Regression Models

As another one said before, I also survived mr. Caffo's courses. He probably is a good researcher and very intelligent man. But SUCKS as a teacher. Dropped the first time and retake it after using other books as sources, then I passed with 100%.

The problem with mr. Caffo, is that he rarely finishes his ideas. He starts talking about a statistical method and concludes nothing tangible or usable, or practical. For instance, when using anova to select models, I had to look different sources to understand how to read properly the F factor. He just enunciates possible tools and mathematical instruments and then expects us to use them to solve practical problems.

I was hard to apply his methods to the course, how hard it is going to be in the real life?

The instructor needs to be replaced.

I took this class last year and don't know if it's changed. I hope so. It tries to cover too much ground given it's only one month long. Another problem was the instructor, Brian Caffo, who seems like a good guy and good researcher, but not an effective teacher. If you love lectures that consist of long proofs and derivations, you'll love Professor Caffo. I prefer more practical lectures, where you're shown how to apply what you're learning to the real world. There was none of that.

Fortunately I already had knowledge of some of the topics covered. Otherwise, I would have been in deep trouble. As it was, I had to take the class twice because I wasn't getting it the first time. In fact, I had to take all of Professor Caffo's classes twice. That wasn't necessary with the other two instructors in the Data Science specialization.

I'd recommend the University of Washington's class on Regression in its Machine Learning specialization, instead.

I agree with a comment above - this class should ideally be completely redone (with a different instructor). The emphasis is on derivation of formulas and techniques, not applications to the real world. Also, the course "textbook" is significantly inferior to the free OpenIntro textbook. Course quizzes and the project were unclearly specified and quirky in focus, while the project's peer grading was a bit random. I survived this course (and Inferential Statistics) by taking a normal university course on statistics simultaneously. Also, I followed up on this one by reviewing logistic regression in the OpenIntro textbook.

As an aside, the Inferential Statistics and Regression Models courses seem almost completely detached from the rest of the sequence, which is a shame.

I've struggled a lot to figure out what are the points of the topics which are explaining in this course but actually it wasn't any relation between them neither any practical usage.
Despite I've learned so many statistics stuffs by having another course by Brian Caffo but in this one, I sometimes felt some hesitations in his voice on how to explain the topic and that just makes me more and more confused (I'd say I have engineering background and not having a strong math knowledge) and in order to have better understanding of the course I also tried to review it more multiple times but even that doesn't change any difference.
So, I dropped the course in the middle of week 3 and switch to another course by the University of Washington which is part of machine learning specialization path.

Regression Models is the seventh course in the Data Science specialization. As with Statistical Inference, it is taught by Brian Caffo and suffers from the same issues as the preceding course. The course covers least squares, simple linear regression, multiple linear regression, regression model diagnostics, and logistic and poisson regression.

The material here is rather strangely-presented. As with Statistical Inference, it is light on intuition, so students will have a hard time applying techniques learned here in an appropriate way without previous experience. Although a fair amount of math is presented for each topic, it is not at a level deep enough for students to really grasp how it works, so the course tries to walk a line behind mathematical rigor and application and mostly fails at both. The course project was relatively straightforward, using the mtcars dataset in R to predict miles per gallon by transmission type (automatic versus manual), with adjustments for other variables in the dataset. Overall I think I spent less than 10-12 hours on the entire course, and I technically took this concurrently with Statistical Inference, although I enrolled in this course only after completing the majority of the previous course.

Overall, two stars. Probably the least useful course in the specialization so far. I have some previous experience in linear regression and have taken a graduate-level course in the area and have published several papers using these techniques, and I did not find this course to be particularly intuitive or useful. A new specialization (Statistics with R) from Duke will include a course covering these topics, so this may be a better choice for someone wanting to learn how to apply these techniques to his/her own data.