good

NA

.

I want to begin this review by thanking Johns Hopkins and Coursera for putting this course together. In the age of "work/learn from home" and "upskilling" it is courses such as these that allow the layman who is unfamiliar with data science to learn basic programming. For that I am grateful. On the other hand, even as this course cost $50.00 (which is reasonable), I still cannot recommend it to other students. As a graduate student, I have had some exposure to R. My statistics class utilized R for HW assignments and basic regression models. Our TA taught us some intermediate coding methods via ggplot2 and dplyr. I primarily took this Coursera course as a refresher. It should be noted that the course description encourages - nay, states - that beginners should do just fine taking the class. Nothing could be further from the truth. The course instructor is clearly a brilliant man who is a leader in the field of data analytics. I'm sure he is also a very well-respected lecturer at JHU. However, I felt like this whole course he was speeding through the content. I had to use 0.75 speed during the lectures so I could hear everything that he was saying. On top of that, he used a lot of R jargon that made the content seem more exclusive to students who already have a background in data science. Finally, there were limited opportunities for students to apply the skills that we learned in each lecture in the actual R environment. He took screenshots of code, explained it in an opaque fashion, and then moved on to the next lesson. That may be fine for some students, but I personally like to 'drive' when I'm learning how to use a car. The one bright spot in this course was the interactive swirl sessions which did allow you to put some concepts to good use. It should be noted that these are optional. In my opinion, if you didn't do the swirl assignments then I can't see how you took anything away from the course at all. Still, the swirl module could be clunky at times and could use updating. In addition, some of the answers in swirl simply required you to copy and paste what you were previously shown, which isn't very challenging. My final qualm with this course is the "programming assignments." While the swirl assignments were probably too easy for the layman user, the programming assignments were insurmountable tasks. There is a HUGE gap between what we learned in the modules and what we were expected to perform in the assignments. I don't understand the logic of teaching concepts, implementing what we learned in simplistic interactive swirl sessions, and then taking on advanced assignments. How does that help the beginner student at all? The teacher encourages a "hacker mentality" but in my opinion, struggling and googling your way to get the right answers isn't learning, it's insufficient teaching. It would be one thing if this course was advertised to advanced users. If that was the case, fine, the student should probably be able to write some of this code. But that was not the case and from the comments, it seemed that many other students also struggled with the program assignments. The teacher also dove into statistical theories in week 4 without explaining them. No offense, but I don't understand how we can possibly grasp linear models and Poisson regressions in a sub ten minute video. In summary, don't take this course if you are beginner or looking for a refresher. The content is outdated, the pace is too fast, and the programming assignments are disconnected from what you learn in the videos. One would be better off watching YouTube videos or taking a Udemy course for free.

I am a very frustrated learner trying to write a constructive review here. I studied the course full time to get a certificate to put in my profile. R programming is considered nearly essential skill, if not fundamental, to data analysis. So I had high hopes going into this course.

Lecture videos: I cannot begin to tell you how many times I fell asleep watching the video lectures. This course is for you if you like listening to someone talking through pages and pages of copy/paste text from a command prompt. I, on the other hand, prefer learning by doing. Sadly, all the doing is clumped into the assignments (more on them later). Also, Dr. Peng makes gross burping/drooling noises in the video from time to time. I apologize if that sounded rude, but try listening to the lectures with headphones and you'll say the same. Video editing for pre-recorded videos is not rocket science.

Assignments: These were the biggest frustration throughout the course. Imagine you want to learn to play the piano (and you play a little guitar now) and you go to the piano class. Your piano teacher talks your ears off with music theory for several hours and then hangs you out to dry in front of an audience in a concert hall. You protest "But I have no clue of how to PLAY", but your teacher says "All the piano greats learned by fiddling around with a "hacker" mentality". You ask "Why did I pay $50 to hack on my own?"

Off-lecture help: If you take the course for the same reason I did, then thank your favorite god for the moderators in the discussion forums (mentors). 1 out of the 2 stars I give belongs solely to them and the fantastic work they do. The stickied posts in the forums offer a bit of help with the assignments (not enough, but still something). And they are also active in answering questions. The other star in my review goes to "Swirl". You will learn way more by doing the swirl exercises than watching the lectures by a long way.

In conclusion, I finished the course with nothing more than a rudimentary understanding of R despite the fine grades. Very little thought seems to have been put into the lectures. I would recommend this only if you want to show this certificate to someone. Otherwise, stay away!

This course is VERY abstract and I find myself rushing through the videos to get to the practice/quiz so that I can trial and error my way through the project..... hoping for the best. Neither am I excited about starting the lesson each week because there is no real world problem/or data set I'm continuously practicing from. The lessons are essentially a reference guide and not a useful approach for teaching. If I wanted a reference guide, I'd just pick up one of the various handbooks/books on the markets that list R codes and their functions. A better way to teach/present this course is by infusing actual, real world examples or cases throughout the lessons instead of just listing a function and talking through its corresponding activity/response. Ideally, the real world example would be introduced in the first lesson in Week 1 and that data set would be used throughout the course(s) to apply and practice newly introduced functions. Teaching from this perspective would likely make the concepts much easier to grasp and importantly, RETAIN. The lessons, as they are currently presented, encourage rote memorization and rob the students of actually applying the concepts/code taught.

Also, I love the idea of this specialization. However, I think the professors need to work more closely with online instructional designers to make the entire series more palatable for online learning. It very much feels as if they have adapted a traditional course on their own without the help of professionals who are skilled at designing online courses. If the aforementioned is the case, the professor(s) should be commended for their efforts, but there is definitely more work to be done to make this an engaging course I'd recommend.

PS - I am taking notes and I have some experience with STATA so this type of coding/anaylsis is not unfamiliar to me. I can only imagine how novices may take to the lessons.

I think it is disingenuous to rate the Data Science Specialization at Beginner level while the R Programming course being rated at Intermediate. It is really a course for experienced programmers. While I have *some* programming background (I won second place in an International Curl Programming contest in 2001), I think professional programmers would have a much easier time with it, although in looking at many online reviews I see professional programmers struggled with it. It did shake the rust off my coding skills, although if I were back in programming instructor mode I would keep the assignments, which were excellent, and work backwards to devise lectures that better supported them, and provide simple exercises that develop the skills needed.

I found that I spent about 35 hours on week 2 between lectures, the assignment, and test, with almost all of that spent on the test. As a busy mature adult the extra 30 or so hours was very unwelcome and I think quite unfair in terms of the stress load of making me wonder if I was going to make it through the material. In looking at many course review comments I see I am not the only one. You need to even out the workload.

When I was teaching JavaScript and Visual Basic at a university from 1999-2003 I found that stepping students through some examples then letting them solo on slightly altered examples while providing support if they were really stuck, then giving them more challenging examples to test their skills, was a very effective plan and helped them maintain their self confidence in the material. You could always give "Challenge" questions that differentiated between the B-plus and A-plus students that force them to develop their hacking skills. Unfortunately, I can't recommend your course until you improve the lessons in the manner I've described.

The only reason I am giving this course 2 stars is because the Swirl modules were awesome. Swirl should be the main form of instruction in this course. I have taken one class using R a long time ago, and have a very very basic understanding of programming -- so not a TOTAL beginner, but basically a beginner -- and I found this course to be absolutely impossible to do using the materials provided. The quizzes and especially coding assignments seemed, for the most part, completely unrelated to the lecture contents. The lectures, in turn, contained all sorts of details that were not needed for the assignments and quizzes and I often felt lost. Lectures were generally barren of concrete examples that have any sort of meaning to the audience -- for example, when teaching about a function, instead of showing an example that actually illustrates how we, as data scientists, might ever want to use the function, its usage is demonstrated in a simple but ultimately meaningless (and therefore forgettable) way. It would have been much more useful to show the basic example and immediately follow by a real application. There were many, many, many functions that one seems to need to have memorized, but the course provided no tools to help facilitate that, or input on which functions to focus on. It was a big mess. I ended up teaching everything to myself by Googling... which is ok, I guess, but why am I taking a course, then? There has to be a better way to usher students through this process, even if the goal is ultimately to get us to be resourceful and find information on our own (like why did I spend all that time on lecture if I was just going to have to look everything up?). Swirl was seriously the only thing that saved it.

I would have liked to see more repetition of functions discussed in the videos that slowly build up to the more advanced questions in the quizzes and assignments. The Swirl lessons are very prescriptive. While Swirl is good in theory, it aids you along too much for it to really be effective. I recommend adding practice assignments that have learners do a particular task independently multiple times with different variations so we can really learn what we're doing. Throughout the course, I would watch videos, understand everything being said, and then be completely lost when it was time to do the assignments. I'd use the forums and Google to complete them, but it felt like there was a disconnect between the videos and the assignments that made the course more difficult/time consuming. If there were an interim step of doing practice questions (not through an overly prescriptive package like Swirl, but one that makes you use solve the problem on your own and struggle just a little bit), I think that would've really help make this course more useful. Also, doing something once in an assignment doesn't mean it's committed to memory, so even though I was able to figure out the assignments, I don't think I've really retained as much info as possible. In terms of grading peer assignments, I don't feel that I know R well enough after taking this course to say whether someone else's code works or not (without running it and assuming that it doesn't match mine exactly). I appreciate the time and effort it must have taken to set this course up, but it fell short of my expectations and I don't really feel that I've gained much (if anything) from it.

This course does teach one useful tools in R. However, I feel like this course needs some serious improvements especially when it comes to the disconnect between the lecture videos and the actual programming assignments. To me it felt like during the lecture videos I was being shown pictures of a car and told what each part does, then in the programming assignment it was like I was being asked to build a fully functional car from scratch. It just wasn't practical. In the future I think to make the Programming assignments more realistic to complete on their own the lecture videos show go more in depth on the tools and functions it's teach. Not only that but i feel like I would have greatly benefitted from watch the Instructor go through sample problems and situations where I could see the tools and functions in practice. Especially for me who is a beginner in R to be able to create some of these fully functional R functions from just the lecture notes would have been extremely difficult. Also, a little note in the future you should adjust the approximate times it would take to complete the programming assignments because i don't feel like they are realistic for new programmers. For example, the approximated time to complete the week 4 programming assignment said 10 minutes and that was just flat out wrong. If you do make some of these changes I feel like this course will be significantly more enjoyable and that the students will learn significantly more.

Be prepared. The lecture portions of the course are quite short and basic. They don't provide much in the way of practical examples.

The assignments often have little to do with the lecture topics (week 3 introduces a whole new concept that you have to learn all about outside of the course to complete). They're decent assignments, and you'll learn from them, but if you don't already have a lot of experience, expect them to take a long time to complete.

I guess my main problem besides this is that there is not enough emphasis on learning the foundations. They introduce subsetting and give you some simple assignments which give you a basic superficial understanding of the concepts and syntax. Unfortunately, that stuff takes some time to really sink in, and you need to sit and play around for hours to get a good feel for it. If they had some optional assignments where you could have some guided exploration of the topic that would be super helpful.

The one saving grace is that there is a lot of help in the forums for each week. The mentors are doing their best to help students out and are generally pretty quick to get back when you ask questions.

My DEAR COURSERA TEAM thanks for providing this course.But I am sorry to say this course is not for the beginners who was not professional coder. I am from background of applied electronics and instrumentation working in Level 4 automation at Oil and Gas domain as subject matter expert. I wanted to use this language for analyzing the plant's day to day critical alarms data. During the phases of the course I faced lots of trouble when performing the assignments. I have to approach some expert coder in order to understand the question and answers. I would feel that team has to modify the course. One more feed back reading method is not at all helpful where as practical was helpful.

Once a learner is accomplishing one course hence material of the course should be sufficient in order to finish the assignments where as this course is not meant for that, I would prefer please modify the strategy. Assignments are not really helpful for learning it simply killing someone times. The reason is during the practical or reading at least same kind of example like assignment, should be demonstrated.

This course seems to have a huge gap. The content presented does not prepare you for the final assignment. The lecture videos introduce R functions and there are interactive demos in R that show how to use them. However, the final assignment requires you to determine which functions to use and how to implement them correctly. Consequently, I had to spend >10 hours finding online examples to complete the final assignment and the code I wrote used functions that were not presented in the course!

I realize that there are multiple ways to write R functions and there are tons of existing R functions to utilize, but what should be presented is one standard way of reading data, organizing the data, sorting the data, etc.

Also, I found the lecture videos to be very uninformative. The presentations need to be more streamlined, focused, and include basic examples.

Overall, I don't feel like I have a firm understanding of R. I certainly do not feel comfortable teaching R to others. I think the course could be improved with more examples of basic R functions, such as lapply and order.

By the end of the course, you do start to develop the "hacker mentality" the course writers want you to develop. That being said, it does seem that they could have done a much better job in the video introducing core concepts that would have given non-programmers a better starting point for looking for outside researchers. A ton of time is wasted in the earliest set of videos on super basic topics, but then the course jumps into some very advanced topics with little attention paid to core concepts except by the mentors in the forums. If these really are "core concepts", as described by the mentors, one would think that the teachers would spend time on those concepts in the videos instead of on more arcane use cases (which are precisely the sorts of thing it makes sense to look for in other sources). It really does seem that the teachers have no concept of what is critical content for beginners, given this odd imbalance of tons of time on extremely basic topics, tons of time on fairly advanced topics, and almost no time on critical intermediate-level content.

The lectures themselves were clear and engaging enough and the swirl practice was helpful, but the learning curve between the information we're presented and the assignments is unacceptably steep. I understand that there is always a gap between lecture and application; however, I should not have to sift through a dozen discussion posts to figure out what is going on in the assignments. Students should be able to complete them with just the functions that were covered in the course materials, otherwise, the course materials are insufficient.

My degree is in education and if I had so poorly prepared my students to complete their assignments, then I would get in a lot of trouble. It seems like this has been a complaint for some time and there have been efforts to improve through the resources provided in the forum, but this calls for either rewriting the assessments so they align better with the content that was covered or a fundamental restructuring of the course content so students are prepared to complete the assessments as currently written.

This course is not set up right, the assignments ask you to do things that aren't explained until the next weeks content, its kind of discouraging. In the end, it does teach you the basics of R, its just too bad that the way to get there is aggravating.

The reason I still only give it 2 stars is because of the quality of the courses itself. There are many, many 1 and 2 minute videos, these could've easily been combined. The teacher seems unprepared in his lectures, he stutters and repeats a lot and makes a weird noise between slides. This shouldn't be necessary with pre-recorded lectures. The assignments and quizzes are also poorly written and contain spelling and sloppy mistakes, which doesn't make sense because the material isn't new. It all just makes it seem like the teacher doesn't really care and just wants to sell the course to a lot of people without putting in much effort.

If you are not following the specialization I would not advise this course for beginners. I'm quite surprised to see the course get such high ratings.

This is not a course for a beginner in programming. If you are interested in learning R, I would recommend going to DataCamp and starting there. The lecture videos were not engaging, and the jump from the lecture and quizzes to the programming assignments was quite significant (as others have pointed out). The lecturer for this course does have a pre-programming assignment on his GitHub repository to aid in the first programming assignment, but this was not easy to locate nor advertised well (had to scroll through discussion forums before I saw someone mention it). The swirl assignments were the only saving grace for a class that was otherwise not engaging or structured well. The class could stand to be restructured: for example, it would have been nice for the lectures to cover the str() function before our first programming assignment. Overall, this class does seem to be a good choice for a "refresher" course if you already have some experience coding in R, but be wary if you have little to no programming experience.

I would only recommend this course if someone were to audit it, and not pay for the certificate and access to the assignments. This course states that it's for beginners, but I would disagree. While the lecture material was helpful for me, someone with little background in programming, the assignments seemed poorly fit to the course. There was a very large gap in what is learned in lectures and what is meant to be applied in the assignments in order to pass. I understand the instructors want to cultivate the so-called 'hacker's mentality' within the students, but I felt I did not get what I paid for. I wasn't anticipating to pay a fee for this course only to have to go and spend a lot of time on google filling in gaps that the instructors did not cover in order to pass, and still not understanding the assignments fully. I think this isn't conducive to great learning for beginners (as this course stated it's geared towards), and for this reason I wouldn't recommend this course.

This course is very bad. I would like to say this course sucks compared with my sql course no matter how it is structured or how the instructor conducted the course. The instructor is very boring and does not explain very clearly. Even when I start, he does not clearly clarify how to start to use R. He was just talking about what the powerpoint shows. I will absolutely not recommend this course if you are beginners. (0 STAR)

I will give a little credit to swirl, a practical tool provided by this course to learn R by yourself thought the instruction of the R program itself. But for the later part of swirl, it's confusing sometimes. However, I learnt much more from swirl rather than from the instructor. I can't understand what he is talking about but you have to make it clear thru using swirl. He said the swirl is an optional part but he does not realized what he shows is much worse than swirl. (2 STARS)

Overall, I don't recommend this anyone, especially if you are beginners.

Poor quality. A disappointing experience.

The lectures are very basic, thought for people that have no experience in programming. This might be ok, but the difficulty of assignments should follow; instead, they can be relatively hard; newbie programmers will probably have an extremely hard time solving them, considering that many of the problems they will encounter are not treated at all in the course.

The course does not focus enough on what differentiates R from other languages: just a few videos on data types, *apply functions, and a tiny little bit of scoping (very unclear). All these areas should have been expanded, and the course should have had more arguments as well.

Finally, the assignments are not enough and their grading system is quite poor: instead of submitting your code and testing it automatically, you should calculate some quantities by hand and select the results out of a few possibilities.

All in all, I was disappointed with the course.

Althought there is a lot to learn here and the instructor knows their stuff, I am disappointed in this course. First, I recommend going ahead and having the computer give most of the lecture, except for maybe the introductions to each week to keep it personalized. The instructor repeats "um", "uh", and "so" far too much (2-4 times per sentence) and it's very hard to focus on the lecture.

Second, the assignments do not match the material taught, and there is no scaffolding. I have a Ph.D. in quantitative sociology and eight years of statistical coding experience (not in R) and I think that is the only thing that got me through the assignments. I'm sure I would have been completely lost if I had no programming experience. It makes no sense to me why the instructor doesn't slowly build the assignments in difficulty, but rather throwing students in the deep end analyzing data from 322 different datasets.

Please improve this course for future students.

This is a challenging course because there is a marked gap between the video lectures and the weekly assignments. This course will be far more demanding of your time and willingness to fail that the Data Scientist's Toolbox. If you have the time and resilience to search out support materials and frequently peruse the Discussion threads, you will be able to find what you need. My peers with programming experience confirmed that there is too wide a gap between the content delivered in the video lectures and the demands of the coding assignments. If you decide to take this course, be sure to do the swirl exercises so you get a feel of how the R functions work. When you grade your peers' work, you might notice that more than a few decided to plagiarize code from others.

The videos need to be redone. Effective instructors know that students learn by seeing examples, not by listening to broad descriptions of what a function can do.

From my past experience, I know it’s not easy for people to take suggestions. But I still want to have a try this time.

I had high hopes for this course, but I am quite disappointed. I think Dr. Peng needs to improve his teaching skills.

1. Good teaching is clear, concise, and right to the point. So please slow down, speak. Don’t mumble.

2. Programming is a practical skill. So the best method to learn programming is to use step-by-step demos. You can talk about a concept for 5 to 10 minutes but I still can’t get it. Use a demo and I can get it right away.

3. Don’t just try to cover the materials so you think you have done the teaching, try to understand how your students learn and make sure they really get it.

4. If you haven’t covered some concept, don’t assume your students will understand it by magic.

A good teacher can make the learning experience effortless and fun, a poor teacher makes it like a torture.

This course is "1-dimensional" and not of high didactic quality. I have followed several other courses ( also Python in Coursera set up) and this one really disappointed me ( I'm in wk 2 out of 4 now). It is a bit of a pitty that the instructor speaks too rapid ( it is ment to be also for non-native US speakers?). The material is hardly explained step by step. So what I do at the moment: I watch the video and than go to Yuotube to find other videos that deal with the topics in an easier and didactic better way.

Examples of good cousres in my opinion are stats and probabilty course/ and the python course

PS "all of the data toolbox" course seem to have the same flaws.

Of course my knowledge level is low concerning programming ( I am a MD, so hardly scientific educated...). Nevertheless I hope this feedback can be of any good.

This course features great Swirl exercises and decent if bland lectures. Unfortunately the homework assignments are highly problematic. They (a) do not build enough upon what is taught, (b) all too often require skills that have not yet been taught, (c) are extremely advanced for what is supposed to be a beginner R programming course and (d) often require you to look up things in Google rather than in the class lectures or exercises. You can easily put 20 hours or more into this course per week and still struggle to complete or pass it. I would recommend beginning R programmers to steer clear until they improve the assignments.

If however you're a more experienced R programmer this might be a worthwhile challenge, but I would caution that the assignments are not well-tied to lectures and exercises.