Don't know about social sciences, but the statistics bible for biomedical research is "Biometry" by Sokal & Rohlf. Maybe have a look at it and I'm sure it will cover social sciences relevant topics as well!

Sorry, I don't think there is 1 "bible" of social science statistics. There are various books that cover aspects such as design, analysis, interpretation, but not 1 that does it all in a great way.

For me Andy Field's Discovering Statistics Using SPSS is definitely the first book I check if I need to clear some issues. I highly recommend it. It's written for beginners, advanced users, students (especially in sociology and psychology) and anyone who needs to understand certain methods. It does not comprise everything, of course, but I find it very useful both for stats/econometrics and for SPSS

Following G.E.P. Box that "All models are wrong, some are useful" we may conclude: There are not "bible" statistical books, some are useful ;-)

@Patrice, can you mention specific books from (the) King?

@Edgar, which one would you say is the best one for analysis then?

@Raluca, this is the type of book I am looking for, but it is a pity that I am not SPSS fan. I started learning STATA already...

@Plamen, I guess some are more useful than others. I hope people with more experience will help me to pick the right one.

@Plamen, you don't have to be a SPSS fan, each chapter is divided in two main parts, one of theory, where he explains the concepts, and one on how to do that with SPSS. You can ignore that second part, although I would read it just for getting an idea on how to interpret the obtained results (especially if you are not familiar with that particular analysis tool), since he focuses on the interpretation more in this part. Besides, even though I'm not familiar with STATA, I'm guessing the results and the way to interpret them are pretty much the same. That is, if he gets the F in ANOVA or the t-test result and the significance level for example, he interprets such a result based on the example he gives. No matter what soft you use to generate the result, you still need to interpret it :) Good luck!

books for learn

Topic, Author

Probability, Walpole or Myers

Inference, Kiefer & Lorden

Sampling, Cochran

Econometric, Maddala

Multivariate Analysis, Hair

Experimental Design, Cochran & Cox

Non parametric Methods, Hollander

Please, stop searching for a 'Bible' anywhere!

This 'concept' simply does not exist.

First you have to determine the exact field of your research.

Then, try to find what do you really need for it.

Multivariate Analysis Techniques in Social Science Research: From Problem to Analysis. http://books.google.be/books/about/Multivariate_Analysis_Techniques_in_Soci.html?id=hS-f5kc_03UC&redir_esc=y

Dear all, let’s start with the observation that also different versions of the bible exist, many versions :-). So to give an answer to the question: I really liked the book by Jacques Tacq. He is a mathematician and sociologist. In his books, every chapter explains one piece of multivariate analysis (EFA, Canonical cor., regression, etc.), but he each time uses the same building blocks, the same way of calculation and same type of explanation. He also goes into depth regarding the mathematics, but it is still explained in a very accessible way. When you go through it (I did it as a student), in the end you have a thorough understanding of each element of Multivariate Analysis and (!) you know how they are different, and similar(!)...

Thank you very much, Carlos and Jurgen!

Does anyone have an opinion on the book "Data Analysis Using Stata" from Ulrich Kohler and Frauke Kreuter?

Plamen.

general books that point to the use of software, do not address the statistics, teaching addresses the procedure in the software there are many elements to consider applying a statistical procedure that, although many do not consider, is proven, the variacionaes in test power. for example, the most overlooked, homoscedasticity, normality, multicollinearity, autocorrelation, among others.

So one thing to learn statistics, and the other is to understand it.

The people I work with, are clear that my contribution as a statistic is not only to determine the method of data analysis, but also convinced that the best procedure. Everyone understands the statistics, but do not handle the breadth of the profession.

If you haven't done so already, I recommend taking a course in statistical computing for the social sciences. The teaching materials in that course may point to the source(s) you seek.

Failing that, below are a place to start, as many instruments in the social sciences are based on subjective assessments (i.e, scales, rulers, and other qualitative measures).

Categorical Analysis by Alan Agresti

Thanks Mary. I actually did a course in statistical analysis using STATA and I was highly recommended to use the book which I add in the attachment bellow. I was going to buy it but it is rather expensive.

http://www.stata.com/bookstore/data-analysis-using-stata/

You may be able to find that book via online sellers, at lower prices for used books in good or very good condition. Good luck!

There are many many books to choose from. My current favourite for comprehensive coverage are:

Field, A (2005) Discovering Statistics Using SPSS (Introducing Statistical Methods series) Sage

(He also has a version covering the same material but using the R package)

Something I have had passed on to me recently was

Bram, U (no date) Thinking Statistically.

This contains no statistical methods per se but is a brilliant introduction to THINKING about statistical concepts.

you can search by internet

Sampling - Leslie Kish, social Statistics - Blalock, Try UN published books or manuals for Statistics.

The Bristol Centre for Multilevel Modelling provides a free online course that is focussed on quantiative modelling for social scientists. It can be followed using R, Stata or MLwiN. It starts with a gentle introduction to why quantitative work is needed and then covers continuous outcome regression, followed by discrete outcome regression. It does this for standard single-level modelling and then for multilevel models. It covers conceptual issues and how these can be implemented in practice through software.

The link is

http://www.bristol.ac.uk/cmm/learning/online-course/index.html

Multivariate Data Analysis (7th Edition) (2010) Joseph F. Hair Jr, William C. Black, Barry J. Babin, Rolph E. Anderson

Some very useful sugestions for social scientists are:

HEALEY - Statistcs: a tool for social research

HUCK - Reading Statistcs and Reasearch

FREUND & WILSON - Statistical Methods

TABACHNICK & FIDDEL - Using Multivariate Statistcs

What is the "Bible" of statistics?

There are various types of books on statistics e.g. descriptive, inferential, univariate, multivariate, variance-based SEM, covariance-based SEM etc.  Perhaps you can explore the following (access the latest edition if you can):

Hair, J. F., Black, W. C., & Babin, B. J. (2010). Multivariate Data Analysis – A Global Perspective, 7th Edition. Pearson Education.

Field, A. (2005). Discovering statistics using SPSS. 2ndEd. London: Sage Publications.

Hair, J. F., Hult, G. T. M., Ringle, C., & Sarstedt, M. (2013). A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM). Sage Publications, Inc.