Good course, though I have to say that the final project was a bit confusing, and I am not sure that the people who did the final project really understood the course and how to create a tidy dataset, as the ones that I looked at did not meet all the principles of tidy data that were outlined. What concerns me is that they all had similar issues, and are all doing peer review of each other - this means that there is no one that can make sure that their answers are really tidy...

The principles of tidy data are well delivered and, overall, the course structure is great. Many great packages were covered (maybe some are a little outdated). Personally I felt that the data.table and dplyr/plyr shouldn't be covered to such depth, just mentioned. The main reason being is that they are constantly updated and often enough deprecate functions/parameters. I believe learning R dialects should be individual's choice and is not required for the data processing in R.

This is an interesting, helpful class. It was challenging, and exposed me to a very wide variety of topics outside R for data analysis, including databases, XML, APIs for getting data. I would have found found swirl type exercises for those topics helpful, as the additional practice really reinforced the lecture material and homework/quiz problems. I also would have found some worked examples or discussion of the homework problems after they were submitted helpful.

Pretty good, I liked the instructor. He explained things better than other instructors and didn't just read off the slides. Some of the quizzes were a bit off in my opinion. As far as I could tell with some of my code and the consistent answers I got, I think the quiz is not up to date with the data. The data is outside the course and is updated fairly frequently, one as early as December 2019 and I think the quiz answers were not updated accordingly.

This course could be a little difficult for people with no programming experience what so ever, even if they took the previous R programming course in this series. Examples are often a little too laconic and not all of the material is as practically useful as the best of it.

However Jeff Leak teaches some excellent data tidying, cleaning and extraction techniques with modern tools and libraries, that I find very useful in my everyday work with data.

This is a tough but important course. I learnt how to get the data from the web sources other than reading files of various formats, manipulate and group the data, and how to prepare a tidy data set for future analysis. There is ample practice to do each of the above. While the discussion forum is a great platform to address our queries, it would be good if there is greater clarity on some of the tools employed especially those in Week 2.

I liked the initial approach of this course, aswell as the resources given. However, I feel that the difficulty the exercises showed wasn't really linked to the concepts: many things were not explained in the lectures, and I had to find the complementary informations in various sites. So, rather than evaluating the understanding of the concepts given in the lectures, the skill to look for further informations was assessed.

The course is really great, it starts of a bit slow initially but then really picks up pace with new concepts and R packages as you move to week 3-4. It really helps you strengthen the basics in what exactly Tidy and Clean data is before you move on to more advanced concepts. The course material needs updating though, some links did not work, and the presentations which are downloadable are not selection friendly.

Great course. Just one suggestion. Thousands of students take this course. They have different internet environments, and the videos were recorded several years ago. It is possible that some of them are not able to download a file from a URL. It would be better if the course could provide files directly in case that students fail to download them, they can still continue to practice other skills.

pretty good examples, good guidance. However again it would be more helpful to start learning from a PROBLEM statement first, moving to an EXAMPLE on how to solve it and then explain how the new information helps you with this in THEORY. it makes learning so much easier and i don't understand teachers that don't follow this human problem solving approach for better understanding and learning.

The course has valuable content, but there is not enough emphasis on how to create a tidy data set. You kind of learn what a tidy data set is (although the definition is vauge), but you would need to see examples of messy data sets and how to convert it to tidy data set. There is one exercise in swirl called tidyr that addresses that, but it would be nice to have also videos on this topic.

Compared with other courses in the same series I observed several problems:

1. The explanation was not good, I had so many doubts that I clarified in other forums. The APIs lecture was too easy compared to the required to solve the quiz. The Dplyr section taught by Professor Roger Peng was the best explained.

2. Links s don't work.

3. The questions are not answered by the teachers.

I like the specialization quite a bit as it contains real world data and difficult enough exercises. This particular course is maybe not as good as the other courses I have taken (1,2,5) as the instructions lack a bit of clarity sometimes. However, the peer reviewed assignments are quite tricky and an excellent opportunity for learning. Took my some serious work to get this course done.

The first thing about the course is that the learning objective was clearer. And the content tied back to it, while also leaving room for self research and study. The project instructions could be a bit clearer, but perhaps the availability of the discussion forum allows this to foster curiosity and community interaction. Overall, it was a worthwhile course.

All Coursera data science courses have been designed very carefully. I found this course very beneficial as it explains the concepts and also tests the knowledge of the learner through tests.

In this course I learnt basics of fetching data from different sources like, API, Text-file, web-page e.t.c. Also I learnt cleaning data using various techniques.

Excellent course! Very useful videos, quizzes, and assignments. Provided the hands-on experience I was looking for. Improvement needs to be done to provide more technical information for doing the quizzes and assignments. Many critical details are being left out and students end up spending way too much time in digging them up via Forum, Google, etc.

Good introduction to getting and cleaning data and very useful learning about the principles of tidy data.

Jeff Leek isn't as good a tutor as Roger Peng and it was a bit frustrating following along at times as no hyperlinks are available for the data. The lessons are just recycled content from Jeff's lectures.

The course project was a good challenge!

Really great course material. I spent way too much time on the exams and projects, because i believe not enough information was given (had to spend a lot of time searching through discussion forums, stackoverflow, help files etc...and while that is useful experience, it was a lot more time commitment than expected from course description)

Honestly, I wanted to give complete 5 rating to this course, because the content of the lecture is well explained. But one feature I didn't like at all and that is Coursera has made it mandatory for the users to purchase the course in order to submit the quiz/assignment which I personally didn't like it. Except that the course is awesome.

Good course, better structured than course 1 and 2 of this programme. However, still a few of frustrating moments when the lecturers all of a sudden use language/jargon that cannot be understood by a beginner with little background in computer science (like me). Final coursework is ambitious, but answers can be found on the internet.

The course is more challenging than the prior two courses, but also very rewarding. I can apply the skills from this course directly to my work, and produce results.

A bit drawback is the some quiz questions and the final assignment can use more detailed instructions. Providing the clarity doesn't necessarily gives away the answers.

It's pretty good. It should provide a high-level approach to "getting and cleaning data"....and how it fits into the high-level roadmap of what Data Scientists do. Some of the quiz questions were written poorly - I expect better from an online course from a prestigious institution. I may have caught a few words misspelled as well.

The assignments were tough for me to handle. There should have been a step by step guide for the programs to be written just as in SWIRL exercises. Otherwise, the lectures are well understood. In general, I am quite fascinated by this course and look forward to learning more and better from other sources as well.

Very useful concepts taught in this course, one would be crawling data consistently as a full time data scientist. But even thought the pacing and project were good, I found the videos a touch boring. Nevertheless, it is part of the whole specialization, which in itself is serious education :)

Good course. I like it. However, it looks a lack of explanations or training for me. It possible, there too much material for this number of lectures, or just I haven't due background. Maybe, it makes sense to add more swirl courses for people like me to gain a proper comprehension.