Text Analysis with R for Students of Literature (Quantitative Methods in the Humanities and Social Sciences) 2014th Edition

Jockers (2014) is a vital source for processing texts of all kinds. I am an educational researcher interested in both qualitative and multi-method analyses, and I have been peevish about using statistical modeling on unstructured texts. This book was a pleasure to read, and I highly recommend it to both students of literature, but also to anyone in the social sciences seeking to find patterns, complexities, and contexts deeply embedded in written texts. It is easily argued that this text fills an important cross-disciplinary methodological gap. Bravo! Well Done!Mario A. Martinez, Ph.D. (Curriculum and Instruction)

I put it down by page 30, and never picked it up again. Too many detours, not enough theory. It's hard to explain what it is about this book, other than I'm a programmer, and have read good and bad books on coding, and this is about the worst. I would recommend the author give some solid theory on text analytics before beginning to show the coding, so we know why we're doing what we're doing. There is no explanation on how you should think about text before analyzing it. The book trickles in lines of formula in R, without you ever getting a picture in your mind as to what you're aiming for. It would be better to buy a book on the theory of text analytics, forget the programming side, and just understand what you'd need to look for when analyzing text. Then, once you have some ideas, the rest of this would make sense. As it is now, you don't know where to file the info you're getting. What's this step for, you ask? You don't know. You have no foundation in text analysis, and are just given steps to write math with R, without knowing what the purpose of the formula is for. Useless!

Engaging writing, with code samples and practices. As for programming, I thought the code quality was somewhat low or sloppy, but Jockers is not a software developer by trade. While reading, I did have a few ideas to solve some text-matching issues across systems, and generally, I found the lack of discipline in the author's approach conducive to flexible thinking about using techniques with R.

You must take the title literally. The book provides an introduction to using R for text analysis, for students of literature that already know text analytic methods. In that regard it is a book for a very limited audience. If you know R and wish to learn text analysis, the book will do nothing for you.

Needs a github repo

Good book

A good introduction to R for anyone interested in text analysis.

I have long been ready to go beyond "out of the box" text analysis tools. This has especially been the case since reading Jockers's own book of criticism, Macroanalysis. I am co-author of a book of historical criticism on modernist literature, Modernism: Keywords (Wiley-Blackwell, 2014), and have been experimenting with using topic modelling as a method for automatically detecting shifting uses of keywords (and key themes) for a proposed second volume. I had Mallet working from the command line, but — after reading Macroanalysis — became frustrated that I couldn't easily chunk texts to particular sizes; perform part-of-speech tagging and extract only certain parts of speech for topic modelling; generate visualizations like word clouds; and output forms of data. For all this, I realized, I needed to learn some programming.Though I have quite a bit of experience with HTML, CSS, and XML, but I had no programming experience prior to reading this book aside from some classes in high school. I knew from my experience of learning HTML that what I needed were (1) some clear explanations of basic concepts and (2) some sample scripts to tinker with and adapt to my own uses. Point (2) is crucial: it's very hard to make scrips from scratch, but I find that once I see a script that's doing something similar to what I want to do, it's relatively easy to customize it to my needs. But, as I know from a few failed attempts to pick up programming from MOOCs and For Dummies books, I also knew that I needed instruction focused specifically on literary analysis. I wasn't interested in learning to program for its own sake; I just wanted to be able to carry out my particular literary research project, which required me to know how to do some programming.Text Analysis in R was fantastic on all counts: the focus is specifically literary, the explanations are very lucid; and the example scripts are close enough to what I needed to do that I was able to customize them. (The chapter on Topic Modelling was of course particularly useful to me.)I read the book sequentially in about a month, doing all the practice exercises along the way. (I read it on the bus on my commute... about 6 hours/week.) By the time I finished, I was ready to fly on my own. In the following weeks, I was able to customize the scripts to my own needs. The modifications were sometimes very significant; I was somewhat amazed at my own progress, I must say. I was able to make the system I initially thought I would need to hire a programmer to develop: a web-based interactive topic modelling browser, in which you enter your own keywords, and the system return the topics in which that word is most likely to appear, showing a word cloud, distribution chart, and list of top texts for each topic). I recently presented my work at a major conference in my field. It went over very well -- and since I did the programming myself, I was able to field technical as well as literary questions.I feel immensely empowered by this book. I have also definitely caught the programming bug. I can't leave R alone now.One final note: some of my programmer friends have turned up their noses at the idea that I would learn R rather than Python, which they all seem to prefer. I chose this book not because I thought R was the right language for me, but just because it seemed like it was focused on the sorts of TASKS I was interested in. Having gone through the book, I'm now fully convinced that R is an excellent language for literary analysis. The ease of outputting tables and charts -- and the ease of performing statistical calculations -- are particularly welcome features of R.

Excellent introduction to R for non programmers, non experts. There may be two different (perhaps overlapping) reasons for this.1) R as a language is easy to get to grips with if you have some idea about the basic tasks you want to do. For me, I was interested in what was and wasn't possible with text analysis. I've tried an introductory book on python, and it was all about how to find the square root of numbers. Whereas this book, you immediately get into the textual analysis that the reader wants to do. I also found R an easy language to debug, when you inevitably make typo mistakes copying the text, and I felt like I learnt a lot from correcting my work and the error messages R gave me.2) The author deserves credit for writing a book that a reasonably intelligent non specialist can understand. Perhaps given his interest in literature, some of this has rubbed off on how to write an engaging technical book. That is, for a technical book it is very readable.I don't feel like an expert, but I've got a really good idea about what is and isn't possible with text analysis. And it's certainly whetted my appetite to try and apply these techniques.

This is a good starter either for students of literature with little or no previous familiarity with the R library of program libraries, or for others with little previous knowledge of literary research but inception or more in R.

V good book. Really liked it.