Want to have a positive social impact as a data scientist?

Speaker 1 [00:02]

Thank you. Thank you.

Speaker 2 [00:04]

Thank you very much for the intro. Yes. Oops, my clicker. I'm really happy to see that there's so much interest in this topic. It's always really inspiring for me to see people interested in having more of a positive social impact in their data jobs, in their data work. And I would like to start this talk with these pretty icons. Who of you knows what these pretty icons represent? Please raise your hand. A few of you. Cool. So for the rest of you, those are the UN global goals. and they represent an agreement by the member states of the UN about what are the currently most pressing issues for humankind. And that means those are the things that the UN thinks or all the member states of the UN agreed on are the things we should be really focusing on if we want to make the world a better place right now. And you may be wondering why I start with these things and not with Say Something Tech because we're on a tech conference and the reason is and that's the other provocation in my talk is one of the things I also think is really important if you remember one thing from my talk is that tech won't save it. So it's not, and I don't mean technology has nothing to contribute to positive social developments. Absolutely, I think it does. Otherwise, I wouldn't be doing this talk. But the reason I put this T-shirt up here is because tech is really not the core thing of what we're doing when we do positive social impact. And so you may be wondering what is the point of all of this. And so what I'm trying to see, and you may remember the keynote from Carla yesterday. She also had this really nice graph where we have from technology coming both a positive direction and a negative direction versus how it might influence social development. And what technology really does is it amplifies social development. So if we have a positive social development, like, say, increasing literacy rate, technology can help us get there faster and help semi-literate people become more literate, et cetera, but also on the negative side. But things like data privacy issues or things like growing social inequality, they also get made worse by technology. And that is really important, I think, to keep in mind when we think about how we can, as technologists, as data scientists, can contribute to social good. This is a pretty complex thing to illustrate. That's why I wrote a blog post about it. If you're interested in it and want to read it in more detail, I don't have the time right now. I encourage you to look at this blog post. And also, so now the question is, now that we have some understanding about that there's both a positive and negative contribution of technology, how do we make sure that what we're amplifying with our work is actually positive and not going in the wrong directions? And there's this quote that often gets thrown around in this context, which is probably not an African proverb, probably somebody just made it up. But it says, if you want to go fast, go alone. If you're going to go far, go together. And that is the other key point of my talk that I would really encourage you, would really be happy if you remember it, is that we should never think about technology in isolation or data science or data technology in isolation. We should always think about it in the social context, and especially we should always strive to work together with those people that actually work on social issues, that are on the front lines, that, for instance, nonprofits, governments, or businesses that do positive social developments. And so rather than think you're starting with the technology, I would really encourage all of you to first start thinking about the social issues. And that's why I started with those goals, because those represent the context and the framework for what current social issues we're facing. And so for the rest of my talk, I would like to dig into a bit of detail how we as data scientists, as data people, as technologists can contribute to those global goals. and but before i do that i would like because this is a very personal talk it's shaped by my personal perspectives i would wanted to give you a bit of an intro about what my background and my perspective is on this and so i've been i've been very passionate about having a positive social impact for probably the last 15 to 20 years and since i graduated high school and i've had a lot wearing a lot of different hats that are somehow contributing to positive social impact more some more, some less. I've volunteered in nonprofits. I've been a corporate employee. I've been a data scientist and a data engineer and a software engineer. I've organized meetups. I've been a diversity advocate. I've been coaching voluntarily, and I've also donated to organizations. And in all these contexts, I've always faced issues like, how can I make my investments of my time, my money, my skills most effective? And that's what I want to focus on in this talk. but this talk is not really about me because I use the perspectives of other people as well so I would like you to introduce Alina Alina is a fictional woman, she's a data scientist she graduated from MSc in data science from the University of Potsdam about three years ago and she's kind of a combination of myself and other people I know that have dealt with the issues of how to have a positive social impact and I'm kind of combining our experiences in this talk so Alina, as I said She's now a data scientist. She works in Berlin in a tech company. They help with travel, so they sell travel arrangements and this kind of stuff. And she works on pricing, and she really likes her work. Her colleagues are really friendly, and their work is super interesting and intellectual stamina, I think. But now that she's approaching charity, some of you who are past charity like myself, You may remember that that can be a point of reflection. She realizes, I really like my work and it's very fun to me, but I would like to do something that goes a little bit beyond that, that's more meaningful. And she's starting to think about what kind of legacy and footprint she wants to leave on the world. And so that's the situation she found herself in. And in this talk, we will follow Alina to see how she dealt with this question, what kind of answers she found. And having a positive social impact is something really personal. That's why I'm not going to give you any answers to specific what you should do. But really what I want you to encourage is I want to give you some questions that you can think about. And kind of a process. I mean, process is a strong word. It's not like something that you have to follow. It's just more some suggestions about what kind of questions you ask yourself to get. find for yourself personally for your interests for your passions for your skills for your time for your family situation and all these kind of things what really works for you and want to do something that really matters to you and so the first thing there's a strategy part to this which is all about figuring out what kind of impact you want and then there's an implementation part to this which is more about experimenting now that you know what you want to achieve how can you actually do that and i use the word experiment very deliberately here because Data science in particular is a very new field, and it's even newer in the social sector. So there's a lot of unknowns, and there's not a true path yet to say, there's not a book you can buy on how to be effective in social work as a data scientist. And so there's a lot of experiments we have to do to figure out what works for us. And again, this whole talk is based on another blog post I wrote. So because I'm only scratching the surface here, I encourage you to have a look at this link and find out and read more if you're interested. Exactly. But first, the strategic decisions, which is to find out what kind of impact you want. And the first thing to figure out is what kind of social issue is relevant for you. If you remember those 17 goals from the UN Global Goals, there's a lot of different things you could be doing. And there's actually more things you could be doing. So it's helpful to focus and figure out what really matters to you. Then the next thing, once you know what kind of social issue you're interested in, is to figure out what kind of organizations do you want to work with, which organizations are impactful, which ones do work that you find where you can contribute. And finally, to clarify for yourself, when do you consider your impact to be meaningful? Do you want to have a really broad impact on the world? Are you happy with a very small scope? What kind of variation of it, et cetera? Let's dive into this. So the first question, again, is, which social issue do you want to address the most? And there's a lot of approaches for choosing your issue. I compiled a few that I've seen people use. And the first one is a very straightforward one, emotional attraction. Sometimes you really know this issue, say, global warming or the environment or feminism or the dying whale. Something is really personally important to you, and then you just know you want to work on this. That's very straightforward, and it's a very comfortable way because you feel very emotionally close to that. but some people are more rational and they worry more about other questions for instance there's a whole movement around effective altruism about how much good can you do maximally with the resources you have and they come up with two different strategies the first one is saving as many lives as you can which is given that we all have limited time limited money limited skills to contribute how can we maximize what can we do to maximize helping as many people as possible and there's a philosopher Peter Singer who says the most ethical thing we can do as people is to invest all that we have into saving people's lives because that's the most crucial thing we can do for other people. And another one that's also from the effective altruism community is this idea of reduction of existential risk. So rather than saving the lives of people that live now, they say the most impactful thing we can do is make sure that humankind survives on the long run. So they look at things that might affect people or humankind on the long run, say global warming, extinction, say being a non-malevolent AI taking over and destroying, fighting us, or a comet hitting the Earth, or some things like that. And they look at how likely are those things and what can we do to prevent them on the long run. And another one that's, again, more on the emotional side is some issues affect you personally. And some people find it really, really meaningful if they can do something against something, say, very common with health issues, something that either affects them personally or affects people they love, people that are close to them, or it affects their communities directly. And finally, if all of these more rational and more emotional strategies leave you still confused and you think you want to have a more abstract impact and you don't really want to work on any specific issue, There's the strategy called social sector capacity building, where rather than working on any specific social issue and working on any specific social problem, you try to make the whole social impact sector a bit more effective. So that's, for instance, there's another talk right after this one from Data Science for Social Good, which I'm a member of and a few other people are here. And we work on helping nonprofits with data science and do their work more effectively. and that's more capacity building because we don't really have any particular social issue directly. So back to Alina. Alina thought really hard about this question because it's really important for her to have something that matters to her and she decided that she went through all these strategies and decided homelessness is something that she finds really emotionally important because every day when she goes to work she's in the subway and she sees the people begging for money and it really depresses her and so she thinks that's really important. On the other hand, she is less concerned about having as many lives as she can because it's a very abstract thing for her, it's very remote. It's a bit too removed for her as an approach. She does think that there are some existential risks that are worth really worrying, especially if she's really freaked out about global warming and she wants to do something about it. And also looking at issues that affect her personally, is she realizes she's in a fairly privileged position. She doesn't have a lot of issues. She's healthy. She's white. She's living in a safe country, et cetera. But she does occasionally experience sexism, and she thinks feminism is an important cause. And finally, social sector capacity building is not an issue for her because she prefers to work directly on social issues. But because three causes is still a lot of things, she thinks she wonders once more what is the most important for her, And then she decides that the things you want to focus right now is on the issue of homelessness in Berlin and also the issue of global warming. So that's great for Alina. And now that she knows which courses are important to her, the next question that we really can ask ourselves is which organizations work on those courses. This is more of a research issue. Obviously, Google is a good starting point, but also there are conferences you can go to. You can ask people around and figure out for the issues that you're interested in, who works on those issues, and which organizations are effective. There's also organizations that monitor, for instance, for social sector organizations who are effective there and give them ratings. So that might be a good starting point. And Alina decides, found four different organizations she finds interesting. It's a small profit that helps the homeless. It's an advocacy group for green IT. because green IT meaning using green energy for service. There's a political party where she likes their policies and there's also an activist organization that organizes the local Friday for Future events in Berlin. Those three organizations she thinks are really worth exploring further. Cool. And the last question of the strategy is for Alina to figure out or for you to figure out which impact is really meaningful to you. So when do you feel you have made an impact, meaningful contribution to the world? And again, that's a question for research because the impact you can have is highly dependent on how big is the issue and who's affected. So there's how severe is the issue, who's most affected, where is it most prevalent. But also it's also a moment for reflection. So as I said before, some people really prefer small impact, some people prefer really large impact. And so that really has an influence on how much time and effort and energy you have to invest. And Alina decides that she really prefers a tangible impact, immediate impact. That's why for her the issue of homelessness, because it affects her community directly and it's very tangible, is the first thing she wants to focus about. And then she does some research and finds that there are around 6,000 to 10,000 homeless people in Berlin, which I think is quite a lot. And so she sets a goal for herself to have 300 of those. in some way or another within her first year of being socially active. Cool. So now that Alina knows what she wants to achieve and on which issues she wants to work and which organizations seem interesting, she can dive into figuring out how she can actually make that happen. And again, there are three steps here conveniently that I think are questions worth asking. The first is to find the approach or direction that you want to work with the organizations that you picked. Then to find out which challenges does the organization face that are worth working on and where you can contribute. And finally, then the step of figuring out how can data size and how can you precisely contribute to that. And first of all, the directions. This is a pretty complex topic because there's a lot of different ways we can contribute. And I have two slides on this, so sorry. and I can only skim the surface of them because they're almost a talk of their own but again I encourage you in this blog post which I'll have again at the last slides you can read them up in more detail so the first thing is and each of those directions works for different addresses a different kind of limitation in the social sector that you're working with and the organization you're working with so sometimes the thing that most limits people to have an impact is that they're liking people, so they just want a lot of people to contribute. Sometimes it's that they don't know what the best approach is. Then a different approach is, sometimes they're short on money. And so all these different approaches, they address very different kind of bottlenecks. And if you want to have a lot of impact, it's really valuable to find out these bottlenecks. And the best way to find them out is usually to talk to people that work in that field, work on that issue and find out kind of what is making their life the most difficult right now. And so one of these approaches which is, sorry, volunteering your skills. Another one is improving the ethical standards in the industry or sector that you're working on. This is the thing that the keynote yesterday focused a lot on, so these kind of algorithmic rules and ethical standards and all these kind of things. Then another one is academic research, which is especially impactful if nobody really knows yet how to work on the social issue that you're facing. And another one is also quite straightforward. is donating a part of your income. Another one, maybe a less obvious one, is advocating for legal changes in the legal environment. So kind of political work, advocacy work, making sure that the laws are supporting, which is especially important if we need to implement on the issue of large-scale change. And last but not least, professional, non-profit, or generally professional social sector work, which is especially important if there's a lot of work to do and it's a well-funded organization that can afford it. And so, with giving all these options, Alina decided to do things. She decides to volunteer at the non-profit that she found, the one that works with the homeless people in Berlin, and to join a political panel. Why did she do this? Well, she spoke with the non-profit staff of the organization she found the most interesting. And they told her, yeah, they receive a lot of funding. They receive enough funding to do their operations. But they're really short on volunteer staff to carry out specific support tasks. And the local laws in Berlin are not always supportive to their work, and so they would like to have some changes, and that's why she decided these two directions may be worthwhile. Yeah, so now that Alina knows how she wants to approach the social challenges she picked is your homelessness, She has more work to do, to talk with the people of the organization she wants to work with. Because now the question is, within this organization, what are the biggest challenges that they're facing? And that's also where it becomes more data science specific. Which are the biggest challenges that you can address, where data science can actually make a meaningful contribution? And thank you. That's, again, extremely relevant. It's context-dependent, very dependent on the situation. So, I really encourage you to have those conversations. But some of the common challenges that organizations face might be in implementing their programs. So, for instance, bringing food and sleeping bags to homeless people. It may be fundraising. It may be advocacy. It may be managing the organizations. It may be staffing, so finding volunteers or people or employees to find to work with. Or it might be monitoring and evaluation. So, a monetary evaluation means that work that helps the organization understand whether they're having the impact they're having and also understanding how well their programs work. So, whether they're, for instance, if they help the homeless, whether they're really helping homeless people. And, yeah, sorry. All of these can be addressed in one way or another with data science, but that's where also the creativity comes into talking with the organization to find out which is the best way to do that. So Alina, let's look at an example again. Alina, she did have this conversation with Robert, the executive director of the organization she wants to work with. And she's asking him, what can I do specifically to achieve my goal of helping 300 homeless people this year? And Robert is a bit confused. And he says, I don't really have an answer to that. We don't track our impact to that level of detail. And so Alina is confused now as well. But what she realizes as a data scientist, well, I can help you figure that out because that's a data problem and that's something where she can actually very directly contribute. So what Alina decides to work on is impact measurement for that organization to help the organization find out how many people they're affecting, what their impact is on those people, and all these kind of data tracking questions. And so for Alina, that's a really big step forward because now she knows what she can work on. She knows what people she wants to work with, what issues she wants to address. And now it's back in the classical realm of more data science questions. So experimenting with the data science tools and the methods that you can. And I'm not going to go into this detail at all. We have a whole conference about this. And I assume most of you feel very comfortable in your job. And now how to work with data science tools. So the only thing I would like to point out here is that, especially when you work with social sector organizations, you might face quite different challenges than you experience when you work in a business setting because the tools and the methods that work in these settings are often more limited, especially in technology. They may have very different data challenges. They may not have much data. They may use very limited and only be able to afford very limited technical tools or maybe no fancy servers for deep learning. But instead, they have other issues. So it's a bit of a different challenge, especially if, for instance, if you say you work your site at a big place like Facebook and you want to make Facebook more ethically responsible, you will have a very different environment. So that's probably the key question here to figure out in the specific context what works. And so that's why we leave Alina. Alina is now really excited to get working. She's designing, thinking about how to track the impact, what kind of service to ask, what kind of data to gather and how to analyze the data. And I would like to leave her here. I would be happy to check in with her in another year and see how things have evolved. And now I would like to move the conversation back to you. And so I would, as I wrap up this talk, I would like to encourage you to remember that tech is an amplifier of social development. So rather than tech being a driver of anything, tech amplifies what is there. And so if we want to amplify the right ones, we need to be really clear for ourselves which directions do we want to amplify. And then we can figure out what is the impact we want and also what can be experimented with to find how to achieve that impact and the other thing i really want you to remember from this talk as i said before is if you want to go fast go alone if you want to go far go together so in other words don't try to sit don't sit somewhere in your room and try to decide what you want to do really talk to other people talk to social sector organizations talk to businesses go to conferences go to events meet up with people that have similar questions and really or join organizations that have a positive impact. So for me, I'm also not working alone. I'm part, as I said, of data science for social good. Can everybody who's also here from the ESS key just wave quickly? So see, we have more people. And also I'm working for a consortium, Council Needs ThoughtWorks, which has both a business focus and a social impact focus, which I'm really happy to know about. And yeah, that was my talk. I'm happy to ask questions if you have any. And again, if you want to read more about this, there's the blog post where I have all the details.

Speaker 1 [24:20]

So now we have time for questions

Speaker 3 [24:29]

Hey, thanks for your talk. I found it very interesting. So my question is, so how does Alina reconcile this kind of activity with a normal job? So you presented as a data scientist, right? Working in a company. And so these are usually full-time jobs, I guess. And if you then want to have maybe also like a private life and meet friends, on top of that, there's just, I feel, very little time left to do this kind of activism that you presented. it um so i think yeah i don't know you can donate it's easy but uh doing like working in ngo or political party is not so easy um so do you have any maybe any experiences from your um i don't know how so yeah how do you do that

Speaker 2 [25:12]

Yes, I do have a lot of experience with that, and I face this issue a lot because I also work full-time, as I said, at ThoughtWorks, and I volunteer for data science for social good. And it is a challenge, and that's why it's really important for you to think about before you jump into anything, how much time can you, and what is the resource you can expend most. Maybe it's money. Maybe donating is the best approach for you. On the other hand, maybe you don't have a family, like I don't have a family that I need to take care of, so I have time to invest. and also maybe on the other hand you decide that working professionally in a non-profit is a good way because then you can combine both your working life and you will earn a lot less money so again it's always a trade-off and you need to be really clear about what kind of trade-offs work for you and know you can't do everything and that's why it's also really important to pick that one issue that you really care about, that one approach and go with that and not spread yourself too far Thank you.

Speaker 4 [26:12]

Actually, I just wanted to chip in on that because having some personal experience with that, I found that it was actually easier than I assumed to just say just two hours every week. That's what I'm doing, and that's it.

Speaker 2 [26:25]

Yeah, great point actually, thanks. More questions?

Speaker 1 [26:31]

Questions, questions?

Speaker 2 [26:39]

If not, oh, sorry.

Speaker 4 [26:42]

Yes, so thank you for a very interesting talk. So one question I have is how to find the best organizations for a specific issue. So let's say I want to go and work on climate change. Where do I look? So I can Google stuff, but then are there better resources which we can use?

Speaker 2 [27:01]

Yeah, that is also a really central question. And I find that Googling is often the best starting point, but don't just Google for organizations, also Google for newspapers on that topic because they will profile, magazines on that topic, they will profile things, blogs will profile things, and conferences will have speakers. And even if you don't go to the conference, it's maybe worse just skimming. If there's a conference on climate change, for example, which organizations are presenting that? Because usually the conference organizers did already some pre-selection to find out organizations that are relevant. So it's a lot of classical digging into information and, again, also asking other people. So Google and this kind of online information is the first starting point, but then really go to their open events, attend their meetups. If they have any, call the organizations or speak to them and write to the executive director or volunteer there. And then it's time-consuming. It's not a trivial choice, but I think it's a well-invested time.

Speaker 1 [27:59]

Questions? Okay, so I have one question. Okay. Regarding data science for Social Good Berlin and the organizations that you mentioned here, what is the data maturity level of these organizations here in Berlin? Do you find that you are helping with data collection and defining problems? What is the stage in kind of general terms?

Speaker 2 [28:25]

Generally, it's not super mature. There are a few exceptions, but most organizations are more on the stage of trying to figure out what kind of data do we have, how can we make use of maybe the random survey that we have or some other data we collected during our operations. So there's a lot of help that we do with just improving data quality, but there's also already, it's a bit of a longer process, so we start with helping organizations improve their data quality and figuring out what kind of problems we can answer with that data, and then we go into more specific events like a hackathon where we actually work on issues.

Speaker 1 [29:00]

Great. We have now five minutes to allow people to switch rooms and just a great applause to our great presenters.