Featured Guest: David Keyes PhD
About the Featured Guest: David Keyes is the founder and CEO of R for the Rest of Us, which offers online courses, workshops, and custom training sessions. He has a PhD in anthropology from UC San Diego, as well as a master’s degree in education from Ohio State, and has dedicated his professional life to teaching people to embrace R.
Note: After Your PhD was not paid for this interview. Below, affiliate links to places like Amazon Marketplace support After Your PhD and keep the site going.
Affiliate Partners
Support After Your PhD by using one of these affiliate partners:
R for the Rest of Us: Featuring David Keyes Transcript
Transcript created with Vizard.AI. Therefore, you may come across some grammar or spelling errors.
Ryan Collins PhD: Thanks for taking the time to come speak with me today. I was really interested to talk to you mainly because of two reasons. One, because you completed a PhD, so you basically left the path of academia and done something else. And that something else is something called R for the Rest of Us. And I would love to hear more about your background and what led you to R for the Rest of Us, mainly from the journey after academia, if you could just kind of give a little overview of that. Sure. Yeah, so let’s see. Actually, I’m thinking back and I’m realizing it’s not totally cut and dry because I ended up doing my PhD. I finished it in, I think it was 2014.
David Keyes PhD: I went to uc san Diego, did a phd in anthropology, but we actually left, my wife and i left san Diego, uh, in 2013 to move to portland where we live now. And I was actually teaching for that year at a small liberal arts college here called lewis and Clark. Um, so i hadn’t actually, the reason i mentioned that because i hadn’t actually, uh, finished my phd at that point. I was still still working on it. Um, yeah. But yeah, so it wasn’t super direct. I mean, also, I never thought that I would be doing R, you know, like teaching people to use R. That was never the plan. But after I did a year at Lewis and Clark and then finished my PhD, I worked at a foundation in Portland called the Oregon Community Foundation, worked in their research and evaluation department doing research.helping the foundation to measure how effective the grant making they did was. And at that point, I mean, I know you’re going to ask how much I used R in grad school. The funny answer is never. Everybody assumes I did, but I actually never did in grad school because I did an entirely qualitative PhD dissertation and I worked on some mixed methods projects, but those were with other people who were using R, so I never had to. But so at the Oregon Community Foundation, I was just using Excel and like got pretty good at Excel. And then I left that job and I was working as a consultant doing what’s called evaluation work, which is basically like applied social science research for nonprofit education government and was using Excel there and just kind of got tired of it.
and wanted to teach myself something. So taught myself R. Absolutely.
Ryan Collins PhD: During that transition, were there any like motivators that made you want to kind of pursue something else? And what did you learn during that transition?
David Keyes PhD: Yeah. I was a little fed up with academia, to be honest, as I’m sure is pretty common. So I had a visiting professorship or yeah, I was visiting faculty for a year and then they hired a tenure track person. I applied for it, didn’t get it. And I just could see like the job market was not that good, especially like in anthropology, not that good.
And my wife and I had moved to Portland and decided we wanted to stay here.
So at that point, it just didn’t really make sense to try to continue in academic careers, especially because I was a little bit kind of ambivalent about trying to stay in academia anyway. So yeah, it just kind of made sense to look for something else at that point. Yeah, I think that’s pretty common with a lot of other PhDs I’ve talked to. Yeah, and so before we actually kind of get more discussion about art and its uses, I was wondering if you could maybe explain just in like how you understand it, what R is for people unfamiliar, maybe people who are like qualitative primarily and kind of interested maybe in using it. Yep. Yeah, so R, I mean, it’s a programming language, but typically I refer to it, I explain it to people as like kind of high powered Excel, which honestly probably sells it short, but I like to find something that people can kind of, you know, grasp onto that they’re already familiar with.
So it was developed originally as a language to do statistical analysis. I very much do not do statistical analysis. I’m a qualitative researcher by training. I mean, I do like the basic in terms of stats when it comes to our, you know, calculating like percent, all descriptive statistics, percents, counts, you know, that kind of means, medians, that type of thing. Nothing more complicated than that. And it’s a little hard to explain to people if they’ve never used it. But what I try to tell people is that R in many ways actually is a workflow tool, at least for me, more than it is a tool for any kind of like hardcore quantitative analysis. I talk a lot about this idea of reproducible reporting.
So you can use R has tools called R Markdown or Quarto, which enable you to kind of combine text alongside code to be able to produce documents really quickly. So the contrast that I give is if you’ve ever done any kind of work where you’re getting your data, making charts in Excel, say, then you copy them into Word, and then you have to kind of like redo it every time you want to do a new report, R allows you to write the code for that one time whenever you have new data, you just rerun your code and you make your reports. When I talk about it as a workflow tool, that’s largely what I’m referring to.
Ryan Collins PhD: Is that primarily what you like about ours, this workflow?
Because I’m curious if you tried other programming languages like Python, do you see any major differences between the two? Because I know the data types can be a little different, the actual jargon and that type of code that you use is a little different. I’m just kind of curious, like, what sticks out to you? Why R is better than what anything else have used?
David Keyes PhD: Yeah, I’ve never used Python. People always ask me that. And I’m like, sorry, I’ve just never used it. I’ve never felt the need to. From what I’ve heard, though, R is by far the most accessible for kind of non-programmer types. So I don’t know if you if that If you would agree with that. I don’t know. Have you used Python?
Ryan Collins PhD: I’ve used both. Yeah, I’ve used R in grad school. It was like political science courses. So yeah, it was a lot of descriptive and some inferential stats. But yeah, I did a lot of database. I always kind of like the… What is the package called that people always use in R? It’s ggplot. Yeah, ggplot. Yeah, I use a lot of ggplot. Yeah, I’ve used Python, but… I feel like I use Python more, but I think it’s mainly just because I see a lot of other industry folks use it. Yeah. But I think it really kind of depends your specialty, too. Like, if you’re specifically in data science or UX, it seems like they sometimes lean towards R. A lot of SEO folks I talk to who are interested in, like, more data type of things are using Python.
David Keyes PhD:: So I think it really just depends, like, your specialty. Yeah. Yeah. And I mean, the types of folks that I tend to work with are, I mean, some of them are in academia or like coming out of academia. And I think R is again, like relatively easy. I shouldn’t say easy, relatively straightforward in terms of a language to pick up. Um, there’s a thing called the tidy verse, which is like an approach for working within R that I think makes it much simpler than, um, what’s called base R. I’ve also just heard with Python, like, in terms of, like, package management and versions, all that stuff. That’s actually… Yeah, I just heard people talk about that as a nightmare in Python. That’s, like, not really an issue in R for the most part.
Yeah, I will say it’s, like, easier to kind of start up, especially with, like, RStudio to, like, kind of just get started with Python. You got to get your environment set up and everything else, so… Yeah, I can definitely see that. And maybe you can speak to this too.
Ryan Collins PhD: Why do you think R is popular in academia? Is there something historical about it? Or is it just the workflow that you mentioned? I’m just kind of curious in your thoughts on that.
David Keyes PhD:: Yeah, I mean, I don’t know totally. Like I said, I never actually used it when I was in academia. But I know it was a language that was developed by academics. And so… You know, just thinking about like the kind of networks where it’s spread, it would make sense that it’s spread within other academics.
That’s my guess, but I don’t actually really kind of know for sure. Yeah, I’m just curious. And so after you kind of learned R, you found all these different use cases.
Ryan Collins PhD: My understanding is you eventually created a website, R for the Rest of Us. I’m assuming there’s some consulting elements in play, also a course as well that maybe someone new to R could take. Can you maybe speak more to the website and how it eventually translated into a book?
David Keyes PhD:: I started R for the Rest of Us in 2019. Initially, it was just doing education. So, um, I was developing online courses, helping people like starting out with our, um, actually at that time, I mean, that was right before COVID. I did some in-person workshops like that fall. I flew around to various cities in the US and Canada and did in-person workshops. Um, but obviously then COVID hit and that was the end of in-person workshops for a while. So yeah, then it, I started a thing called R in three months, which is sort of like a bootcamp for people looking to learn R. That’s a big thing that we do now. It happens twice, twice a year, once in the spring, once in the fall. And then, you know, over the years I had been thinking like, maybe I should write, you know, write a book. And it was about two and a half years ago that actually, well, three years ago when I started getting serious about it, started reaching out to some publishers, seeing like what it might look like.
Well, I even thought about self-publishing for a number of reasons. I decided to go with a traditional publisher and then worked on it for two and a half years. And yeah, it just came out last month. So the book hopefully kind of builds on the, the, Courses, you know, it’s a different format. Obviously some the courses are mostly video based. So some people like that, some people want the book, um, you know, more like the more just reading. So, um, yeah, hopefully they can both help people.
Ryan Collins PhD: I do wonder, in terms of, like, a book, like, do you have a specific audience, just people completely new to art? Maybe, is there an academic audience in mind? Is there, like, a professional audience in mind?
Maybe a hobbyist? I’m just kind of curious about that.
David Keyes PhD:: Yeah, I wouldn’t necessarily, yeah, I wouldn’t necessarily say, like, one kind of segment, because I don’t, Initially, when I started doing this, I mostly actually worked with people in evaluation because that’s the world that I had worked in immediately before starting R for the Rest of Us. But at this point, it’s super like the people that I work with are in all different industries. So it’s intended to be broad in that sense. I tell people that the book is intended either for people totally new to R people with some experience in R who want to learn to use R in new ways. Because what I’ve realized, I used to have a lot of insecurity about the way that I use R because I was like, I don’t do any kind of inferential statistics.
I just do descriptive stats. But then I realized a lot of the things that I do, like I do a lot of data visualization. That’s unique. they’re not unique, but like there aren’t, you know, a lot of people who are doing hardcore stats and are not as gifted when it comes to the database side of things. So, you know, I hope the book can show people, like the idea behind the book is to highlight a range of things that R can do and help people think about it as more than just a tool for statistics. So again, I hope existing R users will pick it up and see new ways that they might consider using R.
Ryan Collins PhD: Yeah, because I kind of see it as, like, about how to, like, do storytelling, but, like, with a different tool rather than just maybe using Excel.
Because I do know, like, with ggplot, there’s a lot of, like, pretty interesting, like, different types of, like, just designs that you can use within it.
Ryan Collins PhD: So there’s, like, you definitely can see, like, a way of just storytelling. And I know, like, in terms of, like, for career transitioning in terms of people that want to use R for like a career, maybe outside of academia or just people that never been in academia. But to be successful, like, are there any other hard skills that you would suggest that pair well with R? I’m kind of assuming maybe like Excel would be a good tool to learn, but are there anything else you’d recommend?
David Keyes PhD:: I don’t know. I mean, I do think like the thing with learning R is you have to,of kind of an understanding of quantitative data and how it works. You know, for example, when I teach people to use R, I assume that they know, you know, different data types. So like, is this data in this column numeric or is it character or whatever? And so, you know, if you are thinking about R, but you’re like, oh, I’ve never done anything in that area, I would definitely think, let me get kind of up to, up to speed on just kind of general skills working with quantitative data. And I think Excel can be a good bridge too. I mean, that was my bridge. That’s how I got to R. But I also think, I mean, you know, this is anecdotal, but I’ve definitely talked to a lot of people who have learned R and it’s really helped their careers just in terms of things that it’s opened up for them.
So, yeah, I don’t know about so much necessarily about complementary skills, but I will say, I mean, this is obviously, I have a self-interest in this, but I think learning R has a ton of potential benefit for people. And like, do you feel that R is like, I know you talk about just like how you educate others about it, but let’s say you were like, you just learned R, not just learned R, but you learned R and then you join a team of people that never used it before. Do you think it’s easier to translate your knowledge to other people? Do you think it’s a tool that will work well in collaborative spaces that eventually people will learn? Because I know with Excel, it’s just everyone already knows Excel.
Ryan Collins PhD: how do you spread R in different workplaces that people may not be familiar with? And maybe there isn’t any. Maybe I’m wrong there. But I’m just curious how you spread R to different people. I mean, I’ve seen that. Right. I work with several teams who have transitioned to R. And I think there’s some nice ways that I’ve seen some of them do this. So some of them will have like a, I don’t know, bi-weekly meeting where, you know, either in person or virtually, the idea is like any R questions that you have, you bring them to that meeting and discuss them. And having someone who’s more advanced is really helpful there because someone less advanced might be like, hey, I’m trying to make this plot and I can’t figure out what I’m doing wrong.
And that more advanced person then can help them. So I think setting up structures like that is really helpful. I also think as much as possible early on, working with real data is really important. like in this R in three months program we do, the teaching materials obviously have to use a generic data set because it has to be something that everybody can work with. But we also then have a piece of it where people are working with their own data and doing assignments with their own data. And I think it’s just so much easier for other people kind of to really see the power of R if they’re doing things that they doing something with their own data. The one other thing, this might be a bit more advanced than what you’re thinking, but we work with some teams.
We’ve made custom R packages. So for example, we’ve made packages for teams that have what’s called a ggplot theme. So that’s just one line of code that they can apply to all of their plots, and it makes them all look on brand. and giving people, what’s that? Yeah, that’s pretty neat. And I was gonna say, I don’t know if you define maybe what a package is, but- Sure. So yeah, I mean, a package is, because R is open source, you can, there’s like the built-in tools of R and people refer to that as base R and that’s like the core code that makes up R. But again, because it’s open source, people can write their own code to have R do any number of things.
And when people write code and kind of package it up, that’s what’s called a package. And there are packages on what’s called CRAN, which is the official place where you can most easily access packages that other people have made. So I talked before about the Tidyverse, which is a collection of packages used for data import, wrangling, data viz. But then anybody can make a package. And I actually have a chapter in my book that talks about making your own package. And we’ve seen teams that have made their own packages. And having things like a custom theme enables less advanced users to pretty easily apply that theme and make all their plots on brand. in ways that, you know, if they had to write that code from scratch, like there’s no way they would, but if they can just write, you know, theme omni to give the name of one of our clients, then it’s pretty straightforward.
Ryan Collins PhD: So if people are interested about learning more about art, either through the website or your book, firstly, like how would they even find this uh are you is the r for the rest of us selling anywhere? Is it on amazon uh somewhere outside of amazon Yeah, it’s on Amazon.
David Keyes PhD:: You can buy it directly from the publisher. They’re called No Starch. There’s also an open, a free online version of the book. If you go to book.artfortherestofus.com, you can actually read it for free. It also has links at the top of every page if people want to buy an actual copy. But yeah, it’s available widely. Yeah.
Ryan Collins PhD: Well, I appreciate you for taking the time again. And I just want to leave the floor open just for like a final question of do you have anything else maybe that we missed?Anything maybe about leaving academia or just your journey in general for those that are interested about like careers outside of academia? Maybe something you could speak to that on.
David Keyes PhD:: Yeah, I will say one thing that people often ask me because now I work for myself and people are like, oh, I want to, you know, I’d love to, get out of academia and work for myself. But I think usually you have to go through an intermediate step of working for someone else. I wouldn’t have even known about R had it not been for the last two jobs that I had before I did this. So while it might be tempting to think, oh, let me just start my own thing. In a lot of cases, I think getting ayou know, regular nine-to-five jobs, seeing what some of the pain points are can really be useful if you do want to do your own thing long-term.
Ryan Collins PhD: Yeah, no, I definitely agree with that. Yeah, I feel like you really just have to learn by observing, and I think you can do that certainly by, like, just seeing how organizations work. So, yeah, especially in my field, like SEO, like, you can’t, like, really consult people if you haven’t really worked with the business to actually do that. So that makes total sense. So, yeah, I’m going to let you go. But again, thank you for taking the time to talk with me on a with after PhD. So, yeah, thank you.
Ryan Collins PhD is an SEO Strategist at Go Fish Digital. Ryan completed his PhD in Media Arts and Sciences at Indiana University Bloomington in 2021. During his time at Indiana University, Ryan eventually pivoted into a career in SEO and Digital Marketing after having informational interviews with working professionals in SEO, working on side projects, and gaining industry experience.