A Really Practical Guide to Using Panel Data

Here’s a really practical guide to using panel data: Longhi, Simonetta, and Alita Nandi. 2015. A Practical Guide to Using Panel Data. Seven Oaks: Sage. Based on course notes of a course taught for many years, this book deserves the word practical in the title. It contains the code and explanations to run your own panel analysis in no time. You’ll have to use Stata, and it only covers the British, German, and American panels.

I would probably recommend using this book in conjunction with one of the Green Books by Sage, e.g. Allison, Paul D. 2009. Fixed Effects Regression Models. Quantitative Applications in the Social Sciences (QASS) 160. SAGE Publications.

What’s great about Longhi & Nandi’s book is that it doesn’t pretend all data are textbook data: it starts with getting the downloaded data into shape and data cleaning. Even though the book is comprehensive, I would have liked it to go into more depth from time to time.

Introduction to Statistics Using PSPP

A few weeks back I argued that PSPP is not (yet) a real replacement for SPSS. I also claimed wrongly that there are no introductions to statistics that use PSPP. I had book-length introductions in mind, but alas no is not quite the right word. Today, I give you The PSPP Guide: An introduction to statistical analysis. This isn’t a proper review, nor an endorsement, simply because I haven’t actually read the book.

Nonetheless, here are some observations (looking over the table of contents). First off, the book does not seem to introduce much beyond PSPP’s capabilities. On the one hand, this is great for the readers, on the other hand, when teaching, there are many things I want my students to be aware of — doing statistics is one thing, reading and interpreting another. I note chapter 6 which sidesteps current shortcomings by using graphing capabilities in OpenOffice. The 2014 version of the book includes factor analysis, keeping up with PSPP. This said, personally, I cannot envisage teaching an introduction to statistics without mentioning logistic regressions.

Given the active development of PSPP I have no doubt that we will see more books like this in the future (and probably from more reputable publishers, too), but frankly, I can’t see myself using a book that doesn’t cover some of the methods I consider essential.