Lauren Burke 00:08
Welcome to episode three of Women in Analytics After Hours, the podcast where we hang out and learn with the WIA community. Each episode, we sit down with women in the data and analytics space to talk about what they do, how they got there and where they found analytics along the way. I'm your host, Lauren Burke, and I'd like to thank you for joining us today.
Today, I'm very excited to have Meghan Hall joining us. Meghan is a data analyst and R educator who has spent over seven years as a data professional in higher education. Meghan is also a public contributor to the sports analytics community. She was previously a data scientist at Zelus Analytics, a company that provides sports intelligence as a service through its innovative platform.
Meghan is also a regular contributor to Hockey Graphs, which focuses on the visualization and analysis of hockey data. Meghan is a self-taught R programmer and is very dedicated to R education, particularly for beginners. She frequently delivers talks, workshops, and classes on the topic. So welcome, Meghan. I'm so excited to have you here with us today, and I really appreciate you taking the time to join us.
Meghan Hall 01:21
I'm so excited. Thanks so much for having me, Lauren.
Lauren Burke 01:24
Absolutely. So you have a super interesting background and while I'm not too familiar with sports analytics, especially hockey analytics, I was born and raised in the home of the Blue Jackets. So I definitely grew up going to a lot of those games. My dad is a big fan, so, it's a really interesting topic.
And I've taken a look at some of the work you've done, and I'm looking forward to learning a little bit more about it from you today, as well as how sports analytics and the open source community kind of come together to make all of those interesting analyses happen.
Meghan Hall 01:56
Yeah, actually, um, the only time I've been to Columbus, it was actually my last travel right before the pandemic started. The blue jackets hosted a hockey analytics conference. So it was the only time I've been to Columbus, it was a great town. So Columbus and hockey analytics are intertwined in my life.
Lauren Burke 02:12
Yeah, I saw that conference. From people that I know that attended that are more into the sports analytics space, I heard it was a really excellent conference and I'm hoping they are able to put that on in future years or maybe even bring in some other local sports teams, maybe the Columbus crew or Columbus Clippers. We have a lot of different variety of sports and potential sports analytics topics in the area.
Meghan Hall 02:39
I would definitely, I would make the journey back.
Lauren Burke 02:41
Awesome. So to start off, I'd like to ask, how did you get started with this area of analytics? Did you have a, uh, existing interest in a certain sport or certain area?
Meghan Hall 02:53
Growing up I was always a baseball fan and baseball has kind of traditionally been at the, you know, cutting edge of at least analytics in the sports world. So just kind of by being a baseball fan and by being, you know, a data person, a nerd, whatever you wanna call it. I was always interested in the numbers side of the game. And then a few years ago I started getting, I started getting into hockey and again, just because of my interest in numbers, I pretty naturally gravitated toward being interested in the analytics side of the game and what data's available and what people do both publicly and privately with that kind of data.
And especially compared to baseball and especially a few years ago, like hockey analytics was way I don't wanna say behind, but just in a much like earlier start in the journey. So there was really a lot of room for, there was still a lot of interesting public work being done, which is in contrast to some other sports, particularly in baseball.
Not that there isn't still quality public work being done in baseball, but the data is so advanced that most of the cutting edge stuff is happening privately. But with hockey, again, there was still so much unexplored territory that there was a pretty active public sports analytics community.
And so just kind of naturally, you know, was interested in data, was interested in some questions. And then realized that the data was available and I could do my own analysis if I wanted.
Lauren Burke 04:09
That's awesome. So did you find analytics first or did you find sports analytics first and come to analytics through that?
Meghan Hall 04:17
To go back, I went to grad school. Actually, I have an MS in Nutrition with a Specialization in Nutritional Epidemiology. And while I was in grad school, you know, I took a few stats classes. I also worked as a research assistant in a research center and really realized that I wasn't particularly interested in pursuing my Ph.D. in that specific area, but I was just really interested in working with data in general.
And so when I left grad school with my MS, I knew I just wanted to work in data in some way, like kind of no matter the domain. And I ended up working in higher ed. I still work at Brown University. That was the side of my first data job, um, over seven years ago now.
And so I got into data first. I was a data analyst for several years. But then really moving into getting interested in sports analytics, kind of on the side was really kind of what advanced, I would say my analytics career, both professionally and I guess, you know, personally outside of work, the stuff I was doing publicly.
Lauren Burke 05:17
I always find it really interesting how people found their interest in analytics or data. I think it's, it's a really unique field because a lot of other areas and industries use data. Like you said, with your epidemiology and nutrition background, you were probably already working with data and doing analysis. Might not have considered it data analytics or data science, right. It was just the data you needed to work with.
Meghan Hall 05:43
Exactly. Yeah. And so, and then I kind of realized that, oh, I could use these data skills and work in, you know, a number of fields, um, that were, that were interesting to me, which was a surprise because growing up, I unfortunately, you know, spent many years like avoiding math, not thinking I was good at math.
And like, I wish I had some, you know, origin story for that disbelief, but I don't really. That's just kind of, you know, how I grew up. And so I'm really sad about that, that I avoided math classes and such in college and kind of got out of all that I could. So it's really surprising to me that I ended up gravitating toward, I don't do a lot of math, but I do work a lot with, you know, numbers and data. So it has definitely been kind of a surprise, but a very welcome one.
Lauren Burke 06:26
That's cool. I, so I actually studied math, so I'm always sad when people say that they didn't like, or they thought they wouldn't be good at math because I think math once you really get into it, especially if you're interested in data and analytics, it helps you put the pieces together and.
Meghan Hall 06:41
I know it, it really makes me sad. And like I said, I wish I could pinpoint kind of where that came from. But even one of the luxuries of working in higher ed is you can often take courses. And actually a couple years ago I took like college level calculus just because I wanted to, because just to like prove to myself that like, yes, you might be, you know, 29 years old at the time, I think. But you can do college math and I did pass.
Lauren Burke 07:05
That's great. I, I love that you did that. Sometimes you need to do things like that just for yourself, so you are able to give yourself that confirmation that you can do these things that you might have thought were hard, but you're capable.
Meghan Hall 07:16
Yeah, and I think, and it's a, and it's a common misconception. I mean, you can get pretty far in working with data without, you know, a ton of advanced math knowledge.
Obviously, I mean, you might hit limits if you're interested in like advanced machine learning and stuff, but there's still a lot of data work you can do without, you know, having a deep calculus background.
But I was still, I really wanted to check that off. So I was really happy that I was able to do that even just again, for myself.
Lauren Burke 07:40
That's so awesome. I love that. So throughout your career, I read it somewhere that you actually transitioned from using Excel to using R for your analysis. Was there anything in particular that drove that transition?
Meghan Hall 07:54
Yes, this is actually, this little origin story is actually the beginning of a talk I'm giving at the upcoming R conference, but I will repeat it here because it's very true. But yes, when I started Higher ed. I was working as a data analyst, not in like, I didn't work in IT. I didn't work on a data science team.
I was really just the kind of data person in like a functional business office and those offices use Excel. So that's what people around me were using. And that's what I was using for the beginning of my analytics career. Which is fine again, you know, Excel is a great program that has lots of uses, but then when I started getting in hockey analytics.
Literally the data files, I could not work with them in Excel on my old laptop. Like they had like a million rows and my computer just like, was not into handling Excel files of that size. So truly out of necessity, I started using R.
Cuz I did have some stats program background from grad school. I had learned R for one class and some other class I had to use SAS. I had to use STATA back in undergrad. So I had some familiarity with programming and then just decided pretty quickly to pursue R because it just had, you know, the most welcoming community. It seemed the most learnable to me at that time. So it was truly out of necessity that I learned R. But again, really, you know, changed my career.
Lauren Burke 09:11
That's awesome. I definitely understand some of those reasons for Excel. From what I saw with hockey data in particular, it's broken down in pretty small time segmentations, right by millisecond.
Meghan Hall 09:26
Yeah, the data that is publicly available. Privately there's some new tracking data that's um, that's definitely what you would classify as quote unquote, big data, but at least the publicly available data, if you think of like one hockey game probably has like 800 records, or rows in an Excel sheet, you know?
So if you're dealing with a full season's worth of data or even multiple seasons you're getting to file sizes that again, at least my little MacBook Air at the time did not like dealing with in Excel.
Lauren Burke 09:56
Right. That's also so awesome that there are so many public data sets and open source like packages and communities in the sports analytics space, that are out there for people that are interested in that, and probably have a lot of good resources for people who are getting started.
Meghan Hall 10:15
Yeah, it's kind of a natural starting point. I think for a lot of people, again, people like me who are interested in sports, who are interested in the number side. Particularly also, I've seen it very popular with students who are wanting to learn more about analytics. Because it's such a, you know, just like a cultural touchstone, it's pretty widespread and there is lots of public data available now, which is nice. It's a really nice learning opportunity.
Lauren Burke 10:38
For someone that is interested in developing a career in sports analytics, do you think those public and open source communities are a good starting point?
Meghan Hall 10:48
Yeah, absolutely. That's a question that I get a lot or anyone who has worked in sports, worked tangentially with sports is like, how do I get into sports analytics? And truly like the number one answer is to do work in public. And the work doesn't even have to be sports-specific necessarily to the sport that you're interested in.
Again, all of the public sports work I have done has been with hockey. And then I ended up actually working for a while as a data scientist at Zelus Analytics. And there I was working with, with baseball data in particular. So a lot of it crosses over sports, just kind of general knowledge. And yeah, absolutely, there are so many data sources available for really whatever sport you're interested in it's true that, you know, the quality of the data definitely varies by sport, but you know, the NFL has really fantastic data. There's a really big community of people who have done a lot of work to develop that data for open source use. And really, doing any kind of projects with open source data, presenting those projects publicly, whether you go to sports analytics conferences, or even if you just, again, have an online presence, have a website or GitHub page or anything is really how most people, not everyone, but most people who end up working in sports analytics, that's kind of how they got discovered.
Lauren Burke 12:05
I think that's a great point, not just for sports analytics, but for a lot of data analytics, specializations, or even just technology fields in general. That you are being able to contribute to open source allows you to put your work out there in a way that opportunities might come to you more than you having to be the one to seek them out.
So I think that's a great suggestion for people that are interested in that potential sports analytics area for a career.
Meghan Hall 12:34
Exactly and I mean both the opportunity I had at Zelus and other opportunities to work in sports that have come to me, but that I haven't pursued for various reasons. You know, all of those came to me just because of my public work. So that's definitely what I would recommend certainly for sports, because that's what I know.
But like you said, it's very applicable across a lot of data and technology-related fields. Which can, you know, on one hand, it does level the playing field. Because all this data is available, internet is open, but it's also, you know, important to recognize that not everyone has the privilege of time to do all of these extra open source projects.
So there needs to be a balance, but it is nice that there is public data available and it is a really great way to showcase your skills.
Lauren Burke 13:18
Right. I like that you mentioned that. I think that's a good point to understand that not everyone has those same resources, but at least with the data out there publicly, people can potentially have the same starting point. So where you're starting is where everyone else is starting with that data. So if you might be new to the field or you're not sure, if you have what it takes to make it, you can at least start and try and do it on your own and see if you can put something together that is interesting and that you think you can make an impact.
Meghan Hall 13:48
Exactly.
Lauren Burke 13:50
So some of the public work that you've done with hockey analytics focuses on two particular topics, goalie pulling and power kills. And you mentioned at one point that that hadn't really been explored before in a deep analytical context. So you mentioned that when you were looking into this, you were trying to answer questions where the necessary information and data wasn't as readily available.
Can you talk a little bit more about how you went about that and how others might go about finding the data they need for less explored questions?
Meghan Hall 14:20
Yeah. I mean to start with right when I kind of started watching hockey. One of the first questions that was coming to me as I was learning the game.
Cuz to give, you know, very brief explanation is if a team is trailing at the end of the game, they often nearly always will pull their goal tenderer in exchange for an extra skater to hopefully, generate more offense and tie the game. But, of course, the downside of that is you don't have a goalie. So if you, you know, give up the puck to the other team, it's very easy for them to score a goal and confirm the win.
So that happens pretty confidently in hockey and for a long time, people, lots of mathematicians, much more advanced than me have tackled the problem of, you know, when teams should pull their goalie. And a lot of the research has shown that teams should actually mathematically be a lot more aggressive than they actually are.
And when I started watching hockey, I was curious if there had actually been any movement in that direction. Cuz mostly, all I saw was well the mathematicians say you should pull it this time, but when like has any of that knowledge actually made it into the teams. And I couldn't really find that anywhere, so I decided to make that kind of my first research project of interest. And there's certainly no need to get into the intricacies of NHL play-by-play data that was available at the time. but it wasn't so simple as, you know, the goalie is gone at this time. Like there was a lot of, kind of more data manipulation that, you know, I had to do to accurately identify those scenarios and then look at it over time.
So I could actually see when this was happening. So again, really great exercise, both because especially at the same time, you know, that's when I was learning R also. So kind of this big complex, not necessarily advanced, but just kind of a lot of complex data manipulation work. Not building models, not really doing the, certainly not doing like the machine learning of data analytics, but more of the data cleaning and data prep was also just a really great exercise.
And similar when I started looking at offense on the penalty kill, which teams that have a offensely aggressive penalty kill, it's often called the power kill. And similarly, the public data that's available in the NHL is not particularly detailed. And often I was interested in more specific things such as, you know, when teams were actually entering their offensive zone, how much time they were spending in their offensive zone, things like that aren't publicly available.
So I often had to, you know, literally watch games like rewatch games on my computer and track those sorts of things. Which is pretty common in hockey. Lots of people who do projects do end up having to do some kind of manual data collection, whether that's, you know, tracking when and where passes happen, or again, these zone entries, which again is just another, I mean if you're thinking about it in terms of the kind of career prospects, it shows that you're willing to kind of go the extra mile for your research question, which is always nice, but also just teaches you more helpful data collection technique.
Lauren Burke 17:19
Right. That is extremely manual data collection. Wow. And yeah, I'll definitely applaud your efforts for that. When I was looking into sports analytics, particularly hockey analytics, I came across a lot of data visualizations, some from yourself and some from a few others. And I felt like, through that, I was actually getting a bit of a better understanding of some of the intricacies of the game itself.
So do you think data visualization is useful in sports analytics, not just for understanding the data, but for helping others to understand the game itself?
Meghan Hall 17:51
Yes tremendously. I just am personally, you know, really interested in passionate about, fairly knowledgeable about data visualization. So, you know, definitely something that I care about, but particularly in sports, there's tons of great data visualization in sports, you know, very common in hockey are shot maps, which are just kind of heat maps that demonstrate either a full game or particular period during a game that show the quantity and the location of shots for each team. And really, it's pretty amazing just looking at, you know, the shot map for a game can give you a pretty good sense of what that game was like.
Even if you don't see any more box score stats. You don't even see a score. Like you could see that, you know, one team was applying a lot of pressure in front of the net and the other team wasn't getting many shots off, et cetera. And especially with the increase of tracking data, that's sometimes publicly available in, in the NFL for certain projects.
It's amazing, some of the data visualizations that people in football data are able to do with, again, the tracking data that you can create these interactive moving visualizations that really track like every player's movement on any given play, which is pretty neat. Which I hope someday we will get that with hockey. There is now tracking data in hockey. So far none of it has been made publicly available, but, you know, hopefully that would change in the future.
Lauren Burke 19:13
That's super interesting. I think just all of the different ways you're able to look at the data and then transform that into not just an analysis, but into that visualization, right. Where you are able to almost replicate that gameplay in a visual aspect with the data sort of layered on top of that is just super interesting and super useful for analysis and just as something that's really cool to see and really cool to look at and learn about.
Meghan Hall 19:40
Yeah, and lots of, you know, lots of good learning opportunities for figuring out how to map data when you're dealing with locations and fields and et cetera. So again, ripe with learning opportunities.
Lauren Burke 19:52
Definitely. Are there any other unique things that you've learned from your work in sports analytics that you've brought with you into your current role in higher ed?
Meghan Hall 20:02
I mean, certainly the like technical skills, again, just becoming extremely proficient. I would say that thanks to working with sports data, again, my experience with hockey data in particular. Hockey play-by-play data can be a bit messy. Again, there's often a lot of manipulations that you need to do in order to get data in the format and cleanliness level that you like.
And really just because of that, you know, I would say that my like true expertise in, in data is, data cleaning and data preparation, which is not really the sexy side of data analysis, but it's very important and I'm very good at it, thanks to dealing with year's worth of messy hockey data.
So I definitely have that, have that to thank.
Lauren Burke 20:46
I mean, I will never play down the merits of being really good at data cleaning, because I think it's a commonly said phrase for machine learning, but also applies to analysis like garbage in equals garbage out.
Meghan Hall 20:58
Exactly.
Lauren Burke 20:59
Yeah. If you, if you can have the data in a good starting point and you are really, really good at that, then I'm pretty confident in saying whatever you put out is usually a good analysis because you have the data in a good place that you can understand it. And, you know, however, you're trying to use it you're at least starting from a point where if you do it correctly, you'll be pretty successful.
Meghan Hall 21:21
Right. And even with just useful experience like working with, again, working so much with the hockey play-by-play data in particular. Doing several projects with it over several years. I just know a lot about that data in particular. Like not even like separate from knowing like the domain knowledge of knowing hockey, but I just know that data really well and like the common pitfalls and common things that happen.
And so all of those again, are really domain transferable. I would say just the experience of like getting really familiar and curious about a data set.
Lauren Burke 21:55
Absolutely. Yeah. I think that's such a good thing to mention about just data and analytics in general, is that as long as you understand the domain you're working in really, really well, then pretty much any technique you're trying to apply can be pretty similar across industry or across specializations, right?
But that the specific domain, you have to understand to know how you can best apply that technique.
Meghan Hall 22:21
Exactly.
Lauren Burke 22:23
You mentioned being self-taught or community-taught in R programming. So can you tell us a little more about the role that the R community played in your learning of the language?
Meghan Hall 22:35
Yeah. One of the reasons that I like once I knew that like, okay, Excel is not sufficient for this anymore. I need to find something else. R really stands out in the community, just because it has such a strong community. So many people love R and love sharing it. And of course, R is, you know, open source, free.
So there's that natural kind of component to it that it's always under development and people are sharing new developments in R, but again, even beyond that, it's a, just a community that I've always felt very comfortable in. You know, certainly as a woman, which is not always true in, you know, data and programming communities. But again, people are, so people are so willing to share their knowledge. There are so many free, like, great resources for learning R in like a variety of formats. And so I just pretty automatically felt really comfortable in that community.
Lauren Burke 23:26
That's great. And you yourself are a big proponent of R education. You've created a ton of helpful resources over the years. You've given talks, developed tutorials. So how did you approach beginning to teach R concepts yourself and make that transition from learner to instructor, especially maybe early on when you might have still felt newer to the space?
Meghan Hall 23:47
I did definitely early on. It kind of happened almost accidentally. Cuz again, I was learning R and I was applying it to the hockey work that I was doing, which I was presenting at conferences and just on my website and such. And I would, you know, be able to answer questions from people in the community about R.
And I ended up writing kind of a really brief, like intro to R kind of series of posts on hockey graphs, which is a site that I have written for in the past kind of focused on hockey analytics. And realized, cuz first I was like, oh, well I'm new, like how can I possibly, you know, teach anyone about R. But the beauty of being new is you are very connected to the beginner experience, right? Like what beginners want. Like even now when I'd say I'm a pretty decent and experienced R user, I still feel very connected to being a beginner because it wasn't that long ago.
And I really remember a lot of the pitfalls that I faced and things that. Didn't find in resources that I really wanted. So, you know, definitely had some strong opinions about how I think R should be taught. Especially from again, I'm always trying to help someone like me in the past, who is curious, who's had a little bit of experience with data, but would not call themselves proficient at all with any kind of coding or programming and kind of how to make that transition less scary.
Cuz I think a lot of people are just automatically like, oh I can't, I can't code, I can't program. I've never done that. But R really provides a nice kind of gentle pathway.
Lauren Burke 25:20
Right. I love that you are considering the beginners and the people that are just starting out, right. Because even if, and like you mentioned, even if you still feel pretty new, there's someone that is at the point, you were a year ago, five years ago, even 10 years ago. And they can benefit from knowing what you would've liked to learn there.
And so I love that you sort of are approaching this as you know your audience because you have been there and
Meghan Hall 25:49
Like I was just there.
Lauren Burke 25:51
Which is great, which is great. That's the best way to teach is when you are teaching to yourself in a way and expanding it to what others might need to know.
Meghan Hall 26:00
Yeah, cuz I mean, R has been, you know, tremendously helpful. Obviously, I've had lots of fun opportunities, like stemming from sports and also just stemming from general kind of R education stuff outside of work. And it's also tremendously leveled up my professional work. Over the time that I've learned R, you know, not that salary means everything, but, you know, my salary has almost doubled.
I have taken on a lot more, been able to take on a lot more interesting projects because I can streamline so much of the repeated reporting work that I have to do through R. And so it has really, you know, been a tremendous benefit to my life and that's even expanding beyond the community. You know, I've made tons of tons of friends because of R. I feel lucky to be, you know, a very small, tiny part of that community.
So I would, you know, like to share that with people as much as I can.
Lauren Burke 26:49
That's awesome. I think that's one of the best things about open source is that you can often find people who are just as passionate about teaching others and learning new things as you are. So it's, it's just a great space to be in if you're really excited about a certain topic or area.
Meghan Hall 27:05
Right, which has really been kind of a constant in my life ever since I have been small. Just really curious, like always wanting to learn new things and try new things. Again, even taking college calculus as a 29-year-old, just because I wanted to prove that I could. So I definitely feel, um, there's a lot of people who feel similarly in the R community and also just devoted to sharing that knowledge, which has been really nice.
Lauren Burke 27:29
Absolutely. And so speaking of speaking, you will actually be giving a talk at this year's RStudio conference. And you'll be speaking on building your own R ecosystem, which is a really useful topic because I feel like a lot of people are in the same boat. And you earlier, even today talked about how you were in that same boat.
So what inspired you to give this talk?
Meghan Hall 27:52
Yeah, I am. The time we're recording this, the RStudio conference is next week. So I'm getting very excited for my talk. I actually submitted, it took me a while, cuz first, when I saw the call for the abstracts go out earlier in the year, I was like, oh, like there's no way, I don't know enough about R to speak at the RStudio Conference.
And then I saw that they extended the deadline for speaking applications. And I was like, okay, this is a sign like you need to at least try, you know, if you don't try, the answer is always no. Um, so I submitted an abstract and was thrilled to get accepted. So I'm very excited to go. I've never been, I haven't been to the R conference. This will be my first kind of R-related conference.
I've been to tons of sports analytics conferences, but this is kind of my first more general analytics conference. I've been to the Tableau Conference a couple times actually, but I'm pretty excited, really excited for this.
But yeah, the main theme of my talk is how to start using R even if no one else in your professional vicinity is using R. Which, again, was a scenario that I found myself in where everyone I knew, because I wasn't working in it, wasn't working on some kind of data team. People were using Excel. And again, the benefits that I have found through my career and through my work in using R have been worth it, even though, again, there are a lot technical difficulties, which I get into in my talk. Of kind of the difficulties of inserting R into, you know, other workflows. But one of the beautiful things about R is even though the R ecosystem itself is so large, you can do and ideally you would do your entire data analysis pipeline in R, and you can do that, which is amazing.
But you can also just pick and choose pieces that work for you and that work for whatever, you know, technical stack that you're working with. Which is, again, one of the things that I think is really great about R, so that'll be the main kind of thesis of my talk.
Lauren Burke 29:49
That's awesome. I hope if anyone is attending RStudio Conference, or if you can check out the videos online afterward, you take a look at Meghan's talk, because I think it's a topic that a lot of people could benefit from, and it probably applies to not just R specifically, right.
Meghan Hall 30:04
Yeah Exactly. And I think actually the talks will be streamed live if you're watching virtually. So yeah, I would definitely recommend checking it out. My talk is on Thursday, the 28th in the afternoon, but again, there's so many great talks. One of the great things about talking at RStudio, actually is that they provide a bunch of speaker coaching.
So we kind of work with small cohorts over the past several weeks to develop our talks. So I've had a chance to kind of see some of the talks in progress and there are so many great talks and again, they can all be viewed for free if you will not be joining us in DC next week. So I would, again, highly recommend that people check those out.
Lauren Burke 30:41
That's awesome. And I believe the speaker coaching through RStudio is actually done through Articulation with Ruth Milligan.
Meghan Hall 30:47
Yes. Which is Columbus based.
Lauren Burke 30:50
Yeah, she's actually the speaker coach for our DataConnect Conference as well. So always willing to shout her out because she does amazing work and helps people prepare in ways that I didn't even know were possible.
Meghan Hall 31:03
Yeah, the coaches that we've been working with have been really amazing. Again, that's a really great perk. So I would recommend anyone, next year, you know, submit an abstract. It's great to be able to talk. And also you get, you know, some free bonus speaker coaching. So it's been very useful.
Lauren Burke 31:16
Awesome. So speaking of resources, this is the final question that I like to ask everyone. If you can tell us what is one resource that has helped you in your career and that you think might help others listening today?
Meghan Hall 31:30
It's not the most exciting answer, but probably a really popular one. But the R for Data Science book that Hadley wrote, that's freely available online has really been my favorite resource. I have it bookmarked. I go back to sections time and time again. Like literally just yesterday, I had to reread something in the factor section. That's truly, I think my favorite resource.
The book is mostly about the Tidyverse, the kind of collection of packages developed by RStudio to use with R, and I work a lot with the Tidyverse, both, personally, professionally, and also in my education work. Cause I think it's a really nice entry point for beginners to handle most of what they need to know about the beginnings of doing data analysis in R. And I really just think it's the best resource for that.
And the other one to give a quick bonus one, it's a little bit more specific. But anytime I'm working specifically in like R markdown or Quarto, and I wanna learn how to do something that I've seen other people do. Really just searching people's GitHub repos. I think is a very underrated learning technique. Again, the beauty of R, so much is open source. Most of the talks, presentations, et cetera, books. Even things that are online that are built through R, most of them have public GitHub repos. And again, you can search through all the code of those to find out the tricks of how people do what they need to do. So whenever I'm building slides or building websites, anything where I'm using R, searching repos has been again so useful for me.
Lauren Burke 33:07
I think those are both really great resources. I love with the first one, the R for Data Science book, you mentioned that you keep coming back to it. Which I think is the sign of a really good resource, right? If it's just something that it's your go-to for whenever you need something on that specific topic.
So we will definitely link both of those in our podcast link section. And finally, how can our listeners keep up with you?
Meghan Hall 33:31
Most easily I think is on Twitter. I am @MeghanMHall. And in my Twitter is linked to, you know, my website, which has links to various projects that I've done, my GitHub, all the talks I've given, et cetera.
Lauren Burke 33:44
Awesome. Well, I really enjoyed speaking with you, learning a little bit more about sports analytics about your journey from Excel to R, and now as a person that is a very big contributor to the R education community. So I'm so glad you were able to join us.
Meghan Hall 34:01
Likewise. Thank you so much for having me.
Lauren Burke 34:03
Absolutely.
.png)
