I couldn't agree more.
I am aiming to have the next installment in my How-To: Python for Data Science blog series posted in July. In the meantime, I highly recommend you check out this blog post by Vincent Granville which was featured in last Wednesday's Data Science Central newsletter. (Yes, this is a cop out tactic. It's summer, give me a break.)
The article is entitled The First Things you Should Learn as a Data Scientist - Not what you Think, and I couldn't agree more with the sentiment and key points of Vincent's post. (See what I did there with my blog post title? Clever, right?)
After observing and experiencing various approaches to teaching data science and analytics, I've come to the following conclusion: You can't teach someone how to use a specific tool set without showing them why and when they need to use those tools. For example, you aren't going to use a hammer with a screw, but you need to explain why that is to someone unfamiliar with the difference between a screw and a nail.
So, check out the post and see what I'm yapping about for yourself: