Data Science in Real Life is messy (About the course)

0
146

What do you think about differences between theoretical and real-life knowledge? What kind of concessions could you have because of the real world difficulties? How does one manage a team facing real data analyses?

Intriguing topic, isn’t it? I can definitely say that for me this questions still topical and interesting after completion of the last course Data Science in Real Life in Executive Data Science Specialization. Let us consider, what you can find in this one-week course expect the central idea that Data analysis in real life is messy =)

The course has the next structure:
– Topic
— Theoretical part – How it’s should be
— Practical part – How it happened

In general, I like this idea of structuring. The approach can help feeling differences between theoretical and practical approaches. However, the information un-specific and broad. There are no technical details, and you get general information during short timeboxes.
As a result, it makes hard to remember the course. I mean that the information from it is very quickly forgotten. After one week, I can hardly recall the details.

Course description tell is that the course can give information about:
1. Experimental design, randomization, A/B testing
2. Causal inference, counterfactuals,
3. Strategies for managing data quality.
4. Bias and confounding
5. Contrasting machine learning versus classical statistical inference

Looks great, but it just looks. In reality, I found too base information there. Too lightly.

I understand that during one week it’s probably impossible to give information about A/B tests, experimental design, strategy for manage quality, bias and other. C’mon, guys, we can spend more time on this, especially when we touch so exiting topics. We know what A/B tests are; probably we wanted to see more useful info and real-life examples.

In summary: the course structured good and can give you lightly information about topics. In the same time, the course is short, the information is too briefly and can be quickly forgotten.

PS:
By the way, good news, there is another specialization from the same authors: Data Science Specialization from Johns Hopkins University. There I found course Developing Data Products. Maybe this course can give additional information. However, this course is number 9 from 10, so you usually should pass other 8 before starting this one, and each course needs 4 weeks.

If you have found a spelling error, please, notify us by selecting that text and pressing Ctrl+Enter.

LEAVE A REPLY

Please enter your comment!
Please enter your name here