I really don't understand why this course is added to this specialization. It is not really linked to the previous courses, and the way to get the grades are all based on peer-reviewed assignment. There is no peers to review my work after submitting the first assignment and there are four such assignment to go.

This is the worst specialization I have taken so far on Coursera, and I have taken several, either from universities or private companies.

The first course is fine for someone who already has SQL knowledge and just needs a refresher, but it is still a course with zero depth in the use of SQL in data science. It is just slides, better explanations of SQL can be found on YouTube.

The second course shows an interesting use of SQL, which is not normally covered. But the structure, assessments, assignments and final project are very open and ambiguous. The explanations are superfluous and you can tell that the course was made in a hurry, with no attention to details and with deficiencies in teaching.

Perhaps the third course is the best, as they present a useful technology and platform.

The final project is a total disappointment, as it is not focused on SQL and its use in Data Science activities, the videos are largely a recycling of those seen in previous courses, so that SQL, which is what the specialization is about, you do not learn something new or have the opportunity to apply it as it really applies in Data Science. Better SQL projects are found on YouTube, but these do not give a certificate endorsed by an educational institution such as the University of California, Davis. Sad that the readings in week four are links to other courses, they did not bother to create material or even reference it in a better way.

Zero material is shared.

After this experience I think I will not take any other course or specialization of this University, as well as not recommend it, I finished it just to receive the certificate, but I think it is not worth spending time on it.

the course is just an assignment, without much guidance and student is supposed to capitalize on whatever they have learned in previous courses. the requirement to use Python is frustrating, it would have been better if the course would have remained focus on SQL. limited people are enrolled in course, it takes weeks for peer review.

It was Ok. It's largely down to you to get the project done to the highest standard you can muster... if you're stuck just use stackoverflow. But through that I learned a hell of a lot, just stick to your ambition with answering your question and plug away at trying to get the answer even if it means googling lots to work lots of stuff out yourself. If you do, you may have a project worthy of being shared with your network or even blogged about.

Some feedback: given this was part of a SQL specialization, it would've been interesting to ensure candidates created a database that they would query themselves, rather than just draw the ER diagram. But to be fair on Course 4, that wasn't taught in the courses before, just seems to me to be an opportunity missed?

Another feedback: it may have been more helpful for the examples to be based in Spark / Pyspark for them to have the feel they build on Course 3 of this Specialization. Again, a missed opportunity if people just reverted to Pandas and not use the Capstone to apply Course 3 teachings.

But as always and perhaps rightly it's up to the learner to construct their project in a way that's meaningful to their learning goals.

Thanks for the course

This course was mainly a Python-course instead of an SQL-course - not only in terms of examples and screencasts, but also the exercises can not be sufficiently worked through with SQL solely. If no pyhton / R / scala and visualization skills are already there, it would take far to much time to learn it in such a small course. Additionally, the exercises did not point out all the nessecairy aspect to fulfill to get the full score clearly in all cases: When they say "use 1 or 2 new metric" they actually mean: "Set up 2 metrics to get the full score", which is odd, because one might instead focus on other aspect on the analysis, when following the first formulation. Again (as in the 2nd course of the specialization) I did not find it fruitful to dive deep into a dummy dataset, that has nothing to do with our actual profession. Additionally, this is far from being presentable in a portfolio. I would suggest providing a specialization filled with the basics and techniques, leave the application up to the real use cases. This is not effectively teachable in a MOOC. Additionally, pointing out to paid courses from the same provider, as had been done in week 4, is not providing resources, it is advertising. I also disliked the fact that all assignments were peer graded. That slowed down progression dramatically and I do not benefit from unqualified feedback from people who are no more capable or even less capable than I am. Also the reviews often where simply wrong. If a course where skills from the previouse ones are being united - which btw. this course not really resembles of - is really needed - which I doubt - then I would suggest presenting more best practices and techniques of structuring data and a project and questioning these knowledge in quizzes.

I think the aim for the course encourage exploration using the learned skill. Which is fine, actually a great idea. But the execution is sloppy. I think the course need to be balanced between exploration and steering as well. It should be designed so that some insights was hidden in the data and with no particular instruction, the students should find out these insights. So they have meaningful goal that they reached as well as a handful of exploration experience from the lack of instruction.

Instead? We have 3 sets of data, which are too narrow even compare to the example given in the lecture. Also the lecture and the assignments are barely comparable, which is more confusing. The peer review is cute and all but not too useful in this kind of capstone project where learners are generally inexperience.

Overall, not too much value was added from the course unless you want to experience the frustration of working with data, which you can get elsewhere. Not recommended

This course flip flops between SQL and Python and expects new SQL learners to be able to follow along. None of the courses expand on one another and instead it feels like you blindly walked into 4 random computer science classes on a college campus. They use so many different tools and programs as well without properly introducing you to them and then once you finally feel like you're understanding the system, the next course has a completely new one. Much of the material also has typos and grammatical issues or sometimes the questions are just randomly cut off. It's as if someone asked AI to come up with a computer programming course that can be completed in a month using a 4-year program as its reference.

This was a great course. It taught me more about SQL in one month than a semester at a top 20 university.

The concept of this course should be appreciated. It is very nice to conduct a project by yourself to understand better what you've learned. Indeed, the course planed is organized so that you can go ahead step by step to complete your project. However, I need to put negative points to technical problems. You see something interesting in the videos, but its codes are not provided and there is no explanation of how you can reproduce what you see in the videos...

This capstone project is run either on Python / Scala or R. It has almost nothing to do with SQL by itself. You can run it in provided Jupyter notebook sandbox OR on your local machine (technically, you can avoid using SQL in total :D, but can also create SQLite DB localy from the datasets... Provided videos are absolutely not useful, as they show text processing in Python. 3 datasets are provided for capstone, to in .csv file (easy to import anywhere) and once in .json (difficult). Do not enroll if you don't have sufficient skills in statistical programming languages On the other side, the provided datasets allow for some interesting inferential analysis.

The content and direction were above and beyond what I was expecting. This is a real class- I spent at least 160 hours on the specialization and have legitimately upgraded my SQL skills to intermediate. The structure of the classes cannot be overstated, as each module builds on each other and the cumulative projects are done in chunks each class (albeit with the last one an exception, there was less hand-holding there and more focus on where we can go from here (Jupyter, Python, Tableau, things like that). If you're out of work due to the pandemic, there is literally no better way to spend your time than learning the skills needed today, from this class. Can't say enough.

This guided project was a nice end to the SQL Basics specialization.

The final course was difficult to follow. The instructor was using a different application than what was taught in previous courses. It was very difficult to see his screen. He was only slightly teaching the Jupyter application he was using, but not with depth. He was more giving ideas of what we could do with our own projects. It would have been better to actually learn more. But I enjoyed the opportunity to practice more and complete a presentation.

Not enough guidance on how to do the assignments- lots of use of python when there is no background of it in previous courses. The videos skimmed through it as well.

Pros: - Real life-like scenario for you to study by yourself and come to your own conclusions and analysis.

Cons: - Some knowledge of Python is required even tho this is a SQL specialization targeted at beginners with little-to-no prior programming knowledge of any kind. You shouldn't have to go to something like stackoverflow to troubleshoot things that weren't even discussed in this or prior courses since it more or less defeats the purpose of enrolling in guided courses like this one. - 4 peer-graded assignments. In my opinion, this is the weakest link in all of Coursera. Especially when it comes to courses targeted at beginners. You really can't expect a beginner to rate some of these, more nuanced projects correctly/fairly simply by giving them a cheat sheet. That's without mentioning language barriers and peer availability (this course is the least enrolled of the specialization, btw), among other things.

Overall, it's kinda disappointing to see a specialization that started so well end (for those lucky enough to get graded) up in such a mediocre fashion. I won't be taking any more courses by UC Davis nor recommending them after I'm done with this one.

I selected the Lobbyists4America dataset. Not to mention, Don demonstrated how to prepare data for analysis using Python, which was not covered in this course and was not a prerequisite. Okay, I skipped these videos.

I imported a tar.gz file into Databricks, but it is unable to build a proper table because the only file formats supported are csv and json. Nobody described how to unpack the tar.gz file into the database environment. Don't tell me to unpack locally and upload it (the unpacked file is 1.6Tb!).

This course is WEIRD. The instructor said 'okay set? let's go' and then the Python scripting came out from nowhere. I was like what in the H*** was he doing in this SQL, repeat SQL, course???

This course is very good. The content teaches you much more than only SQL and open your eyes for a whole new world of opportunities to study. I didn't give 5 stars because the second module was very challenging. The explanations were not that clear and the videos do need some better edition. Nevertheless, I did learn SQL and I feel very confident with the language after taking this course.

The course is supposed to evaluate SQL skills but unwillingly the learners have to use a lot of their Python skills. That would be my only complain.

There is nothing about the Capstone Project that requires the use of SQL, and it might be a disadvantage if you try to use SQL instead of Python or R.