An inclusive solution for teaching and learning R during the COVID pandemic
The COVID pandemic has shaken our teaching and learning approaches in many different ways all over the world. Nonetheless, it has also provided opportunities for bringing creativity into the classroom. In this talk, I will discuss how I have used RStudio Cloud in my teaching during the pandemic and how I capitalized on the opportunities that RStudio Cloud offers to deal with the crucial issues of software installation. Introducing RStudio Cloud in the units has allowed me to work effectively in an online environment to engage, motivate and empower students through their learning process while removing the troubles and hurdles of software installation which is generally particularly challenging in first-year cohorts without prior coding experience. I used RStudio Cloud in a data science introductory unit at Monash University and as a tool to present the usage of R and RStudio for reproducible reporting in another unit on Reproducible and Collaborative Practises. In the latter, I introduced RStudio Cloud during the first few weeks to get the students up to speed before transitioning to using R and RStudio Cloud in their own local machines while using the command line interface, Git, and GitHub as a version control tool for reproducible reporting. I will also discuss how I organized and managed the unit's RStudio Cloud account so that my research associates were also an integral part of the unit delivery to ensure the success of the units. Read Dr. Menéndez's guest post on the RStudio Blog: https://www.rstudio.com/blog/rstudio-cloud-an-inclusive-solution-for-learning-r/ Read more in the follow-up blog post: https://www.rstudio.com/blog/teaching-data-science-in-the-cloud/
image: thumbnail.jpg
Transcript#
This transcript was generated automatically and may contain errors.
of our first RStudio Cloud live series. It's actually the first one here in 2022. So happy new year. Also happens to be my little girl's second birthday today. So a lot of excitement here at my household. Now, in terms of housekeeping, just like last time, we're going to start with a guest presenter for the first half of the session, and then we're going to follow with a live Q&A. Now, no need to wait till the presentation is over. If you have any questions along the way, please just send them on in and we'll ensure that they get addressed. Also, if you do have to hop off early, please don't don't worry, we will be recording this so you can watch it at your leisure. And at this point, I'd like to introduce today's guest, Dr. Patricia Menendez, all the way from Australia, sharing her experience at the Business Stats and Econ Department within Monash University.
Pete, thanks so much for having me here today. And good morning, good afternoon, or good evening, depending on where you are. Welcome. And thanks for joining us for this presentation. I'm very pleased to be here today. And I would like to thank Pete and the RStudio Cloud team for giving me the opportunity to share my experiences using RStudio Cloud with all of you this morning here in Australia. I'm also very much looking forward to have a conversation with you and to hear about your experiences using RStudio Cloud, and also answering any questions that you may have. So without any further ado, let's get started with the presentation.
Background and motivation
About two and a half years ago, I transitioned from working for the Australian government back into academia. And as the chief examiner, I was given the task to teach a couple of units on analytics at the Monash Business School. And when I started to prepare the materials and also to think about the logistics and also to think about the delivery, I tried to recall my experiences both in academia and outside of academia, working with people who were starting to learn R. And I really wanted to reflect at that time, what were the obstacles that the learners first encountered when they tried to start their journey learning R. And one thing that was very clear to me was that learners at the beginning, the first burden they face is the installation of the software, and also to get their first project up and running. I experienced that struggles with people outside of academia, and also with students in academia. So that's what when I thought that having a tool that could enable me to remove those kinds of burdens at the beginning, would be something very useful. And that's what I thought about RStudio Cloud.
When the COVID pandemic hit, we were here in Australia at the beginning of the semester, and suddenly we had to change the whole program. And instead of teaching in classrooms like the one you see over there, which is one of the classrooms that we have in the teaching and learning building at Monash University, we have to move to a remote and online teaching and learning environment. And at that time, I was really very, very pleased of the choices that I made before even I knew that we will be teaching remotely, that was using RStudio Cloud.
And so in Australia, when we had the first two weeks of the semester, and then all the lectures and tutorials they were doing online because of the COVID pandemic. So I'm sure many of you have experienced the same, but when you teach online and remote, the tools that you use really, they make a big difference. And actually, in my experience at RStudio Cloud, it really helped a lot to go through the semester in a very smooth way.
Student cohort and units taught
So let me first tell you a little bit about the student cohort and the units that I taught using RStudio Cloud. The students were both undergraduate and graduate students, and the classroom size ranged from about 300 students to 60 to 80 students. And in addition to the challenges of teaching remote and online, I also had to deal with the issues that the students in the classroom, they have different knowledge levels and different coding experience. And for the undergraduate students, some of them, they were first year students, and the units that I taught were core to their studies, while for some of the students, the units were elective. And that creates, of course, a lot of diversity in the classroom with the students' knowledge and experiences.
In particular, I taught two units, which are focused on R. The first one is Introduction to Data Analysis, where the students, they first, it's an undergraduate unit where the students first encounter R. And as I said before, some of the students, they are first year students, and this is a core unit for them, and some of the students, they take this unit as an elective. So they might come from a different degree, and maybe they have some experience coding in other programming language. This unit is taught at the undergraduate level, but it's also taught at the graduate level for the students of the Master of Business Analytics. And that this unit is totally focused on using R and get the students up to speed with data wrangling, a little bit of modeling, and yeah, the first encounter with R.
The second unit in which I use RStudio Cloud is a unit which is part of the Master of Business Analytics, so the students are all graduate students. The unit is Reproducible and Collaborative Practices. In this unit, the students learn about reproducibility, the importance of reproducibility and collaborative work, and it's also focused on R. We use a lot of R markdown to create reproducible reports, and this unit has also a very heavy emphasis on the usage of Git and GitHub, both through the RStudio Cloud interface and the R interface, and also using the command line interface.
Teaching and learning goals
When I was preparing the units like about two and a half years ago, and then I revisited these when the whole COVID pandemic hit, I wanted for the students and also for myself to have a teaching and learning experience that was efficient and effective, and I wanted to have a unifying learning framework where I could kind of have some control about how the students interact with the software, with R and with RStudio. I also wanted to remove the installation and the first usage burdens, and of course, I wanted all this to run smoothly in a remote or even in a hybrid learning environment. Also core to the program was I wanted the students to have an active learning based on the principles of constructivism, where the students are actually, they learn by doing.
And I also wanted them to have an authentic experience. I wanted them to use practical real examples, and when you are in a remote environment, it's actually very difficult to deal with students having different computers, different access, and so on. So having a tool like RStudio Cloud allowed me to kind of troubleshoot all the examples, make sure everything ran smoothly, and I had the control to actually make sure things work smoothly without having to worry about the type of computer the students have or how, like, deal with the issues that they might have due to the installation or the computer features.
Introducing R, RStudio, and RStudio Cloud
All right, so I told you about the two units that I taught using RStudio Cloud, but of course, these units are part of a larger program. So in our department, all the units in which we teach R, we use RStudio or RStudio Cloud in some of them. So it is very important for the students to understand the difference between R, RStudio and RStudio Cloud. So even in the units where I taught using RStudio Cloud, in the first instance, the students are introduced to R and RStudio, and then we bring on RStudio Cloud, the integrated development environment in the cloud, as the solution which does not require the students to install the program, and in which they could immediately run a first project in a very smooth way.
So the students actually embrace this in a very easy way, and they very much like it. I have to say, especially those students who did not have any experience coding, it's quite powerful when you give them a URL, a link, and then they go, they click, they open RStudio Cloud, and they suddenly can run the first program and create a visualization. That is something quite powerful, which the students really enjoy.
you give them a URL, a link, and then they go, they click, they open RStudio Cloud, and they suddenly can run the first program and create a visualization. That is something quite powerful, which the students really enjoy.
So in the two units that I mentioned just now, the way in which I approach the teaching of R and RStudio is a little bit different. In the unit of introduction to data analysis, I use RStudio Cloud for the entire semester. I will tell you a little bit more about how do I organize the unit, and how do I use RStudio Cloud for the teaching and learning. In the other unit, the collaborative and reproducible practices, I use RStudio Cloud in the first four weeks of the semester. So the students are confident to install and use R and RStudio in their local machines. And if you remember, I mentioned at the beginning that this unit, collaborative and reproducible practices, is very much focused on reproducibility, and on the usage of Git and GitHub. And not only using Git through the RStudio interface, but also using the command line interface. So in this unit, using RStudio for the first four weeks actually empowers the students to feel confident to install the software in their own machines and start using it locally, which is something very empowering, which the students very much like it, especially those who did not have any prior coding experience or experience with R.
How the units are organized
So here, you are looking at the contents of week three, where we teach them the grammar of graphics and dates. So in Australia, the semester has 12 weeks, and usually we use a learning management system to interact with the students. So this is a screenshot of the week three of introduction to data analysis. So for every week, there is a number of links related to the material that we are going to cover, where the students can read and get a bit more information. And then we have the lecture. And then we have a tutorial.
So the lectures, I use the lectures to introduce the concepts of the week to the students. And first, I discuss different functions that we are going to use, different packages, depending on where we are learning that week. And then during the class, I do live demonstrations. So the code is also on the slide, so the students can follow and do it at the same time if they want. But it is not essential that they do it at the same time. I'm more interested to discuss the different functions, do the demonstration, and have a conversation with the students while I do that. So for the lectures, I don't use RStudio Cloud. I use RStudio.
In my experience, it's actually quite important to, when we use RStudio Cloud, to keep this connection between the cloud and the local interface, right? Because some of the things that I have experienced, I have run these units now for two years. So if I, during the lectures, I only focus on RStudio Cloud, some of the students, they have kind of problems later on understanding, you know, that you can have R and RStudio in your local machine, and it's the same, but the way you access is a little bit different. Instead of clicking in the URL, you have to open it in your computer, look for your files, and so on. So I, during the lectures, I do, I use R and RStudio locally, and then during the tutorials, which are the part of the unit where it's hands-on and we do all the exercises and the practice, we use RStudio Cloud. Like this, I find I get kind of like the good of both worlds. The students are familiar about how do I interact in my local machine, and in some instances, some of them, they also install it in the machines, and at the same time, they see the connection with RStudio Cloud.
Tutorials and materials
So that is for the lecture. So each week, we have tutorials, which they run for two hours, and as I said, they are focused on the materials that we have covered in the lecture. So they are, the tutorial is hands-on. We have a number of exercises, and then the students work, they are divided in groups of about 25 of them, and they work through the exercises with, led by the teaching associate. So one thing that we do, which we found is quite useful, is we create, each week, we create videos for the tutorial.
So for each tutorial, we give the students a PDF, which I will show you in a minute, a PDF with all the exercises that they are going, we are going to cover in the tutorial, and also with the link to the RStudio Cloud space. Before, so right after the lecture, we also launched a video, which is, in the video, we teach the students, for example, in the first week, we just, we recorded a five minutes video, where one of my teaching associates really created a wonderful video, where she walks the students through the access to the cloud, right? So in each, for the video, in each week, we give them an introduction about, okay, this is, these are the exercises that we are going to do, we are going to click in this link that will take us to the RStudio Cloud space for the unit, and then that's where we start with the exercises. So the video kind of aims at giving them like a heads up about what is going to happen in the tutorial.
So if you see, we have the content. So in this case, we are going to run four exercises and the workshop objectives, and then the instructions. So if you see there, that is, follow this link here. So when the students click on the link, they are taken to the space of the unit. So, and then once they access the space, the unit space, you see below, that's what they see, in this case is ETC 10.10.55.10, and then they can access the workshop or the tutorial over there. So each week, we provide them with this material, and at the end of the week, in the same space, in RStudio Cloud, we provide them with the solutions.
Once they access the material in the RStudio Cloud for that particular week's tutorial, what they have is something like what you see on the screen. We still, they have the PDF with the instructions, and then they have the R Markdown files, the data, and so on. So we typically give them an skeleton, and then they have the PDF with the instructions of the things that they need to do to complete the tutorial. And if you see in the top left corner, that is the space that I create at the beginning of the semester for the tutorials for this unit. So the students, I share a link with them, they access this space, and then for the rest of the semester, every week, they go to the same link. Unless if for some reason I need to update the link, and then in that case, I let them know through the learning management system.
Managing the teaching team
All right, so in terms of the teaching team, so I'm the chief examiner and also the lecturer, and I have to say a fantastic group of teaching associates. They really, they are a wonderful group, which I think really makes the teaching and learning experience for the students a really wonderful one. So when using RStudio Cloud, it's not different from maybe other, when you teach other units, but I feel it is very important to have always an open communication channel. So the way in which I organize the RStudio Cloud for these units is there is an administrator for the space, which is myself, and then my head tutor has also admin rights to the space. So the other teaching associates, they don't typically have admin rights, and that's like, I want to minimize any possible issue that might happen if someone changed something within the RStudio Cloud space without intending to do so.
So what we do is we create the RStudio Cloud space for the unit, which is where the students work and access, and then we create a clone of that, which we call the teaching associate RStudio Cloud space, where the teaching associates, they have the same materials. Also, they have the solutions, and everybody there has admin rights, so they can play around. However, in the unit space, only the head teaching associate has the admin rights. So if there is some major issue, the other teaching associates, they can either contact me or the head tutor.
This is how the teaching associate RStudio Cloud space look like. So you can see there the admin, the members, they all have admin rights, and one thing that we experience that is kind of like, it's nothing major, but I think it's something that I would like to just point out for maybe some of you who might want to use this. When I first create this space for the teaching associates, I share the, I send them a link to the space, right, and then they join the space. But what happened once it was that one of the teaching associates shared the link with the students to this space, where all the material with the solutions are there. So these things can happen because when the students at first, they start to use RStudio Cloud, sometimes, in general, it goes very smoothly when you send them the link to the space and they join, but sometimes there are some issues, and then you have, they cannot find the link, and then you give them the link. So anyway, there could be confusion when you have more than one space. So what I do is for the teaching associates, once they all join the space, I change the way in which members can join this space to invitation required, just to avoid any issue. The teaching associates for the unit space, they are either contributors or moderators. So the reason for this is that some of the teaching associates at first, they feel a little bit worried that they might change some things in the main project for the tutorial, for example, and they don't want to have moderator rights, but they prefer to be contributors, and that's okay. And as I mentioned just now, it is important to kind of have an open channel for communication. So in our case, we have weekly meetings, and we always have a Slack channel open for any issues that they require admin rights. They can always contact me or my head tutor to deal with those.
Using RStudio Cloud for assessment
All right, so we also use RStudio Cloud for assessment. So we have one space for the tutorials, as I described just now, and we have another space where the students join for the assignments during the semester. So the assignments, I share with them the information via the learning management system, and in the same fashion as with the tutorials, then they receive a link, and then they go to the RStudio Cloud space, and then they start working from there. When they finish their assignment, they download the project from RStudio Cloud, and they upload it into the learning management system on their assignment, so we can keep track of the submissions and everything. And we can always, in RStudio Cloud, we can always go and pick in their projects and see what they have been doing in the cloud once they have submitted the assignment.
We also use RStudio Cloud for the group work that we have in the unit, and also I run a semester assessment with all the students. So the assignments, I launch them via the learning management system. In our case, we use Moodle, and it's something I create an assignment for those of you who are familiar with Moodle, but any of the other learning management system works the same way. I create the assignment, and then there is a link that takes them to the assignment RStudio Cloud space, and then the students, they click on the assignment and they start working on the project. So you can see there, for example, 265 projects were derived from assignment one. They work in the assignment, they download the project, and then they submit it into the learning management system. Also, the reason why I ask them to download the project and submit it is I want to also kind of make sure the project hasn't changed after the submission date in RStudio Cloud.
So the other piece of assessment that I do using RStudio Cloud, which the first year was a little bit like, it has worked very well the two years I made it, but the first year, it was a bit nerve-wracking to have almost 300 students doing this assessment at the same time, and everything had to run smoothly. And as I said, the first year was a little bit more nerve-wracking. The second year, it ran really smoothly. So this semester assessment is a time exercise that runs over two hours. So it's kind of like an exam, if you wish. I launched it via the learning management system. So for this piece of assessment, I create a different RStudio Cloud space in which I invite the students before to make sure everyone has access. Once they access, it is time, and it's the same time for all the students. They start working on the assignment in RStudio Cloud, and then once they complete the assignment, they download the project with all the files, and they submit it again via the learning management system. So in order to deal with any possible technical issues, we have a Zoom call open for technical support. So the Zoom call is not open for the students to try to get help about the exercise. No, it's just for technical support. Usually, there are not many issues, but sometimes some students, they keep being kicked out of their project, and sometimes I have to kind of download, delete them, and create like give them access again, but that's really a minority. As I said, it runs very smoothly, but having this Zoom call open, it gives the students the peace of mind that if something happened, they can always contact us.
Reflections and takeaways
So just to wrap it up, some thoughts. RStudio Cloud really, I think, is great for to kind of lead students in their first R journey. It removes the installation burdens. It actually kind of avoids the problem of dealing with different operating systems or computers with different specs. It definitely facilitates the remote and online learning because I can always see other students cloning the project. Are they working? And if some students have some issue, I can always speak in their project, which is really good.
Also, it gives you kind of like a flexibility to decide how the users access the different projects in terms of how much RAM do you need or CPUs. So if you are working with real-life examples, you can always make sure things work before you kind of open the exercise to the students, and if you need, you can adjust those parameters. I think one thing that is very important is to actually keep the connection. For people like us who are more used to use RStudio, it's kind of, it seems obvious, right? There is a connection between RStudio Cloud and RStudio. When the students, they are first starting their journey into learning R, if we only focus on RStudio Cloud, what I have experienced is some of them have trouble understanding what is an R project because when I give them a link, they already access, they are already in the project. So sometimes at the beginning when they are learning, it's a little bit confusing. That's why I try to use RStudio in my desktop for the lectures and RStudio Cloud for the practical exercises so the students see how the two things are almost the same but slightly different in the way you access to it, mainly because I don't want them to be confused with what is an R project.
I think it's a very empowering tool, especially, you know, like in the other unit, the reproducible and collaborative practices. I want them to build confidence to install the software in their local machine and to also use the command line interface and also interact with it through the RStudio interface. So I find that it's really empowering and some of them, they really love the possibility to go back to RStudio Cloud and try different things with different R versions without having to install all of them in their computer. So it's quite good.
So just as a final thought, like I always think about the teaching and learning experience is a little bit like fine dining, right? It's not only about the food but it's also about the experience and that's what I think RStudio Cloud brings a lot to students that are first learning R. And with this, I would like to thank you all for your attention and take any questions that you may have or any comments. Thank you.
So just as a final thought, like I always think about the teaching and learning experience is a little bit like fine dining, right? It's not only about the food but it's also about the experience and that's what I think RStudio Cloud brings a lot to students that are first learning R.
Q&A
Hey Patricia, that was great. Thank you so much. Really learned a lot, took a bunch of notes along the way and I see we've actually had a lot of questions trickling in. The first question that I think we should jump into, if I'm following the order correctly, is regarding your experience with GitHub. So can you talk a little bit about if you did have to prime students at all or if they came to the table with Git experience? Oh no, no. Actually the students, they had no experience. So some of them, they have no experience even coding R. I would say the majority. So they have to learn R, RStudio, and they need to also learn Git and GitHub. And in that unit, my focus is that the students learn to use Git through the command line interface. So they also use, I teach them kind of like to use Git and GitHub through the RStudio interface, but also through the command line interface. And the reason for that is that if you learn how to use Git through the command line interface, you can use Git with R, with MATLAB, with Python, with whatever you want. So I want them to have that flexibility. So I teach them everything from scratch.
If you can speak to your use of the functionality that allows you as the instructor to access the project of the students. Yes, yes. That is actually, that is a great feature from RStudio Cloud. So this is like what you see on the screen now. That's what I see as an instructor. And you see over here, it says view 265 derived projects. If I click there, what I see is it's the project of each of the students, right? So if I, let's say if I want to go and look at the particular project of a particular student, I just simply look for in the members space, look for the name of the student, and then I will have the list of all the projects the student has. And then I can simply click on their project and I will see, I will have their project in my screen. So it's very easy to access the student's project. And that's something also very useful if they really, I don't tend to overuse it because I want them to create a reproducible example if they have a problem where I can help them. But if they really are really stuck and I need to go and pick in their project, I can do it that way, which is really very smooth and very helpful.
The question, Patricia, is basically what was the budget approval process for a tool like RStudio Cloud or other tools that you use? What was that like at Monash? Right. So when I decided to use RStudio Cloud, first we had a conversation with Pete and the team. I mean, first I wanted to see what were the capabilities. Was this one suitable for our students? And Pete and the team were really very helpful. And then our school manager kind of discussed the pricing and the product that we needed. So everything happened through our department. So it was in our case, it was quite smooth. I remember at first we also had a discussion with the IT department at Monash to decide whether we wanted to use the RStudio Cloud server or we wanted to use the server at Monash. And in the end, it turned out that it was easier and it would work out better for us to use the server at Monash. And that's how we decided to go ahead. So the budget process, first I discussed with Pete and the team to see if the specs were aligned to what I needed. Then I discussed with the department manager and the head of the department. And then from there, they discussed the pricing and the funding and everything else with Pete and the team.
How does RStudio Cloud differ from RStudio server? A lecturer may set up his or her own RStudio server, maybe using a VM, then give access to their students to work online on their RStudio account. Right, right. Yeah, that's a possibility. Of course, it adds more work to you, to the lecturer. So I haven't used it in that way. I use RStudio Cloud using RStudio Cloud server. But what you are mentioning is what I was discussing just now, that we first talk with our IT department if we could create that like a server at Monash to use that. In our case, for our needs, we prefer to use RStudio Cloud server. But you could potentially do that and I'm sure Peter and the team can tell you more about that.
Yeah, absolutely. The other thing I'd add that I think this next question I just shared really touches on is that RStudio Cloud and RStudio server or Workbench are not actually exactly the same. So RStudio Cloud does offer additional features and capabilities that you don't find with the other offerings that lend themselves in particular to academic teaching. So the notion of a workspace, you can think of that as a virtual classroom, that doesn't exist in the other RStudio offerings. The ability to actually shell out assignments and to kind of peek into your students work, that's also specific to RStudio Cloud. There's some other differences as well, but just want to make sure that's clear that RStudio Cloud does have specific capabilities that really help you teach and learn that you won't find in the other RStudio offerings. Now, outside of the actual features, the other thing to think about is RStudio Cloud is a hosted SaaS offering, meaning we're hosting it for you. So from an infrastructure perspective, you don't have to worry about having servers or having the correct IT colleagues that can manage those servers. We're handling that all for you.
Let's see, we have a bunch of questions here. Here's another one related to GitHub. What about pushing to a managed repo? Do you teach that as well? Yes, yes. So the students, they learn how to use Git to create the repository, and then we like I teach them to store that in GitHub. They could use other remote repository. But yeah, yeah. They learn all that in the course.
I just wanted to mention that I realized I forgot to make the GitHub repository with the slides public. So I'm going to do it right now. Yeah, no problem. And I do see that some folks are asking about getting access to the slides. Patricia has been very nice and has agreed to share all this material. So we will ensure to get that out to all of you. Okay, so yeah, you should be able to access the GitHub repository with all the materials for the slides in the link that is there on the slides.
I also wanted to jump into a comment or a question that was made earlier around an academic completely free version of RStudio Cloud. So just to clear that up, there is a free tier of RStudio Cloud that's actually free for anyone, regardless of if you're an academic or maybe, you know, working within an institution, such as, you know, a corporate company, or maybe you're just freelancing on your own as a consultant. There is a free tier that you're more than welcome to use. It is a little bit limited in terms of the amount of time and the amount of people you can collaborate with. But if you've never used cloud, it's a great way to get started. And depending on your teaching use case, depending on how large it is, it could actually address that as well. Outside of the free tier, there are various paid versions of RStudio Cloud. Now, if you're using it for an academic degree granting institution, like Patricia's use case, there is a heavy discount, essentially a 95 plus percent discount, since we're basically just passing on our hosting cost. And we're happy to have those conversations with each of you if you have a use case that you'd like to run by us.
Well, one question that I actually had, I think it's great that you really spoke to the very relevant pandemic nature and how RStudio Cloud was able to help, along with the idea of using this with undergraduate and graduate programs. I know you had a use case where maybe you had 30, 60 students and another where you had closer to 300 students. Maybe you could speak a little bit to how RStudio Cloud scaled for those use cases. Right, yeah. Well, I have to say it scales really, really well, because once you create the RStudio Cloud space for a unit, you share that space through a link, through a URL, right? And then it doesn't really matter if you have 30 students or if you have 300 students. They all just go and click on the URL link and they join the space. So in terms of the students accessing the RStudio Cloud space, it definitely scales up really well. No issues there. Then, of course, the issues come, it's not the same to have 30 students with 30 questions than to have 300. But those are more about the content rather than the RStudio Cloud itself. So I would say it scales up very well.
I'm going to surface a question regarding Markdown. Say something about instructing Markdown specifically. I have found this actually the most difficult for students. Right, right. Yes, yes. It's interesting because in a way, Markdown is very simple, but students struggle a little bit at the beginning. So the way I approach the beginning of the unit, first I teach them about R. Like, for example, in introduction to data analysis, we discuss R commands and then we talk a little bit about reproducibility and then how we can integrate code with text to create a report. So that's kind of like I introduce this sequentially so they understand why do we need RMarkdown or Markdown and what are the benefits of it. And it works quite well. In the unit, the reproducibility unit, that one I start straight away into RMarkdown. So students who don't have a lot of knowledge of R, at first they kind of like, yeah, they need some time to get their head around. But I do the same thing as I do in the other unit, just a little bit faster because this unit, reproducible and collaborative practices, is more focused on reproducibility and creating reports, creating slides, and using RMarkdown with Git to create version control projects and with GitHub.
I'm having trouble locating it to bring it up on the screen, but I know earlier someone did ask about the actual grading experience. So how you went about grading assignments and maybe if there was any type of integration with an LMS, I believe you might use Moodle there, any insight you can share on that front? Right. So in the way that I didn't use anything that integrate RStudio Cloud with the Moodle system. So the way I did the grading for the assignment and for the assessments is the students, they work on their assignment using RStudio Cloud. And then once they have finished their assignment, which is contained in an R project, in RStudio Cloud, they download that project into a zip folder and then they upload the zip folder into Moodle. So like this, I can see who has submitted and if it's submitted on time and so on. And then what we do is we ask the students also to, in addition to submit the zip folder with the R project, they also need to submit the URL of the project. Like this, we avoid the trouble to have to look through the 300 students to find the project. So once we have the link, we just click and we can directly go straight away into the R project. So per se, I haven't used like an integration that kind of brings the marking straight away into Moodle. So we kind of, we have the assignment and then we mark the assignment and upload the grades and the feedback in Moodle.
Great, thank you. I see that someone did ask about a collaboration feature and in the past they heard RStudio Cloud was going to be working on that. So the good news is that we have been heavily working on that. So if anyone's familiar with the experience on Workbench or RStudio Server where you have a Google Doc-like experience where multiple people can edit and collaborate on the code in real time with different colored cursors, we are going to be bringing that into RStudio Cloud. The hope is we'll have that available by the end of this quarter. So please keep that in mind and once it is available we'd love your feedback if you're able to get your hands on it.
Someone did ask how do I get started talking to my college dean in IT about the possibility of hosting RStudio Cloud? I know within some institutions it involves you know first talking to your chair but if you don't mind reiterating how you went about that Patricia? Well I guess you would need to make a case, right? So what I did is I kind of I looked about what RStudio Cloud could offer that would be useful for the units that I was teaching and I kind of did like pros cons and they kind of brought a document with with the reasons why I wanted to use that and why I thought it was something worth to invest on and yeah it was it was quite I have to say yeah our department is quite pro R and RStudio and they kind of they embrace it with no problem.
Another question that I had is I'm just curious this hybrid approach I know you mentioned the idea of starting in cloud for one of the courses to help lower the barrier to entry and essentially doing that for the first four weeks and then getting those particular students probably for that type of course to learn how to install and set up. Have you found that since you adopt a cloud for that actual use case that it's been a lot simpler for students as opposed to from day one having to deal with troubleshooting? Yeah absolutely I really like this the hybrid approach because it really the students who have no experience they are very nervous at the beginning not to mention that in that unit in addition to learning RStudio and reproducibility they need to also learn to use the command line interface git and github. So to use RStudio cloud the first four weeks it really it makes things very easy because they feel when they are in the cloud it's kind of a safe environment they just click on the project the project opens everything runs you know everything is there for them they learn there are different versions of R how to install packages but everything is kind of in a contained environment where nothing really can go very wrong right so that's what once they have used it for several weeks for a few weeks they really feel very confident and we don't transition from like black to white right we the first four weeks we use RStudio cloud and then we we slowly transition to use it on the on the desktop so it really it really helps.
All right here's here's a good one that I think is always fun to talk about. Very challenging for me to convince folks that R is not an instead of Python. Frustrating. Any you know insight you'd like to share around that? Well I have to say in our department we don't have this challenge because we mostly use R for almost all the units. I don't know I think maybe we should put aside this R versus Python maybe Python is good for some things and R is and RStudio is good for for some others so I think yeah I'm more like of there are there are good things in both so I you know this is like the eternal kind of conversation and I really don't have like a very good answer other than well if you want to convince someone to use R versus Python I guess
