 Great, thanks Kevin. So is this working? Can everybody hear me? Yeah, okay So thanks Kevin for introducing me. I'll give a little bit more background about Who I am and where I come from so I did my undergrad grad school and postdoc at UC Davis And then eventually it was forced to move out to the Bay Area, but I love it here So we moved out here two years ago to Pleasant Hill I did my undergrad and graduate work in microbiology specifically in microbial genomics and then I did a postdoc in biology education research and So I'll be bringing some of that to this talk and then after my postdoc I joined data carpentry as the associate director and I've been there for about a year now And I'm really excited to be here today to talk with you all about the work that the data carpentry and software carpentry communities are doing in helping researchers make their work more efficient and more reproducible. So I wanted to get a little bit of Background on people in the audience. So can I get a show of hands? How many of you have heard of data or software carpentry? Okay, good before today. Yes And how many of you have been to a data carpentry or software carpentry workshop and how many of you are data or software carpentry instructors? Okay, or about to become instructors Lauren. Yes So thank you for for that I wanted to get a little bit of an idea of who's in the audience because I don't want to tell you all about What data carpentry is if you've already been involved with data carpentry? so I'm going to give a little bit of An overview of what I'll be talking about today. So first I want to talk Most of the people in this room are already very familiar with the reproducibility crisis and with the urgent need that We have for developing people with data skills and by data skills What I mean is the ability to organize manage and manipulate data sets and that's everything from collecting and cleaning data to Analyzing and reporting data in ways that are reproducible So we'll talk a little bit about that just to make sure everyone's on the same page And then we'll take a step back and talk about what would it be like to live in an ideal world where we didn't have this Problem what would that look like and then can we use that vision to redefine the problem in terms of cultural change? So cultural change is very closely related with what I was doing in my postdoc So I'll bring in some of that background there and talk about some of the analogies Behind the educational work. I was doing in my postdoc and the work that I'm doing with data carpentry now And then from there once we've talked about cultural change and the principles of how cultural change works And we'll talk about some of the ways that data carpentry and software carpentry communities have Helped to use what we know about cultural change to solve this problem that we've defined And then I'll talk a little bit about where we as an organization are headed in the future All right, so a little bit of background about the problem So it's no surprise to everyone in this room that we're collecting lots and lots of data And we're collecting more and more data every day and the velocity isn't going to slow down in the types of data that we're collecting Isn't going to D diversify we're just going to keep getting more and more types of and more complicated types of data so one Example of this from my background from genomics is that with increases advances in Sequencing technology we now have the ability to collect vast amounts of DNA sequencing information so this shows a graph of number of nucleotides in the and in the embal database Between what is that 82 and 2010 so you can see that we've Generated huge amounts of sequencing data and this isn't slowing down ever again so the Thing to take away from this and this isn't just in genomics I'm sure in each of your fields you can think about the ways that big data is affecting your domain and One of the ways that this is affecting us is that? It's no longer the case that once you get a big data set you ship it off to somebody who's specialized in Managing that type of data so big data is now the domain of every researcher Every researcher at some point in their career is going to have to Figure out how to manage and how to analyze this type of data And that can lead to a lot of data pain and frustration so this is the point in your workflow where you've Created and designed this wonderful experiment and you've collected all of your data And you now go to look at and analyze your data and realize you don't even know how to open the file that the data is in or maybe you don't know how to Clean and organize the data and so this can lead to a lot of sadness And a lot of anger and I'm going to refer to this as hitting the data pain wall So we've all been there raise your hand if you've been there Raise your hand if you're there right now Okay So the problem then is how do we figure out how to scale? People with data skills at the same rate as data production is scaling So we know we're going to keep making and collecting more data How do we ensure that we're producing more people who can handle that data as it comes out? and so the Very very simplified answer to that question is training and I'm going to go into a lot more detail about that, but basically It's a simple fact that if we want more people to be able to have data skills at some point in their career They need to be trained for those data skills, and then the question becomes where and how do we do that training? So just one example of how we know that training is Is a need? This is a survey from 2013 from Bramble in Australia where they asked their their community What is the biggest hurdle that you face in your career? So those are the blue bars over there and the biggest hurdles were a lack of expertise and lack of training so this is something that impacts people's ability to do their job and The most useful thing that Bramble can do for its community is offer training and I like to point out that training is Perceived as a bigger need than compute power then access to data and then funding So it's not often the case that anything is more needed than funding, but in this case. Yes Researchers know that they are lacking these skills They know that there are projects they want to do and their problems They want to solve that require this type of skill set and they're looking for and they're they're They're eager to get this type of training So then the question becomes where Does it does this training come from? How can we offer this training to researchers? And how can we make sure that we do it in the most efficient way possible? And I don't think it's going to be contentious at all to say that The undergraduate curriculum is already full to bursting at most universities We can't necessarily just add another course for data skills That's Not even taking into account the fact that a single course is not going to do the trick right so a single course in how to analyze data is probably not going to solve the problem and because we have a deficit in number of people overall who have data skills we also By extension have a deficit in number of people who can teach those data skills at the university levels So we can't just slot this into university courses So I want to take a step back and think about what would it look like to live in a world where this wasn't a problem What would an ideal solution be? so In an ideal world We would learn early these skills for conducting research and analyzing data Maybe not necessarily that early, but pretty early And we wouldn't learn it once because as I'm sure everyone in this room has experienced the technology keeps changing Right, so you learn one skill you learn one tool and then six months two years down the road That's no longer the tool to use so we need to build up a culture Where not only are we creating opportunities to learn these skills early? But we're creating opportunities and supporting people in learning them throughout their careers and This is really a question of cultural change So we're building up the opportunities in the support system for people to become lifelong learners of data skills So data and software carpentry are organizations that I'm going to talk about in a little bit more detail as we move forward, but data and software carpentry in general run workshops to help people Develop these skills in data analysis and software development data Carpentry is more focused on developing data skills and software carpentry is more focused on beginner skills for software development and so I'm going to switch over to talking about the way that these organizations Have taken into account what we know about how cultural change does and doesn't work into our model to help achieve this ideal world so a little bit of background about my Previous experience so in my postdoc. I worked with a group that does science and Technology and engineering mathematics of STEM education research at the We changed our name in the middle of my postdoc so it's a center for educational effectiveness in the Department of undergrad education at UC Davis and one of the things that this group is tasked with is trying to encourage Professors and other instructors at UC Davis to use evidence-based teaching practices So again, I think that many people in the room will be familiar with the fact that getting instructors to change their instructional practices is hard And that this is a difficult cultural change problem And so one of the things that I took away from my postdoc was this appreciation of how cultural change works and doesn't work And the things that are important to keep in mind when trying to undertake a project like this So I'm going to bring in that parallel as we talk about how data carpentry and software carpentry work to change this culture around data skills so one of the most reproducible outcomes in the study of how cultural change works is that you can't change someone's practices unless they agree with you that Those practices are worth doing so it's hard to convince someone to stably and reproducibly Do a specific thing unless they agree with you that that thing is important and before you can get them to Right, so that sorry So to get someone to do something you need they need to believe that it's an important thing to do but getting Convincing someone that something is worth doing is very hard Changing someone's attitude is very hard So the trick that I worked with in my postdoc and that data carpentry and software carpentry Work with is start with people who already agree with you if you're working on a big cultural change problem find the people who already agree with you that this is an important thing to do and Capitalize on that enthusiasm So for data and software carpentry We know that there is an eager body of learners who already agree that these skills around Data analysis and reproducible research are important to have and we know that those People exist and are eager to learn because we have them lining up and flocking our workshops. So we have I have the number here. We've run 600 workshops over the past couple of years and we've had over 5,000 learners at those workshops. So we have Always wait lists and the workshops are sold or not sold out, but they're full within a couple of days of opening So we know that there's this large body of eager learners who are ready to To take on these data skills and to take on these practices And the reason that most of those learners are eager to take on those practices is because they've hit that data pain wall Right, so they've gotten to the point in their research where they know that they need Some work around something that they don't already know how to implement and they're ready and eager to adopt those practices another Lesson that's been learned in the field of cultural change is that it's very difficult to change Practices that are established So if you're working in my case in the postdoc with Instructors who've been teaching the way they've been teaching for 20 or 30 years You can imagine it's very difficult to convince people in that situation to change the way that they're teaching And for good reason it's worked for them for 20 or 30 years. Why should you invest in changing your practices? so The fix and the workaround that we worked with in my postdoc was starting out with Teaching graduate teaching assistants these practices because they're starting at the beginning of their career So same thing for data and software carpentry We end up if you look at our workshop audience Working primarily with early career researchers. So this includes a few undergraduate students but mostly graduate students postdocs and early career faculty And again, this has to do with that data pain wall. So these are people who are just starting to feel The effects of not having these skills These are people who are now working with large data sets and have no idea how to proceed and so they come to these workshops and they're very eager to pick up these skills and now that we have These you know over the past year over 5,000 learners at our workshop Those people are going back to their departments and they're acting as models for these skills in their departments So they can now start to establish Some of the longer-term cultural change within their institutions so by working with the people who are already eager to learn these things and Essentially embedding them within their departments. We can have a more lasting impact than just on the people at our workshops on the third principle of cultural change that I'm going to talk about here is that it especially in educational cultural change one of the most common models is to try to develop a perfect curriculum and then disseminate it and hope that people adopt it and use it exactly as written and We know that that doesn't work there have been studies that have followed up on how many people use those curricula and whether they Get changed and the answer is that they they don't get disseminated as is and so And this is for good reason right so on the one hand Every educational experience is different So there's no one-size-fits-all solution every classroom is different And even if it wasn't the case that every classroom is different It's definitely the case that every instructor thinks their classroom is unique and special so Even if it was possible to develop the perfect curriculum You know ahead of time that it's not going to be adopted as is and kept the same And so the way that we at data and software carpentry work with this principle of cultural change is that our Our curricular materials are developed by the community and they're developed collaboratively and they change constantly as people think of improvements and as people as people update them so We have this Community develop set of curriculum for teaching the skills that we teach and The trick here is that if you're going to Have your community develop your curriculum you want to make sure That your curriculum is Following best practices so that it's using good principles of lesson design and the way that we help ensure that our curriculum follows good practices of Lesson design is by training all of our instructors who are working on the materials in those practices So we run instructor training programs. These are two-day intensive hands-on workshops that some of you have been to where we teach Evidence-based teaching practices and we teach curriculum design principles And then those instructors are the ones who are going in and updating and improving on the curriculum And so again, we're building up this community who can then carry forward those principles of cultural change and We have run over 40 training events for instructors over the past year And have certified 400 plus new instructors just in the past year So you can see again that we have this Expansive body of people who are excited and who are willing to donate their time to the effort of helping other Researchers not have to hit that data pain wall. So we've talked with our instructors and asked them Why are you so eager to volunteer your time for this and that most common answer that we get back is that they went through They went through hell trying to analyze their data in graduate school And they want to help other people not have to go through that problem So they want to teach others how to learn these skills So this is really a grassroots training effort It's developed by practitioners for practitioners. So our instructors are the people who are having these data analysis issues and who are trying to Teach other people how to learn data skills And so I talked about that our lessons are collaboratively developed. They're also Licensed CC BY so anyone can use and adapt them for any purpose and The primary Deliverable of data and software carpentry are these workshops that we organize and deliver So our volunteer instructors work to teach two-day hands-on workshops for for our learners on the curriculum on our data skills curriculum and The and the instructors will often have a pool of local helpers Who are people who also want to help? Our learners develop these skills Okay, so a little bit about the workshops. I talked about their their two days Which is what we've found is about the amount of time that most people have in their schedules to dedicate to An intensive activity of learning these tasks. We know that you can't teach all of this stuff in two days so what we focus on is developing in learners the foundational groundwork for being able to tackle these skills and we try to teach Self-directed learning skills. So we teach them how to read documentation. We teach them how to Decode error messages. We teach them how to search for help files And these are really the things that you end up doing the most as you move forward in developing data skills So we're teaching learners how to be life-long self-directed Learners and we're giving them a set of basic vocabulary and concepts for approaching these skills one of the most powerful pieces of technology that we use in our workshops are These sticky notes. So I just want to give a real quick intro to sticky notes Because I think these are the one of the best things we do in our workshops So raise your hand if you've been at one of our workshops and use these sticky notes. Yeah Raise your hand if you think they're the best thing ever Yeah, okay, but most every most everybody So I have a blog post on here all about how we use sticky notes that encourage you all to go read It's the best blog post ever. I wrote it So what we do is we go into our workshops and all of our learners are using their own machines to To follow along with the material as the instructor is going through so we use our principle called live coding Which means that the students are doing the tasks as they're learning about the tasks They're not getting a lecture on how to do data analysis or data cleaning They're actually doing it in real time and when you have people doing things in real time, of course, you have People running into problems in real time, right? So we have at the beginning of each workshop we give each learner two different colored post-it notes Usually they're red and green, but I'll pretend this is red. So this is my I'm having trouble sticky note And this is my everything is a okay sticky note and the way that these work and the reason I think that they're so Powerful for being so simple is that you've got a room full of students on computers If I'm typing away, and I'm trying to follow along with the instructor, but I'm having trouble I don't have to sit here with my hand raised while I try to type right so I can signal to the instructor or to one of the helpers that I'm falling behind well still maintaining my ability to Keep up with the material and without having to signal to everyone else in the room that I'm lost Okay, so it's a very Low Stress way of signaling to the instructor and these get rave reviews all of our workshops So that's my ode to the post-it note that encourage everyone to read and Again, we can't teach everything in two days But the goal at the at the end of our workshops is to have people who Have come in knowing that these things are important for them to learn But not really knowing what it is that's important for them to learn And then when they come out we want them to have a basic vocabulary and a basic set of concepts that they can build on along with a set of Skills for troubleshooting the learning process So one of the things that I find is so powerful about the data and software carpentry model Is that not only are we producing advocates in our learners and our instructors for data skills? but we're also Providing advocates for evidence-based teaching practices, which is the thing that I worked with in my postdoc and the other part of cultural change that I've worked with so Anyone who's in a university setting and who's had to interact with your Department of religion called Center for Teaching and Learning Anyone who's interacted with the Center for Teaching and Learning will know that it's very difficult to try to Change Teaching practices regardless of how much evidence we have about how we should be teaching And so one of the things that I'm really excited about working with data carpentry on is that we now have this body of 400 instructors who were trained this year Who now know good pedagogical teaching practices and know how to develop curriculum that matches what we know about how Learning works and they're going forward and they're being advocates for this in their departments And they're going forward and teaching in this way in their classes So this is the double impact that I think that data carpentry is having So I want to end just a little bit by talking about Our community so I talked about our instructors And our instructors teach workshops and they help develop the curriculum But we also have other segments of our community that overlaps with the instructors and perform very important roles That help support our mission. So trainers train new instructors And we have kind of a train the trainer model because we're scaling up so fast Like I said, we had 400 new instructors trained this year in order to spread that model and run more workshops We need to empower Instructor trainers to teach instructors how to teach our workshops that helps us scale We also have a mentoring program where experienced instructors help new instructors Learn how to be comfortable teaching make sure that they're up to date on the curriculum and Support them as they plan and deliver their workshops. So this mentoring program is relatively new We started it just a couple of months ago But the amount of support that we've had in the community and the number of people we've had signing up for this really shows that the carpentry community is full of people who are eager to help Support each other and I think that's a really good sign for the strength of our community moving forward We also have dedicated maintainers who manage our curriculum and make sure that all of these contributions coming in from our instructor base Work within the lesson and on top of that We have the new lesson infrastructure group and an assessment network that helps to assess the impact of our workshops So I'll end just by talking a little bit about Where we're going as an organization in the future. So one of the things that we're looking to Move forward on is providing more support for our learners after workshops right now. We have Learners will go to the workshops for two days and then they'll be done with the workshop And they'll be very excited about applying their skills But we don't necessarily in most cases have a local community where they can support each other and practice those Skills after the workshops have ended. We do have a few places where local communities have sprung up relatively spontaneously and Well, they'll have Dedicated help sessions every week for people who've gone to these workshops and help really develop their skills And build their skill set. So that's one thing that we're interested in working towards moving forward is building up these local Chackers and supporting our learners another thing that we're working on now And I talked a little bit about our new mentorship program is helping continue to support our instructors after they've gone through instructor training to Develop their teaching skills and to develop their their skills and working with the curriculum We also are working towards expanding our disciplinary reach data carpentry Workshops right now primarily serve people in the biological sciences. We have curriculum that are dedicated to ecology and genomics We have a new curriculum in geospatial sciences and we are now working on Developing curriculum in social sciences And there's that offshoot called library carpentry that some of you may have heard of that's involved in supporting Digital digital humanities and librarians. So that's something that we're working on in going into the future And we're also hoping to grow in our underrepresented regions. So data carpentry has large populations of instructors in Australia in North America and in Europe and we're growing in South Africa And also in Latin America in Brazil So we're hoping to continue to grow up in those in those underserved regions as we move forward Quick shout out to our steering committees. We just had a new steering committee come on board for software carpentry So those people all just started within the last month and they're very energized and ready to work with our community And then a little bit about our financial support and then with that I know I only took half an hour of my 50 minutes But I'm sure none of you mind that spending a little more time on questions rather than me talking at you So thank you, and I'd be happy to have questions