In February 2021, the City of Hayward, Calif. convened an eight-week Policy Innovation Workshop on Community Safety with the goal of developing and testing policy ideas aimed at improving community safety. Workshop participants include community members and city staff, including representation from the city’s police force and criminal justice system. The workshop was convened in response to a letter from several city council members to Hayward’s city manager. ESAL Founder & Chair Arti Garg is one of the workshop’s community participants. This blog series is a diary of Garg's participation in the workshop. For additional entries please see:

• Pre-Interview Workshop - Policy Innovation through Community Workshops
• Workshop 1 - Moving from "Evidence-Based" to "Hypothesis-Driven" Policy-making
• Workshop 2 - Why Stating Assumptions Enables Policy-making in Uncertainty

Workshop 3 on March 10, 2021
March 11, 2021

I have written before about the importance of empathy in local policy-making. But I haven’t elaborated on what that means in practice. Last night’s workshop, which focused on how to interview community members about their experience of public safety, gave me a new perspective on the role of empathy in hypothesis-driven policy-making...including how best to resolve the tension between the subjectivity inherent in empathetic thinking and the objectivity we, as scientists and engineers, strive toward in our work.

As part of the innovation process, we will be interviewing community members to learn more about my team’s challenge. During the workshop, the facilitators gave us concrete information about how to respectfully conduct interviews. I was struck by how “unscientific”--at least from this physicist’s perspective--the suggested approaches were. Our facilitators asked us to empathize with our interviewees. They suggested that we focus on the interviewees’ emotional reactions to experiences they described rather than on the factual details. We were encouraged to tailor the arc of our interviews to each interviewee; adjusting, omitting, or adding questions based on what we intuit will make the subject most comfortable and open to sharing. We were also told to be mindful that our own group identities (e.g. race, class, etc) might impact what the interviewee is willing to share.

While interviews are not a common practice in my own scientific discipline, I am familiar with best practices in other disciplines. And none of these techniques meet my understanding of their data gathering standards! I remarked in the large group session how different these approaches are from those I’ve used in even less rigorous scenarios like journalistic interviews or when interviewing job candidates or grant applicants. One of the workshop facilitators acknowledged this, observing that these interviews are intended to be “relational, not transactional.” This comment resonated with me.

She was asking me to use empathy to develop a holistic, and ultimately subjective, understanding of the experiences of the people impacted by public safety policy. The interviews aren’t a “transaction” through which we extract information. They are an interaction through which we can learn to better relate to members of our community. The workshop facilitators said we are aiming to develop evidence-informed policies...though perhaps it’s more accurate to say “evidence-informed policy hypotheses.” By interviewing community members, we will gain more “evidence” about their experiences. But instead of seeking to perform a statistically sound analysis that might result in an “evidence-based” description of their experiences, we are seeking to develop an “evidence-informed” understanding of how our fellow community members experience safety in Hayward. We will necessarily need to pair that understanding with our own personal experiences to develop hypotheses about what changes could improve their experiences.

While the physicist in me fixates on the data gaps in this approach, the former congressional fellow and policy analyst knows that this is actually a more rigorous approach than policymakers often use. When I speak about ESAL, I almost always start by reminding people that, “Decisions are made by those who show up.” While policymakers want to use the best information available to make decisions, often that information is limited to the best of what is most readily available...usually made so by the people who “show up” and present it to them. The urgency of decision-making often means they don’t have time to try to uncover what they don’t know. In scientific terms, it means they don’t have time to understand the “biases” in their data. Through these interviews, we are attempting to correct for our biased evidence by seeking out new information. Because we acknowledge that our process for doing this is imperfect, we will rely on our shared humanity to fill in the gaps...and we assume this will help us come up with more effective solutions for our chosen stakeholder group.

This process is messy. The outcome is likely to be heavily dependent on the make-up of the workshop participants. But all of policy-making is like that. It’s why we have elections. It matters who is representing us in the policy process! As engineers and scientists, we may find it attractive to think of ourselves as objective experts providing unbiased input into a policy process. As I’ve argued before, this romanticized ideal is logically impossible to achieve. Scientists are biased, because we are human and humans are biased. The very act of choosing what topics to research or on which issue to engage is an act of bias. In this post, I’ve revealed my own biases that place a higher value on objective, factual information as compared to subjective, emotional information.

Our goal as engineers and scientists who engage locally should not be to maintain an artificial wall of “objectivity” between ourselves and our neighbors. Instead, our goal should be to better understand and relate to their experiences. In doing so, we can bring our whole selves--the part that feels pain and empathizes with our neighbor, the part that makes value judgments and shares those values with our neighbor, and the part that likes diagonalizing matrices even if our neighbor doesn’t--to developing solutions.

In February 2021, the City of Hayward, Calif. convened an eight-week Policy Innovation Workshop on Community Safety with the goal of developing and testing policy ideas aimed at improving community safety. Workshop participants include community members and city staff, including representation from the city’s police force and criminal justice system. The workshop was convened in response to a letter from several city council members to Hayward’s city manager. ESAL Founder & Chair Arti Garg is one of the workshop’s community participants. This blog series is a diary of Garg's participation in the workshop. For additional entries please see:

• Pre-Interview Workshop - Policy Innovation through Community Workshops
• Workshop 1 - Moving from "Evidence-Based" to "Hypothesis-Driven" Policy-making
• Workshop 3 - Approaching Evidence-Informed Policy-making with Empathy

Workshop 2 on February 24, 2021
February 26, 2021

Prior to Wednesday night’s workshop, the facilitators had finalized the “challenges” that emerged from last week’s workshops, asked all of us participants to rank our preferred challenge areas, and assigned us to the teams we will work with for the remainder of the workshop sessions. I was assigned to one of two teams addressing the challenge: “There is a lack of trust between the community and government, including City Hall and Hayward Police Department, stemming from a lack in communication and relationship building and an inadequate recognition on the long-term negative impacts of systemic racism."

The first portion of the workshop focused on developing a sense of trust within our teams through exercises designed to help us surface and share assumptions. We each shared what we needed from ourselves, our team members, and our team coach to make the innovation efforts successful. I’ll be honest. I sometimes find these types of activities a bit hokey. But, upon reflection, I appreciate their importance. When conducting scientific research, we state our assumptions and attempt to account for all the biases in our methods that might obscure our findings. When doing collaborative community innovation, we must do the same thing. We need to identify for ourselves and share with each other the assumptions and biases we are bringing to the discussions.

For the remainder of the workshop, our team brainstormed with the goals of refining our challenge statement, choosing which of the stakeholders impacted by this challenge our team would focus on developing a solution for, and identifying what remaining information we needed to begin hypothesizing solutions. One question many of us had was whether the lack of trust many community members had identified in the survey stemmed from specific interactions with the city government and police. This launched a discussion about how much the origin of the lack of trust mattered. In particular, someone raised the observation that trust is rarely built by explaining past actions.

Being trained as a scientist and having engaged in many aspects of policy-making, I struggle with this question often. How much do we need to understand the cause of a problem in order to develop a solution? In astrophysics, causation matters because it’s a central tenet of our research methodology. We observe the universe as it is, and we attempt to understand whether a given hypothesis about the laws that govern the universe’s behavior can produce what we observe. In engineering, we want to know whether a given perturbation to a system, for example some number of cars driving on a bridge simultaneously, can cause catastrophic failures. In policy-making, we don’t always have the luxury of fully understanding causes. We need to decide whether to close schools and mandate masks without certainty about whether children can transmit a virus or the mode of transmission. And sometimes causal information isn’t particularly important. There are many things we don’t understand about cognitive development in children, but we have empirical evidence of the benefits of early childhood education. So we implement government-sponsored preschool programs without knowing the mechanism by which they produce desirable educational outcomes.

Going back to the question of whether the root of community members’ lack of trust matters, I think our questions reveal that our team is working under an unstated assumption that the solution looks different depending on whether you are rebuilding trust that has been betrayed or building trust for the first time. For myself, I can say that my lived experience relating to other people provides me evidence that this is a valid assumption. And the question is whether we can proceed with policy-making on this basis alone.

After last week’s workshop, I suggested reframing the discussion as “hypothesis-driven” policy-making instead of “evidence-based” policy-making. Another benefit of doing this is that hypotheses need to be contextualized in their underlying assumptions, and this framing provides us the accountability to surface assumptions in our policy engagement as well. It also provides some expediency without sacrificing rigor. We don’t need to answer the question of whether the cause of lack of trust matters. We only need to state that our team is operating under the assumption that it does, and that assumption will inform the ideas we propose and how we evaluate and interpret their impacts.

In February 2021, the City of Hayward, Calif. convened an eight-week Policy Innovation Workshop on Community Safety with the goal of developing and testing policy ideas aimed at improving community safety. Workshop participants include community members and city staff, including representation from the city’s police force and criminal justice system. The workshop was convened in response to a letter from several city council members to Hayward’s city manager. ESAL Founder & Chair Arti Garg is one of the workshop’s community participants. This blog series is a diary of Garg's participation in the workshop. For additional entries please see:

• Pre-Interview Workshop - Policy Innovation through Community Workshops
• Workshop 2 - Why Stating Assumptions Enables Policy-making in Uncertainty
• Workshop 3 - Approaching Evidence-Informed Policy-making with Empathy

Workshop 1 - February 18, 2021
February 19, 2021

Our first group workshop took place yesterday evening. Doing this kind of work virtually is difficult, and I was impressed with the way our city staff used small (3-4 people) breakout rooms to help build a sense of community within a large group. Much of our first two hours were spent helping participants unfamiliar with lean innovation learn more about the process and familiarizing us with the online collaboration tools we would use, such as a digital whiteboard. I’ve been using these processes and tools for years in my work as a data scientist and technologist, and I’m interested to see how their use will be similar or different in a policy innovation setting.

For the final hour of the workshop, we looked at data collected through the community conversations on public safety effort the city had undertaken last year. The data included quantitative survey results as well as selected quotes from interviews conducted with community members. I was struck by how inconclusive the data were, especially if you approached them with too narrow a scope.

For example, the survey data showed that 63% of Hayward residents feel very safe or safe interacting with the Hayward police department. Considering only that number, one might conclude that Hayward is doing well from a policing perspective. When data were further broken down by race, however, racial disparities were undeniable. 50% of residents who identify as Black feel safe interacting with Hayward police compared to 69% of residents who identify as white. Also notable was that fewer (59%) Hayward residents felt safe “in general” than they did interacting with the police, with residents who identify as Black reporting a higher general sense of safety (69%) than residents who identify as white (55%). The survey report also provided some quantitative analyses of the open-ended responses to a question about how the city could improve community safety and reform policing, with the top two responses aligning to “More police presence” (24%) and “More community involvement” (20%). In short, the quantitative survey results left a lot of open questions about what the public safety issues in Hayward might or might not be.

Reading the qualitative quotes in detail, however, patterns became more apparent. Many quotes reflected that residents do not feel heard by city government overall, not just by the police department. There were also several concerns raised about the need for the city government to directly acknowledge and address racism as an issue faced by its residents, without pointing to a specific policy change but rather to a need to feel recognized and supported. And pervading many comments was a clear sense that many city residents do not have trust in our local government.

During the workshop, the facilitators used a common lean innovation technique of asking participants to write their observations of the survey results on (virtual) sticky notes, categorizing those observations, and then asking participants to vote one which of the categories they felt were most important to address through the process. We will get the results of the votes next week, but it’s already apparent that we will be addressing the community safety question through a much broader lens than just policing reform.

After just this first workshop, it’s become clear to me that as engineers and scientists we too often oversimplify the goals of “science policy.” I often hear the benefit of people with STEM backgrounds participating in policy processes reduced to our ability to help ensure that decisions are “evidence-based.” But what does that mean when the evidence that exists is inconclusive? As a physicist, I cannot honestly argue that the process we used to decide on priorities for these workshops met the standards of analytical rigor used in the sciences. As a technologist, though, I can tell you that I’ve seen this approach produce successful outcomes by identifying unexpected solutions to problems.

So, if this supposedly data-driven approach to policy innovation doesn’t meet the criteria of statistical soundness used in science, what do we as scientists have to contribute? Perhaps we need to reframe our understanding of our role in policy discussions away from ensuring it is “evidence-based” to ensuring it is “hypothesis-driven.” As engineers and scientists, we are trained to formulate testable hypotheses for phenomena we observe that can’t be explained by existing theories or principles. Prior to this workshop, many of the ideas (experiments?) that have been proposed by Hayward community members for improving community safety have centered around policing. While those ideas will be part of the discussion, they are unlikely to address the non-policing related community safety concerns the data reveal. We may need to formulate a framework (a hypothesis?) that’s more expansive than policing to address those concerns. And, isn’t formulating informed hypotheses about phenomena we don’t understand what scientists are trained to do?

In February 2021, the City of Hayward, Calif. convened an eight-week Policy Innovation Workshop on Community Safety with the goal of developing and testing policy ideas aimed at improving community safety. Workshop participants include community members and city staff, including representation from the city’s police force and criminal justice system. The workshop was convened in response to a letter from several city council members to Hayward’s city manager. ESAL Founder & Chair Arti Garg is one of the workshop’s community participants. This blog series is a diary of Garg's participation in the workshop. For additional entries please see:

• Workshop 1 - Moving from "Evidence-Based" to "Hypothesis-Driven" Policy-making
• Workshop 2 - Why Stating Assumptions Enables Policy-making in Uncertainty
• Workshop 3 - Approaching Evidence-Informed Policy-making with Empathy

Pre-Workshop Interview on February 11, 2021
February 12, 2021

Yesterday marked the first official activity of my participation in the Policy Innovation Workshop series. I met with two analysts from the City Manager’s Office to go over some background for the workshops and discuss any questions I may have. I was unsurprised to learn that the community members participating in the workshop hold a spectrum of views ranging from the belief that the police force should be substantially increased to the belief that it should be abolished altogether. As we discussed how the city staff were preparing to bridge these differing viewpoints, I realized that for me the essence of democracy is finding ways to reach common solutions from very different starting points.

While it certainly makes discussion more difficult, bringing together people with deeply held, albeit conflicting, convictions can spark new ideas and allow us to find new approaches to longstanding problems. I noted to the city staff members that while my own proclivities in policy hew closer to incrementalism, I know that the path to progress isn’t always straight. Often, it’s pretty winding, and our first attempt to break with the status quo may not always land us exactly where we want to go. I respect those advocating for extreme changes, and I am open to considering ideas that may feel uncomfortable. I don’t know what the next few months will hold, but I’m looking forward to working with my community in a new and immersive way.