
Welcome
Arguing with Data is about a new way to approach introductory statistics: hands-on, real-world, and conceptually-oriented (in other words, less calculation intensive). The Arguing approach merges foundational concepts—randomness and probability, sampling theory and confidence intervals, correlation and regression—with practical research design thinking—such as, “What are the basic principles of good research?”, “How do I design a survey?”, “How do I design an experiment?” The learning goal in this approach is knowing how to evaluate real research (political polls, experiments) for its use of credible methods. Arguing with Data posits that statistical learners should start their journey into this rewarding subject by thinking of themselves, from the beginning, as potential researchers and as community experts.
Quick Access
Arguing with Data Textbook
An introduction to statistics for activists, journalists & policymakers
Learning to use GeoGebra
This activity introduces students to making data displays in GeoGebra.
Learning to use Jamovi
Frequently Asked Questions
I wanted to answer some questions you might have about who I am, this curriculum, and how it can be used.
Yes. I have a B.A. in math from Columbia, a master’s in teaching math, a research-based master’s in communication, and a decade’s experience teaching regular and A.P.-level statistics. I read all 400+ sources cited in the book, plus many others. That said, we can all make mistakes, so please let me know of any errors or omissions you spot!
No, I think it would work for a first course in college. Adult learners might also find this curriculum useful and interesting. I included the answer key for each quiz for the self-directed learner; they can read a chapter, take the quiz, correct themselves, and if necessary, take a second version of the quiz. (I did not include answers keys for unit tests or exams to preserve them for classroom assessments.)
This is unapologetically a regular-level statistics curriculum. Students will only learn a few calculations (e.g., confidence intervals, probability). Instead, students will focus on acquiring conceptual understanding first, for example, the concept of what a confidence interval is and how to think about probability.
I’m a regular classroom teacher, not a major textbook publisher. I’m planning to submit this curriculum in my home state of Oregon for their approval, but I lack the resources to submit it to 49 other states. If you’re at a school where you can use your professional judgment to decide what materials to use, great! You don’t need state approval. If you’re at a school where that’s not the case, might I suggest using this textbook as a supplement? You can pick and choose whatever parts you like.
I think so! (For those of you who don’t know, some states are recommending two high school years of algebra and geometry, followed by a third high school year of math in another pathway, such as data science.) But as stated before, this is unofficial.
It’s my belief—and the opinion of a lot of experts—that a simplistic, formulaic approach to p-values lead to the current replication crisis in several research fields. The better, and frankly easier, alternative is using confidence intervals instead. However, I included both p-values and confidence intervals in the section on the interpretation of research results because p-values are so prevalent. Calculating p-values isn’t covered in this textbook; that’s for students to learn in a future course, if they choose to go on in statistics.
Yes, check out the American Statistics Association‘s recommendations, the K–12 portion of which was also signed by the National Council of Teachers of Mathematics. I’ll quote the A.S.A.: “The ultimate goal: statistical literacy for all” [emphasis theirs], which they define as “having a healthy dose of skepticism about findings based upon data. A statistically literate high school graduate will be able to evaluate conclusions from data and judge the legitimacy of reported results” (p. 5). If you read the report, it’s clear that professional and research statisticians would like to see more discussion in K–12 statistics education of different methods of data collection, including looking at real data that require clean up, as well as how to ask statistically-appropriate research questions.
Everything is copyrighted under a Creative Commons license. The textbook is a no-derivatives allowed copyright, but everything else is free for you to alter as you see fit. All I ask for in return is that you share materials with me: interesting problem sets, worksheets, applications, and the like. I hope that this website can be a community of people who share resources to help us teach statistics in this new way.
Do you have a Lesson plan or teacher resource to share?
Submit your resource or get in touch through our contact form and I can get it added to our resource library page.
About Russell
Russell has a B.A. in math from Columbia, a master’s in teaching math, and a research master’s in communication. He has been teaching for two decades, including regular and A.P.-level statistics, and is currently working at Northwest Academy in Portland, Oregon. Russell has presented at numerous math conferences about statistics education (as well as other math topics). He’s a GeoGebra ambassador and coaches other math teachers regularly in online sessions about this free middle school/ high school/ college software. He’s also got very good Excel kung fu. In his free time, Russell snowboards the sick pow on Mt. Hood (it’s usually cascade cement, tbh). You can read more of his musings on his blog.