Great course for probability and the foundation of statistics.

However, to get the most out of the course it is essential to study the related chapter in the book "OpenIntro Statistics" and do the examples, and exercises first. Then it will be easy to review the contents with the videos.

Therefore, the time required for the course is at least double to triple, at least in my case.

I have background in mathematics and statistics. So I didn't have any problems to understand the different concepts and formulas. A background in R programming including R Studio and Tidyverse, ggplot, etc. is recommended.

The course is an excellent introduction to probability and exploratory data analysis with R. Concepts are well explained and Prof. Mine Çetinkaya-Rundel is an excellent teacher. The materials of this course are very well organized and the concepts are clearly explained. The examples displayed are from real-world data and are very well chosen to further develop the theoretical concepts. Regarding the final project, it required me some time of autonomous work to figure out the right code. Moreover, I enjoyed and learn a lot from this course.

This course had a good balance of easy and challenging content. I like how the reattempt feature of the quizzes doesn't just give you all the same questions again in a different order. But at the same time it doesn't completely change the set of questions. This forces you to go back a really understand the content if you want to maximise your mark. The final project seemed harder than I was expecting, but that resulted in gaining a lot more practice with RStudio, which really helped to learn it well.

Dr. Mine Çetinkaya-Rundel is an amazing teacher! I have never learnt and enjoyed so much statistics!

Syllabus is well constructed and organised, plenty of learning materials (with moderate difficulty). This first course is aimed at beginners (no requirement of prior statistics knowledge). Although I use statistics regularly, I don't really understand them well so I find this course extremely helpful!

Thank you so much for putting together this course of the whole specialisation!

I've not studied at this level for a while so the first couple of weeks were intense. The pace didn't let up but the quality of the presentation material was excellent. I didn't feel quite prepared enough for the project and it took much longer than the estimate to complete but well worth it as an exercise.

Very pedagogical course !

I liked:

+ The detailed Learning Objectives.

+ Good examples in the lectures.

+ The quizzes are great for testing and refreshing one's memory.

+ Overall the course seems very well-focused on the most important foundational items and hammers them in.

I'd appreciate improvement in:

+ Providing more clarity on how much R is expected to be learned for the final project, or lowering the level of R skill expected for the final project.

+ Providing a more relaxed time estimate on the final project. I spent 10+ hours on it, in addition to ~10 hours learning more R on DataCamp.

+ Getting feedback on my work from a professional, not just from fellow students. (I would be willing to pay for that.)

Good course for statistics but not for R

All the political statistic examples were more distracting than enticing in my case.

The sound isn't normalized from lesson-to-lesson so you have to turn up the sound for some and down for others. In certain videos, there are incisions of audio segments, re-dos, which don't have sound normalization easier. An easy fix.

The R itself is taught through text. The videos (in terms of content) are great for people coming back to statistics who've taken it before. The supplemental textbook readings probably aren't necessary for people with familiarity. I give 3 stars because my next-day motivation of 'let's get back to it' is not there. I'm coming from Python for Everybody (coursera) from University of Michigan.

I wish they went over how to use R for beginners in the beginning of the course. I feel like the final project was stressful and piecemealed together from google searches on the web. I think they should give us a list of all the codes in the beginning of the course that we will use throughout the entire course.

The video lectures were an excellent introduction/review to basic statistical concepts. I'm grateful for what I learned. However, if your goal (like mine) is to gain introductory skills with R, I *do not* recommend this course. The labs in Weeks 1-4 were very difficult and there was almost no guidance on how to complete; I struggled through them by finding solutions on StackExchange, etc. But the worst was the final project in Week 5; the material in Weeks 1-4 was absolutely inadequate to prepare the novice user of R. Furthermore, the assignment is "peer graded", meaning you will not receive instructor feedback--fellow novices will attempt to provide feedback on your work. It is the blind leading the blind. Absolutely not worth the money.

Before you start this specialization you should check the start date for the final capstone course and make sure that it aligns with the time you anticipate wanting to take it (add a few days for the annoying bottleneck of getting peer reviews back for your assignment in course 4 since you will not be allowed to enroll in course 5 until all of the first four are completed).

When I made it to the capstone course there was a full month until the start date so all of the assignments are locked until then. I am far from the only one this happened to. Others on the forum for the course are complaining about two-month waits. This seems like a naked cash grab since the specialization is a monthly subscription. Honestly, it would be better if they did just surprise you with an extra $50 charge to do the capstone because at least you would have the option of paying and getting started on finishing up the specialization.

Otherwise, the specialization isn't really maintained any more. For the first three courses it doesn't matter much because it's all pretty easy (my rating for the first three would actually be two stars as stand-alone courses probably). In the third week of the fourth course the instructor is swapped out for a new instructor who is utterly incomprehensible, using terms and ideas that haven't actually been introduced. The accompanying text is similarly incomprehensible. I say this as a math person who has subsequently gone to other resources to learn this specific material. The problem is not that the material just gets harder, but that the quality of instruction drops through the floor. So beware of that.

By the textbook it appears to be a very vanilla Stat 101 course, chapter 2 of the textbook did cover some conditional probability and Bayes theory otherwise very similar to stat 101 course I took in 1978. In other words, not exactly 21st century data science. Taught at high school / college freshman level (one of the exercises was writing mean and standard deviation with Greek letters -- pure busywork). The textbook inaccurately described a data set with several different data types as a "data matrix" a word usage that conflicts with both linear algebra and the statistical language R. This course is an obsolete and lame intro to statistics; I would recommend instead either Stanford's Introduction to Statistical Learning or the Coursera's Data Science Specialization.

This course (and specialization) was promoted as suitable for someone with no programming background. The fact is that the way it is taught, it is assumed that the learner has basic programming experience. I enrolled in the course and would now like to un-enroll since I have no programming background, but the un-enroll option is not available to me.

The instructions for all the labs were confusing and left me lost. The pace was too quick, and I didn't feel like I learned anything.

Lectures by Prof Çetinkaya-Rundel: Excellent content for the time alotted and well-organized. The URL to the "Distribution Calculator" was consistently erroneous in the slides (=: You can find it here: https://gallery.shinyapps.io/dist_calc/. The source code for that web calculator is part of a github R-language project and can be found here: https://github.com/ShinyEd/intro-stats/tree/master/dist_calc . One of these days, I am going to write up an issue on her use of "set.seed(12345)" and see if she takes it seriously! Her code is not bad for a statistician. (=:

Very simple Math. Depending on your background, this could be dissapointing or a relief. Too light for me.

A lot of R-language, RStudio, and R-markup. Expect to dive in. Note that RStudio is still evolving so you can expect a surprise or two along the way (E.g. crashes, hangs). The web version of RStudio has its own issues. I would stick to the desktop version since you have more control of your own desktop. Windows users can gleefully reboot when RStudio hangs - is it RStudio or Windoze?

"Let them eat cake" should not be missed: https://speakerdeck.com/minecr/let-them-eat-cake-first-0a3bbf75-f6f1-42d5-8d2f-ac2ff741611f .

Warning to physical science students: This is not in any way a criticism of the course but there are no examples from Astronomy, Chemistry, Geology, or Physics. Only health and social science stuff. I would have liked to see course content applied to something like data acquired from a telescope such as TESS or Hubble but that's me.

The forum for the last week is clogged with requests to have final projects reviewed, detracting from the intended purpose. I do not agree with having lower division students review each other's work: student maturity level to assess and give feedback is lacking. But, I knew what I was letting myself in for.

Another warning to physical science students: The data for the final project is CDC health survey responses for 2013. Roughly, 0.5 million rows and 330 columns. Lots of non-response values in the data. Not my cup of tea.

Very impressive course, certainly among the top five I've ever taken online. Course design is basically flawless. The lectures are clear, concise and based on interesting examples. The course also comes with a textbook for which you can pay what you want, even a price of zero. The textbook is also very well written and contains plenty of examples to illustrate concepts that are introduced and lots of practice problems too. This is crucial for developing a good understanding of the material taught in a course like this.

I do have one warning about this course, however. The learning curve for the programming aspect of it is very steep. It says no previous knowledge of R is required, but I don't think I would've been able to finish my final project if I hadn't already taken about 15 other courses (mostly on DataCamp) that focused on R programming. At the very least, you should take the free Introduction to R course on DataCamp before you start any of the labs for this course. Ideally, you should also get a DataCamp membership and work through about 60% of the courses on the Data Scientist with R track before even starting this course.

I realize that sounds like a lot to ask, and that I am contradicting the course description that was written by the instructors, but this course makes use of a great deal of the knowledge that is taught on that track, especially the dplyr and ggplot2 packages, from the first module. Dplyr in particular is a wonderful R package that does things I've always dreamed of while struggling to do basic things in Excel, but it takes a lot of practice to get the hang of it you will get much more out of this course if you already have some experience with dplyr (and ggplot2) before starting it.

I've audited several similar courses and found this one to be the best.

First of all, Dr. Mine is just so great at explaining things. There is no doubt that she's one of the best in her area, but she's also born to teach and communicate. She combines all kinds of way to make a concept vivid and clear. I've audited couple other courses, and I took relevant courses back in college a while ago, Dr. Mine is the best out of all the professors I've met at explaining things. This is not in this course but next, but just as an example of how clear she is when explaining standard deviation of sample means. She takes time to combine a specific example, visualization, and simulation, to really make all the points clear. You could try to listen to her on that part in the next course week 1.

Second, the R practice in every week is very beneficial and helpful. The cases used in those practices are fun to work with too. The hands-on experience on R and data exploring is valuable.

Overall, this is a very helpful course for me to review probability that I took a while ago in college and almost forgot, and for me to learn R and get hands-on practice.

This course is definitly suitable for learners who don't have any related background. Dr. Mine Cetinkaya-Rundel has an amiable speaking style and always highlighted the key points in teaching videos, which helped me understand other contents in the textbook. Besides, the time and assignment arrangement of this course are also very reasonable. The only thing I could complain about is a system bug of the Amazon AWS and the grading system. My final project file went blank after the system told me the file uploading is sucessfully completed on August 3, and I got three 0 point from three peers since they only saw a blank file, of which I had no idea, and all I could see is "Grading in progress" on the system. Until the final grading day is over, which is August 12, the system finally reminded me of this horrible thing. I came to mentors, disscussion forum as well as the help center, reuploaded my file, desperately tried to find peers who can still spare some time to review my file, and it is finally fixed today. I got my certificate in the end, but the grading process is really frustrating. Hope this bug won't happen to anyone ever ag

Teaching: I really like the clear and concise teaching style of lecturer and the wide range of simple real-life example used to explain the course content. I’m a social science student, given I’ve studied quantitative research methods before, this course is easy intro to and good refresher of data and probability theory. This course really gives me some confidence to continue to study probability theory, after finishing this specialization.

Textbook: The textbook used in this course is a good supplementary material, although it is not necessary to read the textbook. Course videos have already explained everything that we need to know at intro level. The textbook also covers some extra optional topics that are worth reading.

Course Structure: The course structure is well organized with clear focus in each week.

Assessment: The assessment of quiz in each week is relatively easy. The exploratory data analysis required in peer-reviewed assignment is relatively difficult for beginners. However, the course mentor has drafted an easy-to-follow guide in the discussion section which is really helpful for finishing this assignment.

I liked the course very much, as it really did a good job to refresh my previously learned concepts in basic statistics and probability. The use of RMarkdown was not that easy, but somehow manageable with trial and error, and the knitting to html was also a little cumbersom.

My main problem was that the week 5 assignment was not very clear and demanded more from the inspiration and creativity of the student in coming up with original research questions than on how to handle them and put them in R. As a consequence, the line of thinking risked to be reversed: first you realize what you can handle in 'R' and then, you create a question that can be handled... Perhaps, I would have preferred a first research question, formulated by the teachers, where only the concepts and programming already taught are to be used, before letting us free in the wild. Another consequence of the total open question is that the more fancy assignments that I could review, were those that used more R packages/commands than we were supposed to know already... But still, overall, it was a good course, sometimes challenging, which is good!

Best statistics course I've ever taken. So many Aha! moments I can't count them.

I have struggled for years to understand and get the hang of statistics, at uni, with online courses and at work. With this course (and the following courses) I think I have finally gained a DEEPER understanding of some of the basic but very important concepts of statistics. Lots of detailed examples and no overly complicated maths gibberish (although still mathematically sound!).

The R programming bits run in parallel to the statistics lectures and can be followed (necessary for a certificate) or can be ignored (if you only want to grasp the concepts), but are overall very easy to understand and follow. There is only little background to R as a programming language and the different types of data, lists, matrices etc. To me that's a good thing, as it allows you to use R right away (which in turn makes me more motivated to go back and learn more about R).

I whole-heartedly recommend this to anyone who wants to understand and use statistics!

Very well put-together course.

I like that the course has in-video quizzes as well as practice exercises to help prepare you for the weekly quizzes. The labs for the course are also very helpful.

The textbook that accompanies the course is freely available in pdf format online and the suggested exercises are a great complement to the rest of the course materials.

For those unfamiliar with R, the project is a bit of a leap from the rest of the contents in the course. To get around that, I'd suggest to both use the discussion forum (posts by mentor David Hood are particularly helpful) and to take both the R programming course and the Exploratory Data Analysis course from the Johns Hopkins data science sequence. Those 2 should together be doable in 5-6 weeks and at that point you should have sufficient background to where doing the project in this course (and those in follow-up courses in this specialization) should not be a problem.

Before starting the course, I was not expecting it to be ghat much effective. The statistics classes were informative. However, the main reason that I took this class was because of R studio and implementation of statistics via R studio. First 2 weeks were perfect, the only lacking part was the 3rd week (I think R assignment could be better. Obviously, we did not need much effort to complete Week 3 R assignment which I did not like). Week 5 project, on the other hand, was just on point. We could apply everything we learnt throughout the course; especially, reviewers were strict and gave constructive feedback which I liked a liked a lot. One more thing, some people might want to start learning R studio via video tutorials, which this course does not have. However, R markdown was very good method for learning purposes. Overall, very solid one. Thanks!

I think this is a very good entry level course for those who are interested in entering the realm of statistics.

The learning objectives of each week are well defined and the practice and weekly tests are based on those learning objectives. The videos explain very well each objective in a very convenient, easy to comprehend and interesting manner. For students who want more 'after class' material, the course offers a very nice book, which I personally used and helped me a lot during the course. The book also offers practice tasks at the end of each chapter.

The course project: Personally I think the course project in the 5th week of the course is interesting in a way that you have actual data to work with and use almost everything you have learned during the course.

Overall I think this is a very good course!