1
$\begingroup$

This question might seem fairly simple, but I would like hear your experiences with statistical handbooks for behavioral scientists. Which handbooks have you found to be the most straightforward, clear and systematic (i.e., which handbooks have been praised by your colleagues or students) for the bachelor level of studies. So, I am not looking for books on specific statistical themes, but rather handbooks which offer a general overview of basic statistical concepts and themes important for a behavioral scientist (e.g., sampling, distributions, descriptive statistics, ANOVA/ANCOVA, correlation analysis etc.).

The books do not have to be free!

$\endgroup$
  • $\begingroup$ I thought nobody uses Handbooks anymore, Google is your handbook. That's not the case with a textbook, of course $\endgroup$ – Aksakal Feb 6 '18 at 21:22
  • $\begingroup$ I think there are various senses of "handbook" that might need to be distinguished. One is a supposedly definitive collection of expository or review articles. It's common, although not a definition, that such books are edited; and similarly that "Handbook" is a title word. Another is as an alternative word for textbook written by one or a few authors. $\endgroup$ – Nick Cox Feb 6 '18 at 21:55
1
$\begingroup$

I find Wilcox Modern Statistics for the Social and Behavioral Sciences an excellent principles-based approach to those concepts and topics.

For a different take, there is McElreath's Statistical Rethinking. This book can be life changing if read at the right time.

Gelman (ed) A Quantitative Tour of the Social Sciences is a very applied, practical approach that I would highly recommend too, but it's not a handbook as such.

$\endgroup$
1
$\begingroup$

Kass, Brown and Eden have an excellent book on the Analysis of Neural Data that manages to span lucidly and comprehensively a range from exploratory data analysis to more formal modelling.

See http://www.springer.com/gb/book/9781461496014

This is so good that people who don't work in this territory (I don't) can enjoy it and recommend it.

My guess is, however, that it will appeal more to behavioural scientists who think that say physiologists are their closest neighbours than to those who think the same of say anthropologists.

$\endgroup$
0
$\begingroup$

You may want to check out the statistics textbooks by Andy Field. These are especially helpful for behavioral scientists as they cover common statistical packages (e.g. R, SPSS etc.). The main con (or pro, depending on what you're looking for) is that they very non-technical and are focused on conducting the tests in software packages. They're not the sort of books that would be taught in a mathematics course.

$\endgroup$
  • 3
    $\begingroup$ I find some of Field's examples to be so risque in non-statistical content as to be disturbing and insulting. I would never recommend his book to my students. $\endgroup$ – Joel W. Feb 6 '18 at 20:27
  • 4
    $\begingroup$ I agree strongly with @JoelW on the matter of taste. I'll add that Field's books are often mentioned here (e.g. comments by whuber and myself at stats.stackexchange.com/questions/157217/…) and indeed elsewhere (e.g. comments by efrique at reddit.com/r/statistics/comments/704clc/…) as highly unreliable and packed with errors and bizarre confusions. $\endgroup$ – Nick Cox Feb 6 '18 at 20:36
0
$\begingroup$

I'd recommend either "Statistics for the Behavioral Sciences," 9th Edition by Frederick J Gravetter and Larry B. Wallnau and "Handbook of Statistical Modeling for the Social and Behavioral Sciences," by Arminger, G., Clogg, Clifford C., and Sobel, M.E., from Springer. First is more general and the second a little more detailed.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.