デューク大学（Duke University） による Linear Regression and Modeling の受講者のレビューおよびフィードバック

This course was good. However, compared to the other courses in the specialisation had less content. I would have liked to have videos on logistic regression as well.

Very useful insights and lea

Excellent Course. Mine, the teacher is a great great teacher. The mentors help a lot.

Technical parts, coursera platform should work better

Great introduction to linear regression. Nice, clean R tutorials via the labs. The lectures do become a little monotonous, but there there are linked readings in a nice, open-source textbook if reading suits you better than listening.

nice lecture, but it is really too short, not into too much details.

Really enjoyed this course! It teaches you the theory you need to understand how a linear regression model works, how to check that your model fulfils certain conditions so that it is valid, and how to build and implement your model in practice!

Good but I felt some gaps in the material made it difficult to learn. Also, the quiz questions are focused on attention to detail "gotcha" questions. This can be frustrating.

An extremely good introductory course. A must for undergraduates. The style of teaching is fluid and you learn concepts step by step. For more advanced learners the only drawback I see is that this is, by default, an introductory course.But still for advanced learners it can be a great (and I really mean great) refresher.

excellent

First, this course will enable me to understand the quantitative part of a research. Additionally, this will help a student to understand the essence of performing such numerical calculations and will make us understand the relationship between different variables.

Secondly, this is the need of the hour and such numerical functions are used worldwide so, learning this course will help in almost every field be it 'Management' be it 'Social Sciences' or be it 'Human Behaviour'.

It was a really good introduction to Linear Model, I recommend this course to all people who wants to learn more about statistical analysis

This is a great course for this specialization but don't expect much depth.

Great course. I only wish there was more time spent on dealing with more complex situations such as overfitting.

Good overview of regression modeling. Would have liked to see more on logistic regression. But that's ok, can read it on my own.

Good, but a little "smaller" than the Inferential statistics course (which is very complete). I would have liked to also learn Logistics regression, which I now have to learn elsewhere.

A good course for Bayesian statistics.

Could be more informative

I liked, but I guess it could improve little by including more topics in linear regression analysis.

Contents are easier compared with other courses in this series. Quite systematic and easy to understand.

Clearly presented. R instruction is pretty minimal, so there is a lot of trial and error and googling.

The course is good and the material is presented clearly. The capstone project is very good and makes you really use all the knowledge obtained in the course and the pre-prequisite course Inferral Statistics. My only dissatisfaction is that the course was rather short: only 3 weeks of material and 1 capstone. Therefor it covered less material then I expected. For example, I expected logistic regression to be covered.

Only contains linear regression as opposed to other model types

Provides a good overview, but I felt some loose ends were not addressed in terms of how stringent the conditions need to be met, and if one can use MLR when this is not the case.

Other nits about the depth and breadth of the course aside, I thought it was a good course. The main critique I have to offer is the lack of emphasis of using the power of R. When teaching model selection, the course should have at least provided instruction—or at least a written resource—on how to write the R code for automating forward/backward selection by R^2.* Being a course about using R as well as about linear regression and modeling, it seems like the appropriate thing to do.

(*A classmate whose final project I peer-reviewed used for loops to run the forward model selection based on R^2. That's how I learned about it.)

Covers the basic of Linear Regression, would like to see more advanced material.