Data analysis is at least as much art as it is science. This book is focused on the details of data analysis that sometimes fall through the cracks in traditional statistics classes and textbooks. It is based in part on the authors blog posts, lecture materials, and tutorials. The author is one of the co-developers of the Johns Hopkins Specialization in Data Science the largest data science program in the world that has enrolled more than 1.76 million people. The book is useful as a companion to introductory courses in data science or data analysis. It is also a useful reference tool for people tasked with reading and critiquing data analyses. It is based on the authors popular open-source guides available through his Github account (
The paper is also available through Leanpub ( if the book is purchased on that platform you are entitled to lifetime free updates.
Quick read, filled with best practises (do's and don't when analysing and presenting data). Very succinct and will likely refer to it time and time again.
Pleasant, readable, sensible. This bit's good, tells you exactly how most social science is limited (it stops at inferential, and sometimes manages to mess even that up):
Очень сжатое и лаконичное описание видов анализа данных с базовыми советами и указанием типовых ошибок. Подойдет новичкам для начального знакомства с темой. Бесплатно (или с пожертвованием) можно получить здесь:
I would certainly recommend this book to anyone who starts with data analytics for a readable, short, summary of the main ideas. What I found lacking was the presentation of these ideas.
The heading levels seemed inconsistent and combine subjects of different level on the same heading level (e.g. chapter 3 mixes the basic components of the processed dataset with minor errors).
Also, the book's insights and ideas could be summed up in a nice series of checklists. And it is great that the book actually provides a series of checklists for the data scientist.
Perhaps useful if this is your first course in data analysis, but not useful for anything but the most beginner starting out. Lack of good examples, strange unexplained discipline specific jargon, and overly broad advice that rarely goes beyond obvious. Was hoping this could be a useful resource to build team norms and goals around, but it is not
Taking off from Strunk and White, Leek develops here a handbook for the user of data.
It is a good introduction for someone who is just really diving into thinking about using and presenting data as a thing in itself. Good data and its presentation is a rhetorical tool that is often under-thought. I will surely be using this as a reference book in the future.
A very lucid guide which act as a prerequisite for a deep driven analysis of a problem in hand and make yourself less vulnerable of some problems which you may fall into while analysis and can yield very of the line results.
Good summary of best practices and recommendations with links to more information. I will certainly keep this handy and revisit when working on data analysis.
É bem curtinho mas me mostrou várias coisas novas. Li bem rápido, me prendeu bastante. Foi uma boa ver uma visão bem parcial de alguém de dentro da área.
This was a decent overview to data science, although in some places it assumes knowledge that the reader might not actually have. I think this could become an important reference for someone going further in data science, but I'm not so sure that it shouldn't be revised to be more self-explanatory for beginners.
A concise introdution and instructions about all stages of data analysis. Each topic can be expanded into a much more deep communication but the suggestions mentioned are very practical. I think it's a good starting point if you're a new-comer to data analysis. And it would be helpful to frequently look it up when you're doing the process to make sure you're on the right track.
Read this from a Data Viz/Tableau PoV, extremely interesting to see how statisticians/academics approach Visual Data Analysis. Small, concise, wish it went into a bit more detail with examples, but I guess there are plenty of other books that do that.
Very helpful for what it is, and what it is intended to be -- an extremely concise and pragmatic quick reference, with plenty of helpful pointers to other sources of information for more depth. A very useful companion to the JHU data science courses as well.
Helpful guide to know how to create a data analisys. It just shows the main points to take care of, even a chapter on how to do presentations of data analitics.
Seguindo o estilo do livro de Roger Peng que li antes, o autor aqui faz praticamente uma receita de bolo, com passos detalhados de cada etapa da análise de dados e um conjunto de dicas acerca dos erros mais comuns e como evitá-los. Certamente retornarei a este livro em algum momento da minha pesquisa.