Top Five Reads of 2017

I once heard someone say that all leaders are readers, though not all readers are leaders. I subscribe to the philosophy embedded in this message. Reading is one of my favorite pastimes and is an integral part of my ongoing professional growth and development. I love book recommendations and thought I would share my five favorite reads of 2017 and a brief report of the value I extracted from each.

1. Thinking, Fast and Slow

Author: Daniel Kahneman

Paperback: https://www.amazon.com/Thinking-Fast-Slow-Daniel-Kahneman/dp/0374533555/ref=sr_1_1?s=books&ie=UTF8&qid=1514596382&sr=1-1&keywords=Thinking+Fast%2C+and+Slow

This book offers incredible insight into the human mind. In it, Kahneman examines two systems of thinking: System 1 operates quickly and with little effort, whereas System 2 allocates the requisite attention for computationally demanding tasks. Kahneman offers many examples of how our intuition is often incorrect and makes the case that slowing our thinking lends itself to better decision making. Consider the following example: If a bat and ball cost $1.10 and the bat costs $1.00 more than the ball, how much does the ball cost? Most are quick to conclude that the ball costs $.10, which is of course incorrect. However, when recruiting System 2 to think through this, it becomes clear that the correct answer is $.05.

This book has taught me to be more conscious of System 1 thinking and to more carefully consider what complexities may exist for the problem at hand that may not be immediately evident. To this end, I think it is important that one not assume the most reticent individuals are least knowledgeable or have little to contribute; it may be that they are better equipped to recruit System 2 before speaking – an approach that should be encouraged.

In addition, the book contains some great examples that illustrate how small samples tend to lend themselves to statistical anomalies as well as the centrality of randomness in probability sampling. There is something of value in this book for everyone, and I highly recommend it.

2. Work Rules!

Author: Laszlo Bock

Hardcover: https://www.amazon.com/Work-Rules-Insights-Inside-Transform/dp/1455554790/ref=sr_1_1?s=books&ie=UTF8&qid=1514596337&sr=1-1&keywords=work+rules

I received a signed copy of this book from Laszlo Bock at the hiQ Elevate Conference in late 2016. This book has been all the rage in HR, particularly in people analytics circles, so I felt compelled to read it. One of the key outcomes of this book for me was a piqued interest in TA analytics. It brought to bear the myriad opportunities in the predictive hiring arena by detailing how Google leveraged the People Analytics team to help scale their inefficient and highly selective hiring practices as the organization experienced exponential growth. Consistent with the message in Kahneman’s book, the idea of leveraging science to recruit top talent is rooted in the assumption that instincts and biases prevent humans from doing this well. “Most interviews are a waste of time because 99.4 percent of the time is spent trying to confirm whatever impression the interviewer formed in the first ten seconds” (p. 89).

Given the deluge of signals made possible by contemporary technologies and data sources, the opportunity to derive important insights into organizations’ workforces is unprecedented, and this book has undoubtedly been a catalyst for change. If you sit in HR, this is a must-read.

3. Naked Statistics: Stripping the Dread from the Data

Author: Charles Wheelan

Paperback: https://www.amazon.com/Naked-Statistics-Stripping-Dread-Data/dp/039334777X/ref=sr_1_1?s=books&ie=UTF8&qid=1514596259&sr=1-1&keywords=naked+statistics

I read this book during a trip to Boston this summer. It is a quick and easy read. I developed a Business Analytics course for an MBA program that I have taught the last two years, and I purchased this book hoping to pick up some new, engaging ways of teaching statistics concepts. I quickly found myself gaining a deeper understanding of topics I thought I already knew well. Wheelan’s explanation of the Central Limit Theorem and its role in enabling generalizations from samples to populations is brilliant and the most easily understood I have heard; it has inspired how I now teach this very important and foundational topic. I also enjoyed Wheelan’s treatment of the “miracle elixir” (i.e., regression analysis). Untangling factors that are inextricably intertwined in order to parse out the effects of each on an outcome can be messy, but readers will gain a solid understanding of how regression facilitates this objective after reading this book.

Wheelan also provides some effective examples to give readers a greater appreciation for the complexities in deceptively simple research questions. One example is the positive but misleading association between police officers and crime. While there is a solid theoretical reason to believe that increased police presence will reduce crime, it is also possible that crime could cause police officers in the sense that cities experiencing greater crime may hire more police officers. Another example ponders whether or not attending an elite school will change a person’s life. In other words, do graduates of elite institutions do well in life because they were incredibly talented when matriculating or because the universities add value by making already talented individuals even more productive? If you are interested, economists Stacy Dale and Alan Krueger developed a clever method to test this using a monetary measure of ‘success’.

4. An Introduction to Statistical Learning with Applications in R (ISLR)

Authors: Gareth James, Daniela Witten, Trevor Hastie, and Robert Tibshirani

PDF (free): https://www-bcf.usc.edu/~gareth/ISL/ISLR%20Seventh%20Printing.pdf

Hardcover: https://www.amazon.com/Introduction-Statistical-Learning-Applications-Statistics/dp/1461471370/ref=sr_1_1?s=books&ie=UTF8&qid=1514596891&sr=1-1&keywords=introduction+to+statistical+learning

I first read this book in 2016 and reread it twice in 2017. Having the PDF version on my phone has made it very accessible. Each time I read this, I gain a deeper understanding of the technical underpinnings of some of the most popular and widely employed algorithms in predictive modeling. Most of this text covers supervised learning algorithms, but there is a chapter devoted to unsupervised methods as well. Compared to its predecessor, The Elements of Statistical Learning (ESL), ISLR is a non-technical treatment of the topics, though I wouldn’t classify it as an easy read as I did Wheelan’s text. ISLR covers topics such as Linear, Logistic, Ridge, and Lasso Regression, LDA, QDA, KNN, GAM, Decision Trees, Random Forests, SVMs, PCA, K-Means and Hierarchical Clustering, cross-validation, and variable selection methods.

A quote attributed to Einstein suggests, “If you can't explain it to a six year old, you don’t understand it yourself.” While I am not certain I could explain many of the concepts in this book to a six-year-old, at least not in a way that does them justice, I find it intrinsically rewarding to develop a deep enough understanding of these important topics that I can explain them simply and effectively to those who have never been exposed to them. Regardless of your level of knowledge with respect to these techniques and their implementations in R, this is a great resource.

5. Storytelling with Data: A Data Visualization Guide for Business Professionals

Author: Cole Nussbaumer Knaflic

Paperback: https://www.amazon.com/Storytelling-Data-Visualization-Business-Professionals/dp/1119002257/ref=sr_1_1?ie=UTF8&qid=1514596162&sr=8-1&keywords=storytelling+with+data

I have always had a preference for the technical, quantitative elements of research and analysis over the art of visualizing results. That said, I recognize the importance of effectively communicating the story behind the data to technical and non-technical audiences alike. In an effort to enhance data visualization and storytelling skills, I picked up this book based on the glowing reviews it has received. In full disclosure, the fact that the author was a former member of the People Analytics team at Google may have made the book more attractive to me.

This book is very well constructed and offers many easy-to-implement principles and guidelines that are universally applicable. It surveys well known visuals, such as line graphs and bar charts, as well as lesser known types, such as slope graphs and waterfall charts. It also examines common missteps to avoid, such as pie charts and 3D visuals. I ended up creating a 96-slide deck from the notes and screen captures I took, which I consult regularly. This is now a required text for my Business Analytics course, and a larger share of time is devoted to data visualization and storytelling. I highly recommend this to anyone in the business of communicating with data.

What books do you recommend for 2018?