistency. the db lies out how to go through the eis indicators, what to do the next step and my name and my email and my phone number on that db to call me for any type of question that they might have. but yes, they go through it the same way at mission as part. >> you're keeping track of what the responses are? >> yes. >> this may be a rhetorical question but if you had to create an executive summary off of the 16 slides, what does this

say? >> well, it's kind of saying that our alerts are going down at the moment. why it's going down is kind of why we have the university the chicago. in 2016, we ruled out body camera and in 2016 we also at the same time required that the supervisors sit down with every officer and go through that alert so now you're seeing a pattern of how to gauge something that hasn't happened. they're sitting down and did that face to face talk stop some type of pattern or at-risk behaviour in the future? so it's hard to gauge that, but those two things happen the same year and now we're seeing a drop in the as alerts. >> and i know as we move forward and really, like i say, it's a rhetorical question working with the university of chicago, but ensuring there's that sort of language, some sort of executive

summary in the beginning and/or at the end so that members of the public, they see the numbers as we see the numbers and to the commissioner the numbers jump out because there are so many but if there are summaries that the indicators are dropping and that would be something to think about as we move forward. >> commission alias. >> i want to turn your attends to page 3, where it has the list of associated factors. and some of the factors that are listed, i'm wondering how those factors, sick pay not protected by federal or state laws, voluntary overtime work, how do those factors predict whether an officer will use force or not? >> it's not whether they use force for not but whether there's at-risk behaviour or something happening with that officer that needs to be addressed by the supervisor, not

necessarily use of force. for instance, when the sergeant is looking at the training history or the number of vehicle stops that officer does, if they're looking at the the number of vehicle stops, i stopped two vehicles a month and my partner stops 30, well, then, hi use of force probably is not my use of force won't be the equal of my partners so those are why they have those there. but it's looking at the totality of everything on that alert, not just the use of force of what they're looking at and what they're trying to accomplish. >> i understand that and some of the factors associated factors make sense but one was weather voluntary overtime worked and i'm not sure how those equate into that. >> well, overtime so when that

supervisor sits down and talks with them, what's going on with you? you've had five complaints and that might open tow up to a face conversation to find out what's going on with this officer and why are you working so much overtime? do you need that overtime and now your job and home life is suffering? it's that what they're using for the associated factors. >> on the left side on the flow chart it incomes three levels of review or actually the two that happen before it gets to you, right? >> right. >> so after the threshold activator goes off it's reviewed by the officer sergeant and if the officer sergeant feels that it wasn't something that should have -- the activation shouldn't

have gone off and they decide to closed it, does that information get to you or does it go into the closed file? >> oh, no, they all come back to me. i send them out, they're sent back to me and they say i recommend it closed, i review it and if i don't think it should be closed because i'm saying i don't though, i see a pattern, i can override that and send it back to the station and telling the captain that an intervention should be opened on this particular person. >> why is there a review by a supervisor between the sergeant and you, then? >> so there's a review by the eis sergeant which is me to find out if it's a valid alert. because it could not be a valid alert based on what the machine is spitting out and it's reviewed by the supervisor who i send to the station and i look at it, it's a valid alert. they go over it and he sends it back to me and that verification is by me to verify their work if it's at-risk or not at-risk. >> and then where is the data that shows that this eis system

works? >> i don't think there is data to show it works. [ laughter ] >> aii'm wondering how often it identifies whether an officer will use force. you know, these are pretty graphs and numbers but where is that data? >> that kind of goes to what i was talking to commissioner brookter about, it's hard to measure something that hasn't happened. so whether eis works or not is what the university of chicago is doing. they're trying to say the threshold system may not be the best and should be from the data, science report. so that report which we present at the next quarterly meeting with hopefully delve into this question. >> how long have they been

analyzing this to get us this report. >> we started sending data mid2016 and we continue to send data as we kept getting it in. they would ask for certain amounts and it takes awhile because they're si lookeiloed fs batch and sent them data as we were able to get that and they probably that for a year and a half that we were crunching the numbers and they kind of fell out. >> three years later we get a report? >> yes. a number of things have happened with the university of chicago in that time. but we should get the final report within the next couple of weeks. >> how much are we pay for the system? >> it was under a grant. be didn't pagrant.we didn't pay. >> have there been occasions where you've gotten a report that says this is closed and you send it ba back and no, there'sa

problem here? >> no. director henderson? >> i wanted to some of in context to what was raised by the commissioners but to point out all of the work, this early intervention system is one of the recommendations that the doj had asked us to really focus on and to point out that it's an emerging industry which is why there aren't a lot of programmes around doing the type of work and this answers the question that was brought last week commission dejesus was asking about the system and what opportunities for more training or other alternatives and there's not a lot out there but that speaks to some of the stuff that's in the pipeline right now that is better than the thresh threshold approach which focuses more on a system that could be data-driven and that's what's in the pipeline that i think

sergeant youngblood was references in my office and dba has been working closely with the department as we've been trying to something like this together. i would say that just to reiterate what sergeant youngblood was saying, that report is coming out in the next few weeks. i would imagine that at the next quarterly report of this, you would have something specifically and concretely to look at and evaluate from that programme and i hope that's what we stay focused on, not to throw this out the window because we have something with numbers here and a lot of the parties are at the table. but obviously, i think, the goal is defining something that has a higher or at all a predictive value. >> commissioner dejesus. >> this is complicate and part of this, we need to understand, this is to try to predict at-risk behaviour and to try to predict offices h that might get in trouble down the road.

there's to science to this, but, for example, one of the things i learned at the presentation. you have an officer who has a batteries, resisting arrest, use of force, civil suit, you are getting those kind of ca categories, those are the kind i look at and look behind the scene to what that type of behaviour is. versus if you have somebody late to work, missed the court appointment, things like that, who may have other issues in their life and not necessarily would show a person who may be prone to at-risk behaviour in terms of use of force. we have a threshold activation system and what director henderson is saying we get a lot of fall negatives. that's why you get 3,000 hits and three people being counseled. so the numbers are shocking. on the other han hand, we don'tt a lot of people being counseled

because we like the idea of not a lot of people getting hits. the university of chicago abandoned us and kind of come bab and given a tentative draft report and it's theirs and the department will be working with them on that i guess that will be coming forward soon. and what the chief said, they were kind enough to invite me to see a different system and it more often event-driven system, more holistic in terms of not a check box. ten officers had their guns drawn on one person and that's ten things. it's more holistic in terms of looking behind the scene, looking at the event, looking at the charges, looking at the officer's training, looking at the officer's personnel record, looking at the officer's shift and his unit and how he compares to his unit. so it's a little bit different and something proposed and they were kind of enough to let me sit in on that.

that's what you have. you'll have a checklist which doesn't necessarily help us and then, of course, the university abandoning us didn't help us either. on the other hand, there was another approach but it's new, also new and it's similar to what they use in l.a., if i understand that right, chief. but that might be a better approach for us. so that's something that the department is looking into and hopefully, we'll see if that might be right for us but it's quite expensive and it would also -- we have a lot of data that's pretty complicated but it would have to be added and tailored to our use but more of a holistic approach in terms of trying to analyze an officer's behaviour. did i say that right? >> is the university of chicago

report of any value to us or is it so flawed that it's really just junk? >> we pointed out some things to take a second and third look at. they had a lot of turn-over, and i'm not making excuses. so when we get the report, we'll look at it and evaluate it and bring it to the commission. >> can you give us a brief idea of what the commissions were? >> 18% reliable -- criteria was 18% valor and i forgot the wording and it made it sound like we're failing but we're not because we don't want a high percentage. i don't have it in front of me. but i think he has a staff person. they didn't have all of the stats. they went ahead without our full stats and then they made predictions based on that and so having uncomplete information

and that's one that i remember. i don't remember anything else. do you remember anything else? >> there are a lot of misspells and mislabelings of graphs. >> really? >> during the conversations back and forth, they would tell us this data is from 2016 to 2018. and then during the call, the report said 2017. so they were giving us conflicting information of what time period they were using. the graphs didn't match up. the typos were common and there were a lot of little things but when we had the analysts run numbers to bea compare what they were saying in our report, they didn't match up. we asked them to address that in the report and there was a lot of -- there's a little bit of contention where they were saying they didn't have the time to work on this the way that we wanted to and so -- but now they're working on it. >> thank you. >> thanks very much.

anything else, chief? >> no. >> ready for the next item. >> item 1b, dpa director's report, report on recent dpa activities an announcements. dpa's report will be limited to a brief description of dpa activities an announcements. commission discussion will be limited to determining whether to calendar any issues raised for a future admission meeting. presentation of dpa's audit of sfpd use of force and director paul henderson is present from dpa. >> good evening, director henderson. >> good evening. my normal update, we have 138 cases are open at this point this year versus 112 which was at this point same time last year. in terms of cases closed, we are at 121 cases we've closed at this year versus 107 this time last year. in terms of cases pending, there

are 295 cases pending versus 245 cases. these numbers are reflecting the spike in cases we've had recently. in terms of 17 cases sustained, we are at 17 cases sustained so far this year versus six this time last year. of the cases that are being told passed the 270 days, we are at 228. of that number, 19 toll an 19 td again, these are the cases this the pipeline for over 270 days. cases mediated, we are at four and we were also at four the same time last year. in terms of operationally, what's going on with the department, we are in the midst of our contract negotiations for the new case management system and we've been working on that for months. but i think it's nearing a

close. in terms of strategic planning with our outside consultants, we're continuing to work on that and we're also going through our budget process right now. we've submitted our budget proposals and we are waiting to hear back from the mayor's office. the 19th of this month, the office attended both richmond and taravell meetings and tonight, we have one of the cases from closed session. the courtroom today or in the hearing room today, we have our senior investigator steve ball, two of our attorneys, stephanie wilson and tanetta thompson and my chief of staff, sarah hawkins, in case there are issues to come up in which we can be helpful or the department can be helpful for the audience.

>> mediation does not seem to be very popular and yet, it's actually very effective. what can we do? >> what we have done is just recently, we had the entire mediation team go around with the chief and speak to all of the command staff, talking about what the system is and how it works. and i doubled the staff that was this that department, because i, too, think that the mediation stuff is really important and i'm trying to make it more robust. it's one of the highest rated elements in the agencies and it constantly gets the best feedback out of all of our operations. people really value participating this the process and i think it's beneficial to the department, as well, specifically since it gives the officers an opportunity to hear directly one-on-one from the folks they are interacting with on a daily basis and vice versa. the community feels like they

are given a voice and an audience with law enforcement in ways that they have not experienced in the past and don't typically experience based on whatever interaction they've had that brought them to the dpa in the first place. so the department has been at the table and we are certainly at the table in terms of trying to come up with creative ways to expand that work, but obviously, i think there's more work that can be done. >> it's up to the complainant who comes in. if the complainant is willing to mediate, the officer has to then mediate the case some? well, they don't have to, but if they don't, the alternative is to initiate and proceed with the full investigation and possibly a sustained case and/or discipline. >> there's every reason to do it. >> every reason. >> the problem is the public so far has not been interested or willing. >> it is a com b combination of

things. obviously there are also standards that are restricted from falling in the mediation so if there are serious cases, those are cases that can't be mediated at all and we have taken an expandive view and certainly more expansive than was in existence for mediation, broadening some of the categories. an expect, so, for instance, an officer-involved shooting, obviously that's not going to be a case that could be immediate d but if on the way to an officer-involved shooting, the officer did something inappropriate that had nothing to do with the shooting, there are no longer these blanket restrictions or block tos blocke any cases mediated. i'm trying to get more thing in the pipeline but part of it is educating the department and community about how the programme works to be more effective to get broader numbers. >> thank you.

vice president taylor. >> you hit on my question, is it a matter of educating the public. >> both. >> and the fact this is available to them and potentially this is a good thing? >> it's both. surprisingly at both ends, even when we make referrals, if they don't want to do it, they don't do it. i have a whole talent pool, over 100 folks that are registered with the agency to act as mediators that we rotate and some of the things we've been doing over the past year is training all of those folks and reevaluating who our mediators are, making sure that everyone is unbiased toward whatever and they have specific training to have the best outcomes for the mediation. we've set the table and now we need people to come and sit at it. >> part of your regular reporting to us, can you include kind of facts about the mediation, the percentages of

cases that go to mediation, successfully complete that process? >> the number of cases being referred? >> a referral doesn't always mean that people agree. is that right? >> correct, but it will give you a number to quantify in terms of the numbers coming in and of the numbers that come in, how many get referred. yes, i can come up with stats for you, yes. >> moving on to your next part. >> i was writing notes and forgot. so next, i have a presentation that i have been talking about for awhile. on our audit. and just to make sure that everyone knows what we're talking about, in 2016, voters with prop h came up with audit specifications and requirements for the dpa and so been when i

came in to the department in 2017, summer of 2017, the mandate was there but the work had not commenced yet. and so, this was a big priority of mine coming in with directives from the mayor's office, from the board of soup storesupervisors and from the pe commission from that time period and the community, as well, to get that done as efficiently as possible. this is work very similar to the work the controller's office does regularly and they were whom i turned to initially. so the folks that you guys are about to hear from are from the controller's office and i just want to bring everyone up to speed. it's been awhile since i've had the focus on this audit and i wanted to present something and make sure the public knew what was going on and the commissioners, especially right now, that we have new

commissioners knew what was going on with the work and had an eigh ability to get input ani think that's important and that report will come out soon but i wanted to frame it so that everywhere haeveryone had a bror understanding and here it is now. without further ado, i will have steve legerty present to you and explain about what our audit programme is. >> good evening. >> good evening. tonya could not be here tonight. just by way of overview, tonight's presentation, we're

going to provide an overview of the audit on behalf of the department of police accountability, on the san francisco's reporting procedures for forced data. we'll go over our audit scope, audit methodology and we'll discuss what the next steps are and to allow for time for questions and answers. jusjust by way of introduction,e city services to conduct financial audits, the city auditor's activities. so the work performed is done in accordance with u.s. accountable

office generally auditing sarredknownas yellow books and s is done and supported with a relative quantity of evidence. we're a peer-reviewed shop which means we're meeting the standards of the usgao. we believe we will assist in improving public accountability and public performance and operation. so what brought us to this point, it was the passage of proposition g in 2016 which

aamended 4.136 in addition to the police accountability and it placed the audit mandate on dpa to conduct a review every two years on police officer use of force or how they handled officer misconduct. for the purposes of the first audit, we elected to focus on polygrappolice officer use of f. i think the question that has come up during the performance of the audit is how is the work we're performing different than the admin code, section 96a placing law enforcement recording requirements on the san francisco police department. it provides reports to the mayor, uses of force an arrests. just a side by side comparison of 96a and the audit work we're

performing. our work differs in that we'll be assessing adequacy in the san francisco police department's collection and reporting of use of force data. and tiffany and i will get into that later as we discuss these objectives of the audit. and so another question that's come up, how do we decide to focus on use of force data for the first audit and this is pretty important. the decision to audit use of force data came from a review of other similar work being performed in other jurisdictions across the nation as well as consultation with police accountability. it's important for a few reasons, that it's mandated by local and state laws and it's an indicater for the early intervention system as we covered and can be reviewed by

the police academy regarding use of force and report writing. >> now i'll go through what we are focusing on in our audit. so we selected calendar year 2017 so based on a few factors and one of which is in 2016, the department of justice issued a report and made 276 recollections to the police department, 58 related to use of force. in december of 2016, the police department issued a use of force revised use of force policy in the general order 5.01 providing new guidance on use of force data collection reporting and they also instituted a new form to collect data. so as steve mentioned, overall objective of our audit is to assess the adequacy an

effectiveness of the police department's collection of and reporting procedures for use of force. we created subobjectives. our procedure is well-designed, data complete and accurate, do the police department's use of force policies guide supervisors to review whether use of force is within policy, is available data relevant and has the police department implemented department of justice recommendations related to the use of force data reporting and collection process? as you can see, this is a chart and since we're focusing on calendar year 2017, this is from the early intervention's fourth quarter report in 2017 and as you already know, the police department's force and to get a true sense, there are 3,150 uses of force in 2017.

>> objective number one to determine if use of force procedures are well-designed. what this means from our perspective is, we're evaluating use of force data collection and reporting procedures and determining if they're accomplishing objectivelies and consistently applied by police officers when they're reporting use of force and by supervisors reviewing the application of use of force. other factors that we're evaluating in the course of the audit are isn' are sfpd's cultue data reported and reviewed. >> for the se second objective we're reviewing whether the data is complete and accurate and looking at incident reports and other records such as body-worn camera footage, forces

underreported, overreported and whether it aligns with the incident reports and whether it's consistently applied and whether non--sworn personnel are also applying the use of force collect policies consistently. (please stand by).