This statistics and data analysis course will pave the statistical foundation for our discussion on data science.

You will learn how data scientists exercise statistical thinking in designing data collection, derive insights from visualizing data, obtain supporting evidence for data-based decisions and construct models for predicting future trends from data.

Syllabus

Week 1 – Introduction to Data Science

Week 2 – Statistical Thinking

Examples of Statistical Thinking

Numerical Data, Summary Statistics

From Population to Sampled Data

Different Types of Biases

Introduction to Probability

Introduction to Statistical Inference

Week 3 – Statistical Thinking 2

Association and Dependence

Association and Causation

Conditional Probability and Bayes Rule

Simpsons Paradox, Confounding

Introduction to Linear Regression

Special Regression Models

Week 4 – Exploratory Data Analysis and Visualization

Goals of statistical graphics and data visualization

Graphs of Data

Graphs of Fitted Models

Graphs to Check Fitted Models

What makes a good graph?

Principles of graphics

Week 5 – Introduction to Bayesian Modeling

Bayesian inference: combining models and data in a forecasting problem

Bayesian hierarchical modeling for studying public opinion

Bayesian modeling for Big Data

Taught by

Eva Ascarza , James Curley , Andrew Gelman , Lauren Hannah, David Madigan and Tian Zheng

by
A.b.completed this course, spending 5 hours a week on it and found the course difficulty to be very easy.

It's very unclear who this course is supposed to be for. It skims shallowly into some topics in the lectures, then dunks you into a long technical pdf that you have to read to answer the "quiz" questions. Luckily it's easy (if you're a native English speaker) to skim the technical articles and guess the answers, so I got a good grade despite only understanding bits and pieces. The assignments remind me of high school busywork.

You are often presented with new equation with one worked example, then quizzed only once on it before moving on to something else. There are no projects o…

It's very unclear who this course is supposed to be for. It skims shallowly into some topics in the lectures, then dunks you into a long technical pdf that you have to read to answer the "quiz" questions. Luckily it's easy (if you're a native English speaker) to skim the technical articles and guess the answers, so I got a good grade despite only understanding bits and pieces. The assignments remind me of high school busywork.

You are often presented with new equation with one worked example, then quizzed only once on it before moving on to something else. There are no projects or any practical work that unifies the very different topics discussed in the course. Don't expect feedback from admins or assistants or instructors in the discussion questions. The admins do not respond at all to numerous student requests to correct errata in the site, like the correct answer not being available to choose.

The professors will write equations and say "you don't really have to know the equation", or will tell you abstractly about the ideology they used to do an analysis without describing what the analysis is. The result is, you don't learn the intuition, and the ideology doesn't make sense without context.

Pros: some of the lectures, particularly those criticizing classical/Frequentist statistics, are really interesting. But they only hint at something deeper and do not go into depth, and do not guide you toward entry-level external materials that you can pursue on your own.

Overall: it's an interesting broad survey of topics in data analysis, but don't expect to learn how to actually do even simple data analysis. There is essentially no hand-on component of the course. Much of the meat of the course is in a grab-bag of links to external articles which are either very simple or very technical.

by
Nan is taking this course right now, spending 4 hours a week on it and found the course difficulty to be medium.

This class is a mess - we're unable to download lecture videos and transcripts, the powerpoint slides are not available, the quiz policy has changed midstream (you can now retake once instead of zero retakes). The quizzes are confusing and no feedback is available giving reasons for right and wrong answers. Very little problem solving or real life examples - mostly just professors lecturing at a blackboard - albeit a fancy blackboard that is transparent. Waiting for a professor to write out formulas while they mutter them to themselves is a waste of video time. I can't read the 'blackboard' most of the time if I also want to read the subtitles, and the primary instructor has quite a heavy foreign accent. I'm now looking for a better MOOC on statistics.

by
Ericdo1810completed this course, spending 1 hours a week on it and found the course difficulty to be very easy.

Honestly, I took this course out of curiosity. The name of the course is so catchy, I couldn't resist not to enroll.

When I watch the first videos, I was blown away. The videos were so good! They really did a good job conveying the topic of statistics in the context of data science.

Then I go to the quizzes, to my dismay, the quizzes were so ridiculously simple, one can hardly learn anything from it at all. They ask you to read news articles. Come on, anybody who's interested in Data Science, of course has read at least 5 papers about Data Science to know it is the ho…

Honestly, I took this course out of curiosity. The name of the course is so catchy, I couldn't resist not to enroll.

When I watch the first videos, I was blown away. The videos were so good! They really did a good job conveying the topic of statistics in the context of data science.

Then I go to the quizzes, to my dismay, the quizzes were so ridiculously simple, one can hardly learn anything from it at all. They ask you to read news articles. Come on, anybody who's interested in Data Science, of course has read at least 5 papers about Data Science to know it is the hot field right now.

Then the quizzes in the following weeks were ridiculously simple. I don't even need a college education or even high school education to do them. Maybe if you've learnt basic statistics in junior high, you'll be able to score full marks for all the quizzes.

The last weeks touch on Bayesian Methods, which is nice, at last.

However, I think the course learning is only summed up in the videos + the last weeks.

It was like watching lectures from youtube.

So please don't purchase the cert. It is 99USD. I would recommend you choosing other Data Science courses on edX or Coursera that are offer more rigorous training.

I have taken more than 10 MOOC, and this one is the worst one and really beyond awful. I was really looking forward to take all three classes in the series, but I decided not to continue after completing the first one. I think that the instructors didn't make much effort to design this class, instead they just grabbed random material from their own on-campus classes. I just want my $100 back.

I totally agree with Robert Ritz above. I already have a list of topics that I want to research as soon as I finish the course. This is an introductory course, paving the way to more in depth studying. Indeed you get what you put into this. And if you don't find these articles the slightest interesting maybe you are doing the wrong course.

I thought the lectures were useful for someone that is new to data science. I'm a bit surprised by so many negative reviews. I didn't think they were all that bad, but I am also a newbie so don't know what a "great" course looks like.

Not a single problem set and the main lecturer is unintelligible. I'm extremely disappointed with this course. Will have to finish unfortunately to get that Microsoft certificate. I would rate it zero if I could.

Far too light of a touch in general. Would really benefit from being longer, more in depth and with more practical / real world examples and projects. Worth taking if only for the Bayesian section taught by Andrew Gelman.

Others have said many criticisms of this course. Many of them are correct, but as is so often with learning, you get what out what you put into it. Is the class designed to turn you into a pro data scientist in 5 weeks. Absolutely not. Does it jump around in topics? Absolutely.

The reasons for this is the nature of data science itself. This course is teaching more the mentality and process of a data scientist as opposed to hard technical skills. Anyone can learn how to copy and paste commands into R. Not everyone will take the time to understand what it all means. This course is m…

Others have said many criticisms of this course. Many of them are correct, but as is so often with learning, you get what out what you put into it. Is the class designed to turn you into a pro data scientist in 5 weeks. Absolutely not. Does it jump around in topics? Absolutely.

The reasons for this is the nature of data science itself. This course is teaching more the mentality and process of a data scientist as opposed to hard technical skills. Anyone can learn how to copy and paste commands into R. Not everyone will take the time to understand what it all means. This course is merely 1 in a series of three, and is meant as a high level overview.

In college did you hate your entry level courses because they were boring/the teachers accent/the homework was way too easy? Did you drop out of college? I think your answers to these questions will give you a better idea of what to expect.

I personally learned quite a bit that extended my statistics and data analysis background. What you do with this is up to you.

by
Martincompleted this course, spending 1 hours a week on it.

looks fragmented. why do it when you're not committed to doing a good job?

"This class is a mess - we're unable to download lecture videos and transcripts, the powerpoint slides are not available, the quiz policy has changed midstream (you can now retake once instead of zero retakes). The quizzes are confusing and no feedback is available giving reasons for right and wrong answers. " agreed.