# Foreword

These are exciting times in statistics and data science education. (I am
predicting this statement will continue to be true regardless of whether you are
reading this foreword in 2020 or 2050.) But (isn’t there always a but?), as a
statistics educator, it can also feel a bit overwhelming to stay on top of all
the new statistical, technological, and pedagogical innovations. I find myself
constantly asking, “Am I teaching my students the correct content, with the
relevant software, and in the most effective way?” Before I make all of us feel
lost at sea, let me point out how great a life raft I have found in
*ModernDive*. In a sea of intro stats textbooks, *ModernDive* floats to the top
of my list, and let me tell you why. (Note my use of *ModernDive* here refers to
the book in its shortened title version. This also matches up nicely with the
neat hex sticker Drs.
Ismay and Kim created for the cover of *ModernDive*, too.)

My favorite aspect of *ModernDive*, if I must pick a favorite, is that students
gain experience with the whole data analysis pipeline (see Figure
0.2). In particular, *ModernDive* is one of the few intro
stats textbooks that teaches students how to wrangle data. And, while data
cleaning may not be as groovy as model building, it’s often a prerequisite step!
The world is full of messy data and *ModernDive* equips students to transform
their data via the `dplyr`

package.

Speaking of `dplyr`

, students of *ModernDive* are exposed to the `tidyverse`

suite of R packages. Designed with a common structure, `tidyverse`

functions are
written to be easy to learn and use. And, since most intro stats students are
programming newbies, *ModernDive* carefully walks the students through each new
function it presents and provides frequent reinforcement through the many
*Learning checks* dispersed throughout the chapters.

Overall, *ModernDive* includes wise choices for the placement of topics.
Starting with data visualization, *ModernDive* gets students building `ggplot2`

graphs early on and then continues to reinforce important concepts graphically
throughout the book. After moving through data wrangling and data importing,
modeling plays a prominent role, with two chapters devoted to building
regression models and a later chapter on inference for regression. Lastly,
statistical inference is presented through a computational lens with
calculations done via the `infer`

package.

I first met Drs. Ismay and Kim while attending their
workshop at the 2017 US Conference on Teaching Statistics. They
pushed us as participants to put data first and to use computers, instead of
math, as the engine for statistical inference. That experience helped me
modernize my own intro stats course and introduced me to two really
forward-thinking statistics and data science educators. It has been exciting to
see *ModernDive* develop and grow into such a wonderful, timely textbook. I hope
you have decided to dive on in!

*Kelly S. McConville, Harvard University*