Book release: Statistics for Linguists

10/31/2019

My new book, Statistics for Linguists: An Introduction Using R is scheduled to come out on Nov 15. You can already pre-order it on Amazon (US, UK, DE)!

The market for statistics textbooks is crowded, so what's special about this one?

The book is written in an accessible and friendly style, similar to my mixed model tutorials.
There are lots of hands-on examples with real data from actual studies. These examples do not require special linguistic knowledge, which makes the book accessible to many different types of researchers, especially psychologists and cognitive scientists.
Everything is based on the linear model framework.
The book encourages the adoption of open reproducible research methods.
The R code is written in tidyverse style.
Each R function is explained in detail.

The book's explanations have evolved in response to extensive feedback from dozens of statistics workshops that I have taught. So I hope that they are clear and engaging!

Who is the book for?
I hope the book is useful for advanced undergraduates, but also MA and PhD students. And even experienced researchers and data analysts may find some helpful ideas in the way I approach some of the concepts looked at. The book is designed to work as a first introduction to statistics and R and should prove valuable for statistics courses.

New blog!
Incidentally, this is my first-ever blog post. Watch this space or follow me on Twitter (@BodoWinter) for updates on new posts. I plan to write about academia (such as my experience as a General Editor at Language & Cognition), stats (including hands-on tutorials), and linguistics/cogsci topics such as gesture, iconicity, and sensory language.

Conference!
If this is not on your radar yet, join us in the buzzing city of Birmingham for the UK Cognitive Linguistics conference to participate in this exciting interdisciplinary feast of language and cognition research!

9 Comments

Keramat Nouri

11/25/2019 08:23:53 pm

Hi Bodo,
Congratulations for your new book, Statistics for Linguists: An Introduction Using R.
I am professor of Biostatistics in Canada and would like very much to use your book to teach linear models to the master’s students of public health (e.g. Epidemiology, Nutrition, etc.)
I haven’t seen your book yet but read a couple of your nice tutorials.
In fact, I am not still sure if your textbook going to be a suitable choice for my target audience (i.e. public health students). That is, I would like to know if examples of your book are going to be easy to understand by my students (and myself) or one really needs to be familiar with Linguists. Would you please kindly advise?
Thanks,
Keramat

Bodo Winter

12/6/2019 07:00:50 am

Thanks for your kind words! Check it out! I think the examples are accessible to a very wide audience, including public health. I deliberately chose examples that are easy to understand.

Keramat

12/10/2019 10:39:35 am

Many Thanks.

johnly lee Easley link

1/21/2020 05:12:59 am

i am very usefull book and very healthy (nutrition etc..)

nobody

3/4/2020 08:47:09 am

an alarm for a possible typo?
page 75, it says "...the model shown in Figure 4.3a." but I couldn't find Figure 4.3a. (there is Figure 4.3.)

Other than this minor error, I have been really enjoying your book!

George Wolford

8/4/2020 08:04:38 am

I've been using your pdfs (regression and lmms) in my stat course for a few years now with great pleasure. I just purchased your new stat book in paperback with the hope of using it in my course. I like the book but there are missing figures and other oddities. For instance the on Page 83, the book refers to Figure 4.9 but no such figure exists. Other places refer to color such as 'salmon' but there are no colors (probably not surprisingly). Is my book flawed? Will there be a second printing?

George Wolford

8/5/2020 10:01:14 am

So in section 8.2 in your book (categorical * continuous interactions). you mean center SER but you don't mean center POS. I've always done both to remove correlation of main effects from the interaction term. I reran the analyses. I first made POS a factor and specified contr.treatment. I get identical estimates and values to the ones you report. I then changed POS to contr.sump and the p values and estimate for SER change quite a bit (much more sig). The values for POS and the interaction don't change. I think you need to mean center both to get the correct main effects.

Jahir

8/10/2020 04:40:44 pm

Hey Bodo,
I was trying to following your mixed effects tutorial, but can't seem to find the data at the link provided: www.bodowinter.com/tutorial/politeness_data.csv

Is it possible to have the data?

Thanks.

Natalie

10/24/2020 02:59:29 am

Dear Mr. Winter, I'm a student in linguistics learning statistics and R. Your book is very helpfull, thanks a lot for your great work - I get used to R! The 1st 6 chapters were a content of a statistics course in the uni, now I'm working thru the material alone. I have a question to the chapter 8: how is the graphics 8.4a built? I made 3 predictions for ARC_z =-1.5, 0, 1.5 and can plot fitted lines for every case, but how put them together?
A comment to the exercise 2 (8.8.2) you suggest hand -coding of the type treatment coding (0 and 1), but in the chapter you used sum coding (-1,1). Coding manualy as in the execise will lead to results different from those in the chapter. In order to get same results one should code manualy with -1 and 1 instead of 0 and 1. Am I right or have I overseen smth?
Thank you in advance for your answer!

Book release: Statistics for Linguists

Leave a Reply.

Archives