It's a very good course for starting to learn about linear regression. Just be aware that the quality of this course is a bit lower than the previous two. There are fewer videos, the book material is shorter (less suggested exercises and the chapters cover fewer things about linear regression) and some quiz exercises of week 2, which should only cover simple linear regression, have some questions about multiple linear regression which is the 3rd week's topic.

Also, as in the previous two courses, the emphasis is on statistics, not programming with R. This means that if you already know statistics and only want to learn how to use R, there are probably better courses out there for you. But if you want to learn or improve your knowledge of statistics, and also learn how to use R, then do take this course. I think that it's much better to start learning R by actually doing some statistical work and seeing first hand what the software is capable of doing with only a few lines of code, even if you don't fully understand the code's syntax at first.

With all that said, if you take the course PAY ATTENTION TO THE LECTURES, READ THE CHAPTERS and DO THE SUGGESTED EXERCISES. I can't stress this enough. If you don't do all of that, you won't learn as much as you should, and it's painfully obvious that some students didn't do all of that when you review their final R projects. Also, take your time with that final project because that's where you will actually learn some things about R and use what you have learned about statistics (you will have to use google to learn how to code some things properly).

The course is good regarding concepts and theoretical exercises, but poor regarding applying new knowledge in R. Since the course is introductory, an instruction how to install R and a list of R functions without clear explanation how they should be applied in general regression situations makes me explore other sources to learn how to apply those concepts (e.g. DataCamp, CRAN-RProject, etc) and then get back to learn theory? Sorry for expectations but course should provide a full and integrated package of knowledge and skills, especially for beginners.

Furthermore, no Machine Learning (ML) is covered as a tool to run a regression.

My proposal is to provide an algorithm with a comprehensive example how to run a regression using R. From data to final model, step-by-step.

The mathematical depth of this course, is insufficient even at its targeted level, and therefore a lot of practical manipulations of the data, and fine tuning of the model could be had if a week more has been put into this course.

Easy does not equate fun, after completing this course, I left the specialization.

The course is great and I got the certificate for finishing the specialization. I'm a bit sad though.

There used to be a 4-th and a capstone course in this specialization, but they removed them from the requirements. The courses used to be a bit more difficult with a final project, they made them optional too.

I really hope they can improve the course content instead of removing things. The sense of achievement drops greatly knowing the courses get watered down in difficulty.

Very good course. while it does not cover everything. the teacher does a great job explaining things in a simple manner. My feed back would be to move ANOVA into this module.

I enjoyed the course and learned a lot overall. However, we were extraordinarily unprepared for the final project. The dataset was full of characteristics that were explicitly not covered in the course. I had to do a ton of outside research in order to complete the project and understand what I was doing.

Additionally, the forums are completely unhelpful. I never got a reply to any of my questions, and saw many other unanswered questions while browsing.

Great instruction on stats, however the R portion a weekly project that is largely self directed, very little instruction.

I do like the Duke way of teaching statistic, very clear and easy to understand. The final project is interesting and you can learn a lot while doing it, but it won't be enough to do it using the knowledge from this course only, you need to learn from online researching along the way.

The course is structured in a very informative way, it is easy to understand and at the same time difficult concepts are presented in a very easy way. The course instructor is awesome.

fantastic course on linear regression, concepts are well explained followed by quiz and practical exercises.

though you need to complete the prior courses to understand this.

I really liked the way theoretical knowledge is combined with real study cases

Really great course, clear and easy to follow. Highlight recommended.

Very good course and material overall - although a little lighter in content compared to the previous one (Inferential Statistics). I think you get a good intro to linear regression and getting to apply the knowledge in the assignment is very helpful. Beware though as the message boards are empty and it's very unlikely that you will get any questions answered. That's a serious issue with this class.

Overall it covers everything that you would want this course to.

However, I was a complete beginner when starting this course, and as a result I regularly got confused (especially when it came to coding in R)

3/5 stars for me

Compared to other courses in the specification, this course content is too shallow and brief.

She just started with wk 2. There should have been more explanation and videos in week 1...not very interesting. I think statistics you need to take in person.

it provides a superficial knowledge. A deep understanding of subject can not be gain from this course

The course required no prerequisites, but no one can complete the course without actually completing the courses preceding this in the specializing series.

Not suitable for beginners

This is arguably the best online course I have ever done. The teacher and the way she drives home her point her spot-on. Until now, I have struggled with most aspects of linear regression, e.g model selection, model fitting and interpretation of the results. Undertaken this course has cleared all these shortcomings and I can't wait to start analyzing my PhD project. I appreciate the opportunity afforded me by Duke University and Coursera for participating in the course, it will indeed help me in my academic pursuit and lastly many thanks to the facilitator of the course Dr. Mine Çetinkaya-Rundel

I appreciate.

Adedayo Michael, AWONIYI