# Thread: Intro Statistics books too superficial?

1. ## Intro Statistics books too superficial?

I am sure many of you are familiar with the undergraduate "Intro" college textbooks on Statistics. Sometimes I feel these books are "cheating" students, I myself being a student. By "cheating" I mean that they seem to lay such a superficial foundation, that not many really ever seem to understand what they are doing or how to apply it successfully. This leaves the student with the ability to just work "textbook" problems and not real world problems. I would hope that such a gap is not too large, or could be minimized quickly!?

Do you feel these books are too superficial? One book I have is the one called Introductory Statistics by Neil Weiss. The book has a lot of content in it, but itself even mentions it is not meant to be overly complex but intuitive.

So my question basically is, what resources do I use to actually learn why and how the statistical tests, sampling techniques, regression methods etc. all work the way they do?

I look at the more complex books on Amazon, and they look too complicated. So the problem is...the textbooks are either too complicated, or too simple. So what do I do!?

2. ## Re: Intro Statistics books too superficial?

Originally Posted by StilloftheNight
I am sure many of you are familiar with the undergraduate "Intro" college textbooks on Statistics. Sometimes I feel these books are "cheating" students, I myself being a student. By "cheating" I mean that they seem to lay such a superficial foundation, that not many really ever seem to understand what they are doing or how to apply it successfully. This leaves the student with the ability to just work "textbook" problems and not real world problems. I would hope that such a gap is not too large, or could be minimized quickly!?

Do you feel these books are too superficial? One book I have is the one called Introductory Statistics by Neil Weiss. The book has a lot of content in it, but itself even mentions it is not meant to be overly complex but intuitive.

So my question basically is, what resources do I use to actually learn why and how the statistical tests, sampling techniques, regression methods etc. all work the way they do?

I look at the more complex books on Amazon, and they look too complicated. So the problem is...the textbooks are either too complicated, or too simple. So what do I do!?
The introductory books are supposed to be very high level. Nearly every chapter could probably taught as at least one course (some would even require a sequence). If you find things you're interested in, keep digging and borrow a book from the library if you're unsure about purchasing it first.

In any case, the best advice I can give is this: apply techniques to real data sets and figure out how to handle real issues. I find that going through the derivations of certain things helps me to internalize concepts, too. If you have access to at least Excel, but ideally a stats program (less clumsy), try messing around with and simulating things (central limit theorem, repeated sampling to generate CIs to show that X% of all X% CIs will capture the parameter value, mess around with bootstrapping estimates, etc.). Asking questions and then trying to simulate them or work with real data (i.e. what happens to the sum of squared errors when I drop an independent variable from a model? Can I force R-squared to 100% (1) with meaningless variables? Why can't this happen with Adjusted R-squared?). Use the computer in junction with your text (and a pencil and paper to work out some things by hand).

If you're still at school, consider taking more stats classes (applied ones might be more interesting at first, but it's always good to get theory to help with the real "why" behind things).

Hope this helps!

 Tweet

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts