Comentários e feedback de alunos de Advanced R Programming da instituição Universidade Johns Hopkins

I have taken several courses on Coursera and this was by far the worst. First the course contained no lectures. The authors had basically copied segments of their textbooks along with code samples as teaching material. I could have done better by simply reading the textbook myself without paying for the course. Finally the way the final assignment is graded is terrible. It is as if the course organizers have completely washed their hands of any responsibility. You have to grade three peers so that your assignment will get in line to be graded. But there is no time line and the process is completely arbitrary. Students have to actively seek others on the class forum to grade their assignment (they will return the favor ). I really expected more from a course offered by JHU.

Lacking in content. The OOP portion of the final project was poorly constructed. The instructions were unclear and the exercise seemed futile. We took a data set and further complicated it by abstracting into various object classes. I think a better example could be used. Since there are already plenty of ways to do the things our objects were tasked with it seemed like a futile exercise. I think a different type of real world object should be used, perhaps one that is not easily stored in a data frame.

Coverage of R Classes is good and helpful, but on the whole this class is a huge disappointment.

The ENTIRE course is text, apparently adapted from the creators' book. They also have the audacity to suggest you buy the book as well! Save yourself the trouble and just buy the book, or don't bother with either.

They also very much push the "tidyverse" school of R programming... it's useful to know and have in your toolkit, but there are many other approaches to R programming.

Also, support was very bad... Left a question on the forum about their code which didn't execute as expected. It has been two weeks since I left the question, and have since completed the course, but still no response.

Finally, there is no mention at all of vectorizing code, which is what R is optimized for. As one of the most basic tenets of R, it is surprising that an "advanced" course makes no mention of it at all.

Much of this course is well-constructed, well-communicated and clear. Sections from weeks 1-3 were attuned to my programming skill level and the material was communicated generally so that those with experience in other languages could also follow along easily. I had no technical problems with submitting assignments or quizzes and I found the assignments with swirl to be effective in reinforcing the material from the readings. My biggest criticism of this course is about Week 4's content and its respective part in the final assignment. The content in Week 4 concerns Object-Oriented programming and this material was really hard to absorb. It seemed to come out of left field and the readings did not feel as clear as they were in other sections. The readings essentially say "Here is what object-oriented programming allows you to do [create/manipulate classes and objects]" but then goes back on itself by recommending that you do not use object-oriented programming to create custom classes or data structures because the R community already knows what data structures they like. Personally, I had a lot of trouble understanding this content because it was so briefly explained. This did not help me by the time I had to complete the object-oriented programming portion of the final assignment. It is very clear from the class forums that other students have found that section of the assignment particularly difficult for years. Overall, this course was excellent in teaching functional programming principles and application, but the object-oriented section slightly spoiled my experience of the course.

I was really disappointed in this course. Most (if not all) of the materials are copy / pasted from a book that is free online. While I can get past that, my biggest concern is the actual content and assessments.

The content covers really basic examples and explains the concepts superficially, then the assessments expect students to be able to apply those concepts to a higher-level problem. As a university professor (and a cognitive psychologist), I completely understand the purpose of having students apply the information they've learned in a novel situation or environment. BUT that is only useful if the student has foundational knowledge on which to build. This course is akin to reading a dictionary definition, then being expected to apply it to an intermediate-level problem that resembles nothing you've ever seen. In the case of this course, the majority of my (and others', based on the forum discussions) learning was outside of the course. I have a hard time believing that any of which I "learned" searching for answers is going to stick with me long term.

The final assessment does not lay out the requirements (on which you are later graded). There are multiple things you get points for having in the code--even though a handful of them are not required for fully capable code--that are not specified as requirements in the final assignment. This isn't so much about the course per se, but something I find irritating and an impediment to learning in general.

That being said, I found the swirl assignments useful and engaging, and the course content was fairly easy to get through.

Overall I think more relevant examples would be useful, as would a more-detailed explanation of some of the concepts. If you take this course, I'd prepare to spend at LEAST a few days at the end of the course scouring the web for more relevant and detailed answers to your questions that were brushed over, if even addressed at all, in the course materials. It'd be better to go look for examples while you're learning instead of waiting until the assessment, though (if you're interested in decreasing the tax of the final assignment and increasing the potential for long-term learning).

The first three weeks assigments were way too easy and the answers could be found in the material.

The last assignment was very poorly described and the material was very limited in explaining OOP. That meant it took a lot of time figuring out what exactly was needed to complete the assignment and how to do it.

Disappointing after a great first course. Hoping the next will be better.

I was very disappointed with the course. Course material lacked dept and the final project was too complex relatively to the course material and examples shown.

What is the point of this course? Without any lectures, I'm not getting anything other than I could from just buying a book. I'm out.

The class does not cover all material needed for the final assignment in depth enough. I had to do extensive research outside of the class materials to be able to complete the assignment. I learned a lot, but I shouldn't have to do too much outside work to complete assignments for a class I paid for.

The last homework is ambiguous, and it doesn't have enough information available to solve it

It is good way to get to improve your R knowledge.

Instructions/learning material is informative, however, you should be prepared to put your google skills to work in order to complete some assignments. Overall I'm satisfied but I would appreciate a bit more attention to detail in the learning materials there are many typos and general grammar issues that break the concentration and some times require the reader to stop and guess what is being said from context.

I also believe the 4th week is lacking sufficient content for the learner to complete the assignment. There was a large amount of research required outside of the course materials in order to complete week 4.

I think the course introduced some useful packages like purrr and concepts like map/reduce, but I think the explanation into the different OO systems could be improved. I struggled a lot with the final assignment and really depended on external sources of information to complete it.

This course has a very clear goal. However, the implementation of this goal needs significant improvement. The first three modules were very easy to master. However, the last module was disproportionately difficult, making it unreasonable for learners.

This course introduced me how to deal with the real world challenges in R programming while developing applications. The solid guidelines and methods detailed in Mastering Software Development in R book by Roger D Peng presents and specifically in this course what it means to do robust programming in R and at the same time be aware of its limitations and error handling techniques.

A very good course. There are minor bugs with the swirl() connected with the issue swirl() expects to find the answer not in the global environment (or any special swirl() environment as well), but in stored file in the temp files memory on your PC related with the current session. Maybe it will be a useful knowledge to someone :) Alex

Solid overview of the topics in the course description. Does not go into much detail but provides a very nice foundation to build on. The course book is and will be a handy and useful resource, as it allows you to revisit the course materials (minus the exercises) without having to navigate through the course on the platform.

This course is fantastic. It explains two very important concepts - Functional Programing and Object Oriented Programing in simple yet detailed manner. Assignments are helpful in finetuning the OOP concepts. After completion of this course, I am pretty comfortable with R as a program

It is a short and good R course. It goes right to the more advanced stuff. Even a long time user of R will learn new things. The final project is revised by the same students which in my opinion makes it more of a challenge. I've never wanted a homework to be as perfect as this one.

This is really a good course to get started with the R language if someone has an idea about basic programming in other scripting languages. Course content is well organized and the pace of the course is also good enough to follow up easily in busy schedules.

Great curriculum. However, I feel the OOP examples could have been designed from basic - intermediate - advanced levels. Else, it is good to have a separate module on advanced 'OOP' using S3, S4. More examples we work on, better we will get at it.

This course helped the programmer understand some OO concepts and many other features of R that I may not have encountered. Very good, and the final assignment was good. It was worth putting the extra time in to do it correctly.

I like this course as it offers some really practical instructions to programming in R, and I think it's a course that well complements the online book Advanced R written by Hadley Wickham, which focuses more on theories.

Very meaningful class and it teaches so many things, with so many exercises and test exam that basically force us to leverage every bits of learning that we directly learn from the class. Definitely would recommend.

Excellent course, I recommend it. Very good labs in swirl, good readings from an open text-book. As usual, there is a huge leap from the readings and labs to the final assignment, which is a bit stressful.