Skip to main content

Clear-Sighted Statistics: Preface: Clear-SightedStatistica_Preface

Clear-Sighted Statistics: Preface
Clear-SightedStatistica_Preface
    • Notifications
    • Privacy
  • Project HomeClear-Sighted Statistics
  • Projects
  • Learn more about Manifold

Notes

Show the following:

  • Annotations
  • Resources
Search within:

Adjust appearance:

  • font
    Font style
  • color scheme
  • Margins
table of contents
  1. Clear-SightedStatistica_Preface

Except where otherwise noted, Clear-Sighted Statistics is licensed under a
Creative Commons License. You are free to share derivatives of this work for noncommercial purposes only. Please attribute this work to Edward Volchok.

Preface

I. Note to Students

Clear-Sighted Statistics is an open-access introductory statistics textbook that you can read for free on your computer, tablet, or smartphone.

The title Clear-Sighted Statistics stems from a simple observation: When we visualize data, the results of our analyses become readily apparent. My experience teaching introductory statistics classes tells me that you will be more successful learning statistics when you learn to visualize the data.

Throughout Clear-Sighted Statistics, I use data visualization to help you clearly see how to analyze the data and clarify the results of your analyses. Data visualization deal with creating graphic representations of data.

Microsoft Excel is used to demonstrate the statistical techniques covered in an introductory statistics course. Embedded in this ebook are numerous Excel files for the examples shown, as well as end-of-the-chapter exercises. Please explore these files.

You may ask why I chose to use Excel in Clear-Sighted Statistics. There are, after all, far more capable programs for statistics. Excel is a powerful and widely used spreadsheet. It is probably on your computer now and will be on your office computer when you get a job. Developing strong Excel skills may help you get a job and advance your career. I know that my Excel skills greatly helped me build a long career as a marketing consultant.

CUNY students get Microsoft Office 365 for free. This suite includes Excel, Word, Powerpoint, Access. CUNY students can find the link for Microsoft Office 365 on Blackboard. If you are not a CUNY student, you may be able to get Microsoft Office 365 for free or at a discount from your college.

Excel and the rest of the Microsoft Office applications run on Windows and Macintosh computers. You may be able to get Microsoft Office to run on some Chromebook computers. If you have a Chromebook, check this link. You may also be able to view Excel files on some Android phones and tablets, as well as iPhones and iPads. On more powerful tablets like the iPad Pro, you may even be able to edit Excel files. Check this link for details.

In addition to Excel, G*Power, a free program, is often used. G*Power is an important tool for Null Hypothesis Significance Testing. It is an easy-to-use application for determining sample size and calculating a priori and post hoc statistical power and the probability of a Type II error, which is a false negative. Here is the link to download G*Power: G*Power link.

Unfortunately, G*Power only runs on Windows and Macintosh computers. It will not run on Chromebooks, phones, or tablets. That is why online statistical power calculators are also demonstrated when appropriate. There are several such calculators. The ones from Statistical Kingdom are used in this book. Here is the link to the Statistical Kingdom power calculators.

II. Note to Faculty

You may ask why anyone would devote two years writing an introductory statistics textbook when there are so many on the market. My growing frustration with introductory statistics textbooks lead me to write Clear-Sighted Statistics. When I started teaching statistics at Queensborough Community College in 2006, I liked the textbook the department chose. Then in its 10th edition, this book compared favorably to those I remembered using as an under-graduate and graduate student in the late 1960s and 1970s. My department’s assigned textbook included graphics that made it relatively easy for students to understand statistics despite its occasionally obscure, abbreviated, or missing presentations of important concepts like p-values, effect size, and the calculation of Type II errors and statistical power.

When I started teaching statistics, I began reading the current literature on statistical methods. As a result, I grew increasingly dissatisfied with this and many textbooks that a steady stream of publisher’s representatives hawk. There have been important advances in statistical techniques since the presidency of Richard Nixon. Why have these advances not filtered down to introductory textbooks? The science of statistics, in general, and the practice of null hypothesis significance testing, in particular, have been hotly debated for over fifty years. The methods covered in introductory textbooks are increasingly considered inadequate. I frequently wonder how undergraduates who used the current crop of statistics textbooks will be able to understand even the most basic quantitative analyses that appear in peer-reviewed journals. Clearly our students deserve better from us.

It is not an exaggeration to contend that the science of statistics is in the midst of a paradigm shift, as evidenced by editorials and articles in the March 2016 and March 2019 issues of The American Statistician. In their forward to the 2019 special issue of The American Statistician, Ronald L. Wasserstein, Allen L. Schirm, and Nicole A. Lazar wrote:

The [2016] ASA Statement on P-Values and Statistical Significance stopped just short of recommending that declarations of “statistical significance” be abandoned. We take that step here. We conclude, based on our review of the [43] articles in this special issue and the broader literature, that it is time to stop using the term “statistically significant” entirely. Nor should variants such as “significantly different,” “p < 0.05,” and “nonsignificant” survive, whether expressed in words, by asterisks in a table, or in some other way. 1

The editorialist point out that the authors of the 43 articles in this issue “do not sing with one voice.”2 Reading these article will remind you of a line attributed to the great statistician John Tukey, “The collective noun for a group of statisticians is a quarrel.”3 As paradigm shifts go, we may be entering what philosopher of science Thomas Kuhn call the third stage, the adoption of a new paradigm.4 As Kuhn points out, paradigms are difficult to change. And, at this stage, the future paradigm is still murky.

Since my student days, there have been important technological advances for students of statistics. When I first took statistics as an undergraduate in 1969 and as a graduate student in the early 1970s, no one had a handheld calculator. I vaguely remember using a cardboard slide rule to calculate square roots. Today, every student carries a calculator on their cell phone, and computer software allows students to analyze large amounts of data quickly. And yet, many of today’s introductory textbooks devote a few passages to Microsoft Excel and statistical software like Minitab. A few textbooks are built around Microsoft Excel.

Excel shares the same shortcomings as many introductory textbooks: Its data analysis tools fail to address effect size, calculating statistical power and probability of Type II errors among other issues. This is disturbing because the American Psychological Association’s manual of style has for years urged authors to report effect size and statistical power. Excel also lacks tools for conducting post hoc analyses for omnibus tests like ANOVA. While missing some built-in statistical functions, Excel can, of course, be used to compute these measures. I discuss these “missing” analyses throughout ClearSighted Statistics. To enable students to calculate these often-ignored measures, I have included custom Excel workbooks.

I have relied on Excel since serving as a beta tester for its pre-launch version in 1984. It is a useful tool for both data analysis and data visualization. With Excel, I have built useful models that addressed my Fortune 500 clients’ myriad marketing problems. However, as much as I like Excel, I am aware of and frustrated by its shortcomings, some of which I point out in Clear-Sighted Statistics. These flaws are the reason why I use SPSS and G*Power for my own research, even after having conducted preliminary analyses with Excel.

You may ask why not incorporate more sophisticated applications like SPSS, R, or Python into ClearSighted Statistics? My answer is simple. Most of our students do not have access to SPSS or the time to learn R or Python in a fifteen week introduction to statistics course. And, most likely these applications will not be on their employers computers. In addition, I have had many conversations with business executives who advise my department on curriculum. When asked what is most important for students to learn, their top answer is Excel. Consequently, I used Excel, not dedicated statistical applications like SPSS, R, SAS, or Stata.

I have also incorporated G*Power into the sections on null hypothesis significance testing. Many researchers use G*Power to estimate the required sample size for a desired level of statistical power, as well as calculating the achieved statistical power. G*Power is a free application from the Heinrich Heine Universität Düsseldorf. It runs on Windows and Macintosh computers, but not on Chromebooks, phones, or most tablets. Here is the link to download G*Power: G*Power link. To assist students who do not have access to a Windows or Macintosh computer, I also use an online power calculator when appropriate. There are several online power calculators. I use the ones on the Statistical Kingdom website.

I assume that students are numerate; that they can perform basic arithmetic dealing with exponents, square roots, and negative numbers. Sadly, some students lack these skills. Please refer students who need to polish their basic math skills to Appendix 1, Math Review.

Success in introductory statistics should require more than the ability to calculate a simple test statistic using a handheld calculator, or the completion of publishers’ adaptive learning programs that rely on multiple-choice questions. Instructors should demand more of students. We must foster our students’ quantitative reasoning skills, as well as their information and data literacies. Students must become accustomed to explaining their analyses and findings clearly and succinctly.

Helping students hone their quantitative and statistical literacies is one of the biggest challenges in teaching introductory statistics, as well as one of its greatest rewards. In our data-drenched world, students need these skills as employees, consumers, and citizens. To this end, I encourage you to engage students in critical discussions of the analyses we teach. I also suggest short writing assignments—memoranda—that require students to report the results, limitations, and importance of their analyses.

In Appendix 4, I have posted Powerpoint lectures for all 19 chapters of Clear-Sighted Statistics. I have also included pdf files of these lectures. The pdf files show four Powerpoint slides on each page.

III. Acknowledgements

I thank Queensborough Community College and the City University of New York for granting me fellowship leave for the 2019-20 academic year. Without this sabbatical, I could not have completed Clear-Sighted Statistics.

I am also grateful to CUNY’s Professor Matthew Gold, PhD, and Jojo Karlin for helping me mount Clear-Sighted Statistics on the CUNY Commons. In addition, I must thank Queensborough Community College librarian William Blick for his help finding appropriate OER repositories, and to Megan Wacha, CUNY’s Scholarly Communications Librarian, for her invaluable assistance in mounting Clear-Sighted Statistics on CUNY Academic Works.

I am deeply indebted to the following people for reading Clear-Sighted Statistics and its ancillary files. I am grateful for their keen insights, helpful suggestions, and enthusiastic encouragement. Needless to say, I alone am responsible for any errors in the text.

Professor Dona Boccio, PhD

Professor Boccio of Queensborough Community College’s Mathematics & Computer Science Department reviewed Appendix 1, Math Review.

Professor Leslie Ward

Throughout my academic and professional life, I have been fortunate to have found superb librarians. These librarians helped me find needed resources and thereby contributed to the success of my projects. One of the finest research librarians I ever had the pleasure to work with is Queensborough Community College’s Leslie Ward. Ms. Ward reviewed a draft of Chapter 4: Where do Data Come From? I have also spoken to her about the OER movement and the Creative Commons License.

Dr. Arthur Beckman

Arthur Beckman, a fellow political scientist and advertising executive, read several chapters of Clear-Sighted Statistics.

Professor Emeritus Jonas Falik, PhD

Dr. Falik, the former chair of the Queensborough Community College Business department who had taught the department’s statistics class since the department initially offered it until his retirement in 2014, read nearly the entire manuscript. I am grateful to him for his comments, suggestions, and encouragement, as well as for giving me the opportunity to teach statistics. His wit, collegiality, and leadership are sorely missed in our department and across our campus.

Ms. Rosanna Volchok

Rosanna Volchok of the New York Academy of Sciences read several chapters and the Powerpoint presentations that accompany them. She has proven herself to be a careful proofreader and a constructive critic, which are among the many reasons that make her father proud.

Ms. Karen A. Frenkel

Karen A. Frenkel, technology journalist, documentary filmmaker, author, and my wife, read and edited drafts of Clear-Sighted Statistics. I greatly admire her endurance for doing so, and for her many other fine qualities that I have relied on daily since I met her in 1996. I would be lost without her by my side.

IV. About Edward Volchok

Dr. Edward Volchok is a professor in the Business department at Queensborough Community College. He has been a member of the faculty since 2006. Before joining the faculty, he spent 28 years as a marketing communications consultant. Dr. Volchok began his business career working in advertising account management at the New York City offices of Foote, Cone and Belding and Ogilvy & Mather. He was a founding member of two boutique consulting firms that focused on Category Management. He also worked as a senior marketing researcher at a major promotions agency and direct response agency. Most of his professional career has been devoted to providing Fortune 500 companies with innovative marketing solutions. Some of the companies he worked with were: Bristol-Myers Squibb, Campbells, Clairol, Coca-Cola, Cotton Incorporated, General Foods, Häagen-Dazs, J. P. Morgan, Kellogg’s, Kraft, Life Savers, McDonalds, Nestlé, PepsiCo, Quaker Oats, Sandoz, Schering, United Technologies, Warner-Lambert, and Wrigley’s.

Since 1999 Dr. Volchok has taught at New York University’s School of Professional Studies, Stevens Institute of Technology’s graduate business program at the Howe School of Technology Management, CUNY’s New York City College of Technology, and CUNY’s School of Professional Studies.

Dr. Volchok earned his Ph.D. in political science from Columbia University.

Reference

1

Ronald L. Wasserstein, Allen L. Schirm, and Nicole A. Lazar (2019) Moving to a World Beyond “p < 0.05,” The American Statistician, 73:sup1, p. 2. https://www.tandfonline.com/doi/pdf/10.1080/00031305.2019.1583913?needAccess=true.

2 Ronald L. Wasserstein, Allen L. Schirm, and Nicole A. Lazar (2019) Moving to a World Beyond “p < 0.05,” The American Statistician, 73:sup1, p. 1.

3 David R. Brillinger, “…how wonderful the field of statistics is…” In Past, Present and Future of Statistical Science, Edited by Xihong Lin, Christian Genest, David L. Banks, Geert Molenberghs, David W. Scott, and Jane-Ling Want, (London: Chapman and Hall/CRC, 2014). https://www.stat.berkeley.edu/~brill/Papers/copssbrill.pdf.

4 Thomas Kuhn, The Structure of Scientific Revolutions, (Chicago: University of Chicago Press, 1962), p. 91.

Annotate

Creative Commons
Powered by Manifold Scholarship. Learn more at
Opens in new tab or windowmanifoldapp.org