Welcome to the fourth edition of Statistics for Nursing: A Practical Approach—I can’t believe we are on the fourth edition of a book I never set out to write! The idea for this book first began when I was teaching in our registered nurse (RN) to bachelor of science in nursing (BSN) program. I realized how many of our highly experienced nurses had taken statistics but didn’t leave the class with an in-depth understanding of the essential concepts we regularly use in clinical practice. Later, I discovered how many nurses did not return to school simply because they were so intimidated by the idea of taking statistics. I started putting together content because I really believe that if we teach statistics to nurses within the context of nursing, it is easier to understand and more valuable in the long run. I don’t really care if you can plug numbers into formulas—a computer can do that. Instead, I want you to understand when and how to use the concepts and ideas in statistics to improve your patient care. That is the focus of this book.

Whether you are an RN going back for your bachelor’s degree or a nurse ready to tackle advanced practice, this book was written for you! I see the incredible gifts you bring to our profession and want to remove the statistics barrier to both returning to school and going even further with your education, training, and practice.

The pandemic has brought public health to the forefront of our practice, so this edition expands on some of these concepts. It also adds a few new features. One of these, titled “Where Students Often Make Mistakes,” highlights common errors. These are points I would make in class and highlight by putting three stars next to them. You know they will show up on your quiz or exam, and you’ll kick yourself if you miss them! I’m calling out these points in this edition to try to help you avoid making these common mistakes.

I’ve also added some new computer applications. Often, computer applications are the most stressful part of learning statistics. This edition introduces a statistics program that helps make learning a new computer application a little less stressful: Intellectus Statistics. This is a web-based statistical software program designed to help nonstatisticians learn to conduct research and complete statistical analyses appropriately. When I first began using this program, I was thrilled to think about how helpful it would be for my students, and sure enough, they have done some great work with it. Often, trying to learn statistical computing is a challenging and frustrating process for nonstatisticians. This program simplifies the process of learning the software while helping students produce and understand the actual statistics content.

I rarely hear anyone say how much they love using statistical software. My brother Brendan, who authors the “From the Statistician” sections in the text, is the rare exception to the rule here. It is not uncommon to find him smiling and writing SAS code at the same time. I do not understand how that happens! Take it from someone who found the process of learning statistical programming truly painful: Intellectus Statistics is a vast improvement for beginners. There is still a learning curve, though, so give yourself the necessary time, and expect a little bit of frustration until you master the process. Once you do, you will find it to be quite a powerful tool to have. Then talk to your friends who are taking statistics elsewhere and hear about what they struggle with while using other programs—you will thank your lucky stars that you are using Intellectus Statistics!

This edition has updated research articles to help you see how each analysis technique is used in actual practice. I also added an article to Chapter 7 so that you can see some of the background work that goes into putting a study together in the first place. There are now more review questions at the end of chapters as well. No matter how many there are, I keep hearing that you want more! We also expanded the test bank for instructors, which will facilitate the release of more practice questions from older editions as well.

Part of evidence-based practice is ensuring that published evidence is of high quality and does not reach any unsupported conclusions. This is why academic journals rely on outside reviewers who are content experts. Articles go through a peer-review process, and the feedback provided is given to both the journal editors and the authors of the articles themselves. It is critical to the profession that expert nurses serve as reviewers. I encourage all of my students to consider offering their expertise as reviewers for nursing journals. I have found that many nurses are a little intimidated by the idea of doing so. Part of the job of a reviewer is to recognize when something about the analysis is concerning. Many of us assume that published literature reaches legitimate conclusions, but this is only the case if the peer-review process has been thorough and complete. Recognizing issues in submitted articles, abstracts, and grants helps ensure that only accurate and well-designed studies are disseminated and incorporated into care standards. For this reason, this edition introduces a new section called “What Went Wrong?” These short excerpts ask you to consider something that has been submitted for review but has a significant concern within it. You will use your statistics knowledge to detect the concern and report why the submission should be reevaluated before being considered for implementation, publication, or funding. When you only see what is published already, you may not realize how often submitted work isn’t up to par. Recognizing these issues helps the entire profession and may make you more comfortable stepping into a professional reviewer role.

As always, I appreciate hearing from students and instructors using the text. Your thoughts and ideas have often been incorporated into new editions of the book, and it always makes me smile to hear that another nurse is doing well in statistics and understands the concepts. I do sometimes receive critical emails or reviews as well. One such email came from a student who felt that presenting gender variables as binary was not inclusive and made people who identify as nonbinary feel unrecognized and diminished. I never intend to make anyone feel devalued in anything that I write, so I would like to extend my deepest apologies to anyone I made feel this way. I believe the world is made up of many unique and valuable individuals, all of whom deserve the same opportunity to live, love, and prosper.

The writer of this email did identify one of the inherent problems with categorical variables: they are, by nature, not inclusive. No matter the attribute, once we create categories, there will always be people or observations that don’t identify as one of those categories. Some researchers choose to handle this issue by creating several categories for the groups they identify as the largest and using a catchall group like “other” or “none of the above,” whereas others identify the dominant category and group all the rest in the nondominant category. For example, if the variable is marital status, the researcher may opt for the categories of “married,” “divorced,” and “other,” or the researcher may simply use “divorced” and “not divorced.” In either case, you can see that the categories lose some nuance. Asking open-ended questions that allow for all ranges of the attribute to be expressed, such as “I’m living with someone,” “I’m legally separated,” or “I’m widowed,” provides more information. It becomes clear that an accurate category doesn’t exist in the options provided to these subjects.

Depending on the study, the type of information lost by using the categories may be important or not. For example, if the study is about how eating breakfast affects calories consumed at lunch, the marital status of the subjects might not be important at all. However, if relationship satisfaction is the outcome of interest, marital status may need to be explored in more detail. Researchers also must recognize that categories are not fixed over time. For example, if you ask about marital status in 2020 and the subject reports being married, that doesn’t mean the same subject will be married at the conclusion of the study in 2022. Researchers must make every attempt to design valid and useful categories under the circumstances, but we also have to recognize the limitations inherent in categorical variables.

Why not just create lots and lots of categories to include as many people as possible in an accurate category? Unfortunately, the more categories you have, the more inadequate cell counts you may create for your analysis. You can exaggerate differences that may not be critical to the study hypothesis and increase your chances of missing a difference between your study variables that really does exist. On the other hand, if you compress the categories too much, you increase your chance of reporting a difference that doesn’t really exist because the observations in the compressed category now appear more alike than they truly are.

A researcher needs to realize that anytime they use categorical variables, there is a loss of inclusivity and a potential to over- or underrecognize differences that exist, and some groups may be favored or insufficiently responded to because of the created categories. These potential negative ramifications of categorization must be balanced with the value of finding meaningful information within the data. Whenever possible, include scale-level variables and/or meaningful categories that you can support with the sample size you collect while also realizing that human beings and observations never fit into categories perfectly because of our broad, unique, and remarkable diversity.

I hope you all find the fourth edition of the text useful as you travel along in your professional career. It is my greatest wish that each of you finds and follows the path that takes you to your fullest capacity and that you both value and treasure what you mean to yourselves, your families, each other, and the world. I have watched so many former students and colleagues do exceptional things, whether that involves changing national health policies or nursing newborns in the middle of the night. Your gifts, large and small, make a difference every day. Be proud of who you are and all you do. The world would not be the same without your caring presence and all you give to your children, parents, partners, colleagues, patients, and friends. Be kind to yourself. Being a nurse is not an easy path, but your service changes the world every day.