Coursera's online classes are designed to help students achieve mastery over
course material. Some of the best professors in the world - like neurobiology
professor and author Peggy Mason from the University of Chicago, and computer
science professor and Folding@Home director Vijay Pande - will supplement your
knowledge through video lectures. They will also provide challenging
assessments, interactive exercises during each lesson, and the opportunity to
use a mobile app to keep up with yo...

Coursera's online classes are designed to help students achieve mastery over
course material. Some of the best professors in the world - like neurobiology
professor and author Peggy Mason from the University of Chicago, and computer
science professor and Folding@Home director Vijay Pande - will supplement your
knowledge through video lectures. They will also provide challenging
assessments, interactive exercises during each lesson, and the opportunity to
use a mobile app to keep up with your coursework. Coursera also partners with
the US State Department to create “learning hubs” around the world. Students
can get internet access, take courses, and participate in weekly in-person
study groups to make learning even more collaborative. Begin your journey into
the mysteries of the human brain by taking courses in neuroscience. Learn how
to navigate the data infrastructures that multinational corporations use when
you discover the world of data analysis. Follow one of Coursera’s “Skill
Tracks”. Or try any one of its more than 560 available courses to help you
achieve your academic and professional goals.

Learn how to draw conclusions about populations or scientific truths from data. This is the sixth course in the Johns Hopkins Data Science Course Track.

Instructors

Instructors:
Jeff Leek, Roger D. Peng, Brian Caffo

University

University:
Johns Hopkins University

Instructors

Instructors:
Jeff Leek, Roger D. Peng, Brian Caffo

Rankings are based on a provider's overall CourseTalk score, which takes into account both average rating and number of ratings. Stars round to the nearest half.

I had been enjoying this specialization up until this course. I haven't thought about stats in a while so I am having to use google to learn the concepts that are poorly explained in the lectures. The content and course projects are interesting but the delivery is pretty awful. Can't wait to get through this course and move on with the specialization.

I never review classes, but this is just unbearable. The lecture and course material do not lead you to a place where you can understand the theory and apply it to the questions. This course will not prepare you to perform statistical inference, but it will force you to memorize lots of formulas, but good luck guessing which one to use, because this course won't teach you to figure that out for your self.
I think it is a disservice to any students or future employers to include this course in the specialization. Either spend the time to teach fundamentals, or spend the time to teach advanced theory and have prerequisites. Trying to squeeze all of Stats into a four week MOOC is doomed to be a muddled failure.

I'm so disappointed with this course. Normally I won't give review even if i like it or not. I was so disappointed that i needed to give review. The author is like a reading machine. He reads text and formula. There is no explanation at all. I don't know whether he will be able to understand if he goes through his video. My humble opinion to Data Science specialization is to remove these two courses and start creating the course with different author. I'm going through different online site to learn the statistical inference.

If you are familiar with the topics presented, it's most likely too easy. If not, the instructor only provides marginal explanations that will make you understand it. It only shows that the instructor knows his stuff. He is not able to bring the content across.

Much of the lectures videos were poorly done. The slides were just snippets of code and formulas. The instructor did not provide clear explanations that would aid any student who doesn't have advance statistical knowledge in understanding. I spent most of the time googling for other videos to watch just to pass this course. Roger Peng did a great job with previous courses. This one pales in comparison in term of the quality of materials delivered. I'm glad I got through it without giving up.

I completed the first two courses in the data science specialization. While I think there were some significant limitations to those courses as well (assuming a higher level of knowledge than was indicated, boring lecturers, assignments that weren't covered by the content), this one is much worse. I was planning on eventually doing the data science specialization, but after the first week of this course, I requested a refund and have decided to slap together other similar courses that are not so terrible. If you are a formal statistician and you want a boring half review, this course is for you. Otherwise, I would steer clear...

I think Brian Caffo is doing a fantastic job. I learned quite a bit . For instance I never know at that variance formula shortcut before. Moreover , the material is supposed to be complex , this is a statistics course and there are many nuances and exception Monseigneur Caffo is trying to convey.Let's focus on the learning and less on pointing fingers. Thank you! Please do keep up the great work Brian

I passed this course with a %100 grade, but I might as well not have taken it at all. This course is TERRIBLY done. Firstly, from a delivery standpoint, it's horribly unpolished. The concepts are introduced in a rushed half-complete way, lectures often start and stop mid-sentence, the mathematical notation is incomplete and sometimes wrong, many of the SWIRL programming assignments throw errors and terminate halfway through the assignment, the class notes are very messy, and in some of the homework, you're asked to answer identical questions twice. As a lecturer, Brian Caffo wanders and stutters a lot and breezes through very incomplete explanations of crucial probability and statistical topics. When discussing how to implement statistical methods (t-tests, ANOVA pdf functions, probability distributions) in R, he doesn't really give an organized introduction to it, he simply plops code snippets into his lecture slides and stumbl...
I passed this course with a %100 grade, but I might as well not have taken it at all. This course is TERRIBLY done. Firstly, from a delivery standpoint, it's horribly unpolished. The concepts are introduced in a rushed half-complete way, lectures often start and stop mid-sentence, the mathematical notation is incomplete and sometimes wrong, many of the SWIRL programming assignments throw errors and terminate halfway through the assignment, the class notes are very messy, and in some of the homework, you're asked to answer identical questions twice. As a lecturer, Brian Caffo wanders and stutters a lot and breezes through very incomplete explanations of crucial probability and statistical topics. When discussing how to implement statistical methods (t-tests, ANOVA pdf functions, probability distributions) in R, he doesn't really give an organized introduction to it, he simply plops code snippets into his lecture slides and stumbles through them without explaining much about what key R stats functions are and the nuances of using them. In the previous courses of the track that Roger Peng teaches, you get used to very well rounded explanation of R functionality. Do not expect the same in this course. Because of how poorly done this course was done, I've decided NOT to complete the rest of the John Hopkin's data science track. I recommend to the creators of this track that Brian Caffo be removed from this track and the statistical inference course completely reworked by another instructor.

The instructor is terrible, he simple talks through equations and does a poor job of relating the information to people without a statistical background. Very disappointed in this course so far.

Such an impractical way of teaching statistical inference. Like others, I too have taken other courses in the Data Science Specialization and felt that I completed those courses with a solid practical understanding of the course material, which gave me new tools and skills in my professional life. Statistical Inference and Dr. Caffo's other course, Regression Models, merely provided an outline of what I had to go and look up and learn elsewhere, because I found so many of his lectures totally un-intuitive. He gets too wrapped up in the math, and it makes the course material frustrating and impenetrable, when actually many of these concepts are not so complicated. Disappointing.

Brian Caffo was not very articulate in explaining the core principles of Statistical Inference. Instead, it seemed to be one long list of formulae after another with very little in the way of interpretation or intuition as Andrew Ng does so well in his Machine Learning class. In many ways, this course was the lowest point of the Data Science Specialization offered by Johns Hopkins University.
Be prepared to look up material outside this course to really understand the subject and apply the learnings.
You can pass the course itself fairly easily by judicious use of google and the course ware but whether you really assimilate the subject matter and can make use of it is another matter entirely. Some of the other courses in the specialization do give you that learning (especially R programming, Exploring Data, Developing Data Products)

Very poor delivery. Assumes that the person taking the course is knowledgeable on the subject. The instructor is obviously reading from pre-prepared material with a monotone. Extremely uninteresting lectures. No explanation and the instructor even says that we should simply take his word for without questioning, One of the worst lectures.

too many concepts to cover in short period of time. Concepts are hard to follow and understand. For those that do not understand statistics or mathematics, even worst Quizzes and projects too difficult especially those that are not full time eg working. There is not interactions with the instructor whatsoever. Try to understand from writing is difficult. Instructor spoke too fast and too conceptual and do not give plain example or scenarios.

These reviews seem to be from students who were disgruntled because they couldn't pass, encountered more work than they expected, or expected the instructor to provide scintillating real world examples.
I though the class was excellent and well-integrated with the Data Science Track.

I thought it could be a good opportunity to go over some basic statistic concepts in a short time but it turned out as a total waste of time. I will just follow the Duke's course. The instructor and content are not really ready for such a course. Those courses from John Hopkins really lowered the reputation of the university as well. I feel sorry for the students actually paying thousands of dollars to this school.

There is a massive disconnect between this course and the previous courses in the Data Science specialization.
The previous course (which mainly teach R programming) basically start from scratch, assuming no prior knowledge. If you take them in order you build up from zero to more and more competent.
This course, on the other hand, absolutely cannot be taken by someone who has not already studied statistics. Dr Caffo constantly refers to concepts which he has not previously introduced, provides half explanations and often times simply recommends remembering things by heart (instead of explaining them - which would be a whole heap better).
If you first went through everything about probability and statistics at Kahn Academy (where the explanations are awesome) then you might be able to follow this course... although everything of value that you know at that point - you learnt on Kahn Academy!
It's a real pity. It looks l...
There is a massive disconnect between this course and the previous courses in the Data Science specialization.
The previous course (which mainly teach R programming) basically start from scratch, assuming no prior knowledge. If you take them in order you build up from zero to more and more competent.
This course, on the other hand, absolutely cannot be taken by someone who has not already studied statistics. Dr Caffo constantly refers to concepts which he has not previously introduced, provides half explanations and often times simply recommends remembering things by heart (instead of explaining them - which would be a whole heap better).
If you first went through everything about probability and statistics at Kahn Academy (where the explanations are awesome) then you might be able to follow this course... although everything of value that you know at that point - you learnt on Kahn Academy!
It's a real pity. It looks like Dr Caffo has gone to a lot of effort to prepare various materials, a book and a swirl tutorial. However that sheer volume of material in no way compensates for the lack of organization of the material. It's like he's trying to paint the walls before having built the house.
I was planning on finishing the Data Science specialization but having taken this course it feels like it's just too much of an uphill battle to take anything useful away and my time would be better spent on a pure stats course elsewhere, that was better organized.

Sorry Brian, I know you describe your teaching style as different than Roger and Jeff, but my honest opinion is that this is a nice way of saying they are helpful and you are not. I'm on my second course with you and have several with each of them and other coursera instructors. I considered leaving the verified capstone curriculum because of how bad your courses are. Please take this a constructive, I'm sure you are competent in your field and a great person but teaching is an art and you are missing the target. Instead of focusing on math proofs (I'm a mathematician and the walk through of the proofs are not providing intuitions), focus on R and the relevant packages to get the job done, Then provide a high level intuition, not a lot of examples of how unclear ideas can be expressed equally as unclear ideas. In the end I'm sticking it out, but I'm not even going to bother with your videos or lecture notes and I feel like my m...
Sorry Brian, I know you describe your teaching style as different than Roger and Jeff, but my honest opinion is that this is a nice way of saying they are helpful and you are not. I'm on my second course with you and have several with each of them and other coursera instructors. I considered leaving the verified capstone curriculum because of how bad your courses are. Please take this a constructive, I'm sure you are competent in your field and a great person but teaching is an art and you are missing the target. Instead of focusing on math proofs (I'm a mathematician and the walk through of the proofs are not providing intuitions), focus on R and the relevant packages to get the job done, Then provide a high level intuition, not a lot of examples of how unclear ideas can be expressed equally as unclear ideas. In the end I'm sticking it out, but I'm not even going to bother with your videos or lecture notes and I feel like my money was wasted on your two courses (again not because you aren't competent in your field or a good person) because you are not focused on the practical aspects of teaching. Please let Roger and Jeff peer review your courses and completely redo them with you. I use peer review at work heavily and even though it can be an ego blow, letting others direct and accepting it as a learning experience is crucial to serving your clients.

Way too much stuff for a 4-week course. I went away for a week and thus dropped it. But I was invited to do CTA so I will be doing that instead for a couple of months, while I read the statistics textbook! I aced the first quiz, mostly by research, of course. I had statistics in grad school back in 1990, but it was just as hard to remember this time.
Maybe I will skip this course altogether and take one of the others that people have recommended! I have otherwise been very happy with my Data Science courses. (The lecturer makes the course, though; I stopped my initial course on volcanoes because the lecturer was impossible to listen to.)
Note that I am taking these courses for fun; I am retired and not looking for a job.

The instructor means well but I struggle with the manner in which he presents the material. I have taken the 5 courses that lead up to this one but this has been the hardest. I found the lectures provided by the course "Data Analysis & Statistical Inference" helpful in understanding this subject.

Many of statistical inference topics are inherently bit abstract, but this course makes them even more dry. Brian please put some effort by taking some good example - have a look at khanacademy how sal makes it interesting. In fact look at your own HW help videos. Once you bring in that flavor and add the context of data science and r environment you will have a great course but in current shape it is not at all professional. Had to give half stars as I'm forced by this form.

This course has changed a lot it is now much better fully deserves half a
star. In the new version of this class the videos are much improved, now
there's a guitar in the background that randomly appears and disappears as the
lecturer mumbles over formulas without really telling you what is going on.
There's also a "project" where you have to do a simulation and then some work
that you really never learned in the lectures. You need somebody gifted to
explain statistic topics to a general audience and make them really understand
the theory, apply it to the practice and make it fun. This is not the case.
The examples include diseases, tooth grow, height of children and more
diseases. You have to work hard to find examples less attractive than those
for a class.

I was expecting to get reacquainted with the subject matter and be able to
utilize it in my field (and other fields) of study/research. I did. I was
actually able to have a better understanding on some areas where I had doubts
before. The course is being shaped to be able to reach those with little to no
knowledge on the subject matter or for those who have found the subject matter
difficult in the past. I believe the course is nicely getting shape.
The lectures go through lengthy theoretical explanations, which may seem
difficult for some unfamiliar with the subject. However, these explanations
are followed by many case studies in which you can see and understand how the
statistics work conceptually and also how to obtain such statistics and apply
them to the questions being asked. I give specific attention to the fact that
with this course, the student will understand "when to apply what and for what
purpose".
As with all other ...
I was expecting to get reacquainted with the subject matter and be able to
utilize it in my field (and other fields) of study/research. I did. I was
actually able to have a better understanding on some areas where I had doubts
before. The course is being shaped to be able to reach those with little to no
knowledge on the subject matter or for those who have found the subject matter
difficult in the past. I believe the course is nicely getting shape.
The lectures go through lengthy theoretical explanations, which may seem
difficult for some unfamiliar with the subject. However, these explanations
are followed by many case studies in which you can see and understand how the
statistics work conceptually and also how to obtain such statistics and apply
them to the questions being asked. I give specific attention to the fact that
with this course, the student will understand "when to apply what and for what
purpose".
As with all other materials and courses, finishing this course will by no
means make the student a specialist in Statistical Inference, but the student
will gain the necessary information to start working with SI and Data Science
projects.
In my experience, it is with continuous work in the related field that a
person becomes specialized in that field. Taking this course (stand alone and
with the Data Science Specialization) will give you the information necessary
to keep learning and applying what is learned.

The team that teaches this specialization is some of the best and the most
brilliant researchers in the field of bio-statistics. Their intention to offer
this type of specialization, and in particular this Statistical Inference
course is highly commendable. However, the content is "densely" packed into a
brief 4-week course. This makes the learning experience a difficult one for
people who are trying to acquire depth in the subject matter. To add to this
challenge, the style of presentation is very verbose, and it is too difficult
to follow the densely packed slides and the delivery without feeling stressed.
I found this course to be very valuable in advancing my understanding of the
subject, and working with R. There is a lot of ground covered, and I can't say
much of it was "internalized." But, there are core set of concepts that really
sunk in, especially through the project exercises.
If someone is looking for a pedagog...
The team that teaches this specialization is some of the best and the most
brilliant researchers in the field of bio-statistics. Their intention to offer
this type of specialization, and in particular this Statistical Inference
course is highly commendable. However, the content is "densely" packed into a
brief 4-week course. This makes the learning experience a difficult one for
people who are trying to acquire depth in the subject matter. To add to this
challenge, the style of presentation is very verbose, and it is too difficult
to follow the densely packed slides and the delivery without feeling stressed.
I found this course to be very valuable in advancing my understanding of the
subject, and working with R. There is a lot of ground covered, and I can't say
much of it was "internalized." But, there are core set of concepts that really
sunk in, especially through the project exercises.
If someone is looking for a pedagogically charming course, the one offered by
Duke is very good. I find that they both get at the subject from different
perspectives, and I learned different things from both of them.
Some of the negative comments of example topics (health related) are unfair.
The school focuses on biostatistics and the professors are experts in this
field. You can't blame them for picking examples from their field.

The first half of the class I learned almost nothing since I've taken
statistics as a math major (albeit a few decades ago).
Somehow I passed this class, but I do not consider that a mark of learning.
Don't take this course. Take the one from University of Texas on EDX called
Foundations of Data Analysis. Or failing that, the one on EDX from Karolinska
Institute (Exploring Statistics with R), or possibly the Coursera one from
Duke (though that is a much longer MOOC).

Many other reviews have already outlined the piss-poor quality of teaching
that Brian Caffo has provided, so I won't re-hash their valid complaints much.
He jumps all of the place and provides no real scaffolding (interesting how
much I've learned about teaching approaches in this track, due to the
ineptitude of all three professors). I completed the first 5 courses with
"with Distinction", but after seeing this class I'm moving on to courses on
other platforms that are wildly better (Udacity has a "nanodegree" in Data
Science that will be available this fall that sounds much better and the
quality of their courses far surpasses Coursera). The lack of support from
Coursera also leaves a lot to be desired. They tell you to take your issues to
the forums. The professors for this track do not participate in the forums at
all.

Statistical Inference is the 6th course in the John Hopkins data science
specialization track, which is basically an introduction to statistics in R.
The course covers many different topics in the span of 4 weeks from basic
probability and distributions to T tests, p values and statistical power. The
lectures take the form of slideshows with a lot of dense mathematical
notation, small text and mediocre voiceovers. The course tries to cover too
much ground too fast and the material isn’t presented in a way that is easy to
understand or engaging. I don’t think the lecturer’s face was shown once in
the entire course. That’s not to say there isn’t good information in the
lecture slides, but the presentation and execution are poor. If you’re looking
for a good introduction to statistics that uses R, try Duke’s Data Analysis
and Statistical Inference. Udacity’s “Statistics” is another solid option that
is self-paced, moves a bit slower and...
Statistical Inference is the 6th course in the John Hopkins data science
specialization track, which is basically an introduction to statistics in R.
The course covers many different topics in the span of 4 weeks from basic
probability and distributions to T tests, p values and statistical power. The
lectures take the form of slideshows with a lot of dense mathematical
notation, small text and mediocre voiceovers. The course tries to cover too
much ground too fast and the material isn’t presented in a way that is easy to
understand or engaging. I don’t think the lecturer’s face was shown once in
the entire course. That’s not to say there isn’t good information in the
lecture slides, but the presentation and execution are poor. If you’re looking
for a good introduction to statistics that uses R, try Duke’s Data Analysis
and Statistical Inference. Udacity’s “Statistics” is another solid option that
is self-paced, moves a bit slower and does not require programming.

This is my review for the entire Johns Hopkins Data Science specialization on
Coursera. These comments apply specifically to this course and generally to
all the courses. At this point I've only seen six of them but I imagine the
three yet-to-be released courses will be similar. First this was a great idea.
A grand introduction to data science basics and methods from some real
experts. However the execution was sorely lacking. Bottom line the courses are
superficial and not worth the time compared to other data science mooc
offerings. The courses are at best very light introductions. Are you going to
learn statistical inference in a 4 week course? No. Machine Learning? No. R
programming? No. Most people realize that but maybe not everyone does. I feel
sorry for the students who take these courses and afterwards believe they have
any real knowledge of these subjects. Besides the thin content, the
instruction itself is bad....
This is my review for the entire Johns Hopkins Data Science specialization on
Coursera. These comments apply specifically to this course and generally to
all the courses. At this point I've only seen six of them but I imagine the
three yet-to-be released courses will be similar. First this was a great idea.
A grand introduction to data science basics and methods from some real
experts. However the execution was sorely lacking. Bottom line the courses are
superficial and not worth the time compared to other data science mooc
offerings. The courses are at best very light introductions. Are you going to
learn statistical inference in a 4 week course? No. Machine Learning? No. R
programming? No. Most people realize that but maybe not everyone does. I feel
sorry for the students who take these courses and afterwards believe they have
any real knowledge of these subjects. Besides the thin content, the
instruction itself is bad. The instructors often seem to just ramble as if
they haven't prepared at all. And when an instructor says 'You can learn about
this on Wikipedia', I can't help but feel like 'What am I listening to you for
then?' Dr Peng's lectures were better than the other two but this was still my
overall impression. I've taken and passed many other Coursera and edX moocs.
Usually the content and instruction is excellent. I'm sorry to have to write a
negative review but tbh these courses were simply a waste of time, especially
when you consider the many excellent alternatives, like Data Analysis and
Statistical Inference from Duke, Machine Learning and Statistical Learning
from Stanford, and The Analytics Edge and Introduction to Probability from
MIT. Update July 2014: It's my understanding that these courses have been
revamped. I have not taken any of the new offerings. So the criticisms I made
above may not apply anymore.

The course was very similar to reading an extremely boring mathematical
statistics handbook. The presence of the lecturer did not bear any additional
value, as he was not making too much effort to use educational methods to
convey the material. He used very few (and generally uninteresting) examples,
and did not structure the lectures to capture attention. The explanations were
purely mathematical, thus difficult to comprehend, at least for me with a non-
math background. The lecturer clearly not designed this course to an "educated
general audience" (i.e. I presume most mooc takers).
In order to understand the core concepts, I finally ended up on other moocs
and sites on the same topic. BTW some of these external sources used no
complex formulas, but were able to explain key concepts in very short time,
using examples and graphical presentations. I hope this course will improve in
the future, there is certainly a lot of room for ...
The course was very similar to reading an extremely boring mathematical
statistics handbook. The presence of the lecturer did not bear any additional
value, as he was not making too much effort to use educational methods to
convey the material. He used very few (and generally uninteresting) examples,
and did not structure the lectures to capture attention. The explanations were
purely mathematical, thus difficult to comprehend, at least for me with a non-
math background. The lecturer clearly not designed this course to an "educated
general audience" (i.e. I presume most mooc takers).
In order to understand the core concepts, I finally ended up on other moocs
and sites on the same topic. BTW some of these external sources used no
complex formulas, but were able to explain key concepts in very short time,
using examples and graphical presentations. I hope this course will improve in
the future, there is certainly a lot of room for that.
From the third week, the course started to become better, even useful in the
fourt week! So hang on!

This was obviously a weak course in data science specialization. I am not
going to repeat prior posts comments. However, the professors are going to
improve the course. I hope it will reach an acceptable level.
I completed the first two weeks quizzes just with my knowledge and course
slides and not watching videos. For the third and fourth quizzes I used slides
only .I am expecting to get 75 out of 100 but frankly saying I did not learn
anything. Just passed it. I can spend my time on Duke's Data Analysis and
Statistical Inference which was highly recommended on coursera forums.
If you are going to complete data science specialization track , you will have
another alternative from John Hopkins University. Go and check it out.

Rankings are based on a provider's overall CourseTalk score, which takes into account both average rating and number of ratings. Stars round to the nearest half.