Preface

I have dedicated this book to my statistics students, former and future, all of them. Future, because it’s all for them; they’ll be the ones making use of it.

Former, because over the years they have been showing me (and, in many cases, telling me in no uncertain terms and with great emotion) how their first experience with statistics went. Because, somehow, along the way they have also taught me how to teach statistics to them. Not to a mass of generalized “undergraduate social science students in an introductory stats class,” with my initial preconceived idea of these students’ abilities, prior knowledge and needs, no — but to the actual them, the very real people I see in my classes. During the almost ten years of “SOCI 2365 Introduction to Social Research Statistics” instruction at Kwantlen Polytechnic University, I have learned how best to approach teaching stats to my students, in accordance to their actual academic needs and their actual academic abilities.

So who are the students in my classes? (Forgive me, now I’ll have to generalize after all.) The typical student in my introductory stats class tends to be there because they have to (the course is compulsory for our major, along with a handful of others); is majoring sociology; is likely “not very good with math” and, therefore, has delayed taking the course as much as possible because, understandably, they are terrified. I could have used “she is” instead of the gender-neutral “they are”– I typically have more female than male students. That is not to say that students not fitting this profile don’t take up my class; they do, and they’re not few. This example simply gives me the opportunity to give you a taste of what the book will be about: statistics and sociology.

See, “tends to”, “likely”, and “on average” are all terms with specific statistical meaning (as much as they can be misused and misinterpreted in conventional, everyday usage) — but you’ll have to go further into the book for that. However, I can easily tell you that I also have students, many of them, who are not majoring in the social sciences, are in their second year (as they are supposed to), are great with math, and who find the course easy. Of course, many of my students are also male. Obviously, none of what I just said contradicts the description of my typical student (and if it’s not obvious, you definitely need this book). The “typical student” description is simply based on a brief statistical profile of an average class I usually have. The various characteristics I listed may or may not be statistically associated with each other, not to mention anything about causal association. (Were you perhaps thinking that, say, women in my classes are the ones “not good with math” while men “find it easy”? I actually never said, not even implied, that. But now you see how easily statistical information can be misinterpreted and how statements based on statistical information can be taken to mean more than they actually do.)

Why sociology though? The description above can lead us to a few questions (i.e., we can formulate hypotheses), like, are students majoring sociology (or other social sciences, except economics) really more likely to say they are “not good at math” than, say, students in the natural sciences? For that matter, are women on average more likely to major in social sciences and humanities than in the STEM (science, technology, engineering, and mathematics) fields? The answers to these questions  can be found through statistical analysis (both are “yes” by the way) but the explanations (or theories) — i.e., why we observe the relationships between gender, major, and perceived math ability — are profoundly sociological.

In a similar vein, throughout this book I will bring up questions of sociological relevance, I will refer to sociological theories, research and findings, I will give sociological examples, and ultimately I will use sociological data.  

Why does that matter? Stats is stats, right?.. Hmm, yes, and no — and in the case of applied statistics, as the current text is, rather no. Yes; if you go by the table of contents, you’ll see what one typically sees in a generic introductory statistics book (for social scientists); statistics is a set of tools, and it can be presented as generically and as generally as possible. However, like any tool, its value is higher the more specialized it is (you can take an ailing tooth out with a hammer yet arguably it’s better to use specialized dental equipment). Like any tool, it also matters what it is used for and how.

In other words, in this book the statistics instruction will be specialized: from a sociologist (granted, herself specialized in social statistics) for sociologists. (If you are neither a sociology student or sociology instructor, you can take this as sort of a caveat emptor clause: buyer beware.) To the extent that sociology itself is a rather broad discipline and its use of statistics is equally as broad, one could use the book as an introduction to social science statistics. However, I do not go out of my way to engage in statistical instruments more frequently used in, say, criminology or psychology (i.e., small-size court case data, or experiment data, etc.).

I’ll give you a different example: If you open an introductory psychology textbook, you will likely find a chapter on Sexuality and Gender. Yet “gender” and “sexuality” are also huge topics in sociology, and any introductory sociology textbook also has a chapter on them. There will be some overlap in the treatment of the topic by the two disciplines, but you’d be wrong to expect everything — or even most — to be the same.

Simply put, psychologists and sociologists generally tend to ask different questions, to approach a topic differently, to have different concerns, to have different preferred methods for collecting and analyzing (quantitative) data, and to even reach different conclusions, and to therefore offer different theories (as one would expect from two separate disciplines). Why wouldn’t we want specialized statistics for each discipline?

Think of this book as a crash course in statistics. As such, I make these promises:

1) I promise to include only what is absolutely necessary.

2) I promise to skip on fluff and padding and any other material that is not strictly relevant to the exposition.

3) I promise to avoid repetitiveness as much as possible and instead explain everything only once but slowly and patiently.

Given my promise, this book provides a necessarily brief introduction to statistics. It is also a conventional introduction in that, as almost all such books, it does not include all there is about some of the more complex concepts, i.e., it is not entirely truthful.

Don’t get alarmed by this admission. Rather, think of this introduction as your first date with statistics. No one tells all and bares all their secrets on a first date, do they? (…Or it might be their last.) Some things need to be revealed at a later time, once you’ve come to know your love interest better. Statistics is like that too. Some advanced concepts and relatively new developments in the discipline would only make sense to you only after the initial period of getting to know it has passed; then you can learn more “truthfully” and understand in what way and why the tools and concepts were simplified when they were first introduced to you.

And if you never get to “a second date” with statistics, never fear. What you will learn from this brief introduction will be quite practical “in real life” and still will serve you well. (You’ll just know there is more to what you’ve learned — but that’s the case with everything, no?) You will learn the basics of summarizing data and extracting useful information out of it; how data can be manipulated and how and why not to do that; how and when you can generalize from data and the limits to your generalizations; what role probability and uncertainty play in statistics; how to interpret basic statistical information; what to look for in existing statistical reports; and how to execute a basic statistics report on your own. You will learn how to talk about statistics, and how to write about statistics. Finally, you will learn where to go from here, should you ever feel like going on a second date with statistics after all.

Given the purposefully streamlined content, some will not like this book. If you are an instructor (or a student) looking for theoretically comprehensive and expansive introductory treatment of statistics, this is not the book for you — but you also know many such books exist, freely available online or otherwise. Statisticians will likely be severely displeased by some of the things missing here, as compared to a truly conventional introductory statistics text.

But this indeed is why this book exists at all: to only include what I’ve discovered my students need in order to have a basic working knowledge about the most useful and most frequently used simple stats tools.