Hacker News new | past | comments | ask | show | jobs | submit login
Stats without Tears (2016) (brownmath.com)
124 points by ghosthamlet on Dec 20, 2017 | hide | past | favorite | 19 comments



The intro is promise:

DON’T PANIC!

Douglas Adams’ The Hitchhiker’s Guide to the Galaxy bore a “large, friendly label” with those words, and that’s also my message to you.

I don’t see any reason for students to be afraid of statistics. It’s no more difficult than any other technical course, and it’s much more practical than other math courses. The mathematical details are here for those who want them, but I lean heavily on technology to relieve students of the “grunt work”.


"Don't Panic" is generally true when learning most bodies of knowledge. Well any situation really, but especially math and science.

I did my PhD in Physics, and I've found it's very easy to become overwhelmed by a new field of math, physics, or even a large new code base.

But eventually after you bang your head enough on the math or the code base you realise it's not nearly as hard as you first thought.

Everything comes down to a large number of small steps. Each baby step makes sense on its own, like each single line of code.

But the feeling of being overwhelmed can cause panic and anxiety, leading people to lose their confidence that they can understand the new baby steps present in whatever they're looking at.

When I was a TA back in my PhD, I would sometimes tell my students that studying physics is like jumping off the Empire State Building. If you try it in one big jump it's likely to end in disaster, but if you break the 1000 foot height into 2000 steps you can easily get down to ground level.

Unfortunately some sub-par professors and teachers, or poor text books, can exacerbate this feeling of inadequacy by resorting to complicated definitions or explanations when simpler ones would suffice. Sometimes I think it's people wanting to sound smart and toot their own horns, other times it's just poor communication.


I think statistics is hard.

In my country, it is taught to freshmen and it uses the Central Limit Theorem all the time. That theorem is deep and hard to prove so student are forced to take it at face value. Other math courses do not work like that.

This is done precisely because statistics is so useful.


How does it compare/contrast with:

- Wasserman's All of Statistics? (which I haven't read/worked on)

- Downey's Think Stats and Think Bayes? (which I've read in bits and pieces and have found amazing)

- Any other well known Stats or Probability text?


I'm a non-mathematician stats newbie but here is my take:

Strength of this book seems to be that it is very approachable in a "plain english" kind of a way. This is the first time I have seen it and I'm looking forward to reading it.

Strength of _All of Statistics_ is that it is fast-paced and approachable even though it is essentially a math textbook. You only need a bit of algebra to make sense of it. That book exudes the feeling that the author wants to get through all this theory as quickly as possible to move onto other exciting things. (_All of Nonparametric Statistics_ by the same author is actually less approachable and shouldn't be taken as a simplified "the good bits" subset.)

The stats book that I have had the most value out of is _Statistical Modeling: A Fresh Approach_. It's fundamentally about making sense of different kinds of data using a computer. http://www.mosaic-web.org/go/StatisticalModeling/


Regarding All of Statistics, the author states that the 'reader is assumed to know calculus and a little linear algebra'. It's a good book, though think you'd have a tough time with only 'a bit of algebra'.


Although old, I really like Principles of Statistics by Bulmer. Duke’s MOOC is pretty good btw.


Looks like a good intro, has reference links to keywords/definitions, problems with solutions, text-searchable.

References TI-83 or TI-84 feel dated.


You might think that, but high schools in the United States are largely still using these devices, even purchasing new ones.


At the university I work at, these calculators are a common item on students' desks because graphing calculators are verboten (until calculus II).


> References TI-83 or TI-84 feel dated

Yeah ... you would not believe how entrenched these are. Mobile phones are not allowed during testing, so they just stick with old faithful.

Also look up the current prices for TI-84s ... it is ridiculous considering the worlds current status with hardware.


> References TI-83 or TI-84 feel dated.

Now I feel dated. I think there are a few of these still circulating through the younger branches of my family.


It looks nice. Did anyone read it and have an opinion?


I feel like the impact of this e-book would be magnified if the author simply took the effort to improve its design and layout.


One of the problems with Free books is that not just the writing falls to the author. They typically also do answers to exercises, marketing, and here layout. It is unrealistic to expect that one person will be able to do those all. (All this goes double if the text is significantly interactive.)

I know that there are people who do well in all categories, but expecting that all Free texts will have such authors means expecting that there will be few such texts.

There needs to be a way for people who have skills in such areas, and who believe in a project, to get together.


Isn’t it trivialising the effort it takes to figure out the best layout, by adding the adjective “simply”? It’s a little offensive considering this looks like an amazing resource out for free.


You're right. My bad. It takes a significant amount of effort to design a nice/beautiful layout for an online book. Definitely not something to be underestimated.


Could you give some examples? It looks like a math textbook which is already pretty inviting. Could it be better? Perhaps black text on a white background?


Maybe that's not their area of focus ;) but I do agree! Maybe someone can help.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: