This is an archive of information/links from past StatChat meetings.

Our next StatChat is Thursday September 26, 2019 from 6:00–8:00pm at Macalester College (Olin-Rice 301). The agenda for the meeting will include:

**6:00–6:30 Settle In & Eat****6:30–7:45 Guest Speaker:**Tim Erickson will present on the using the Common Online Data Analysis Platform (CODAP).*Be sure to bring your computer as this will be an interactive session.***7:45–8:00:**Discussion, wrap-up, and future StatChats

We hope to see you there! If you plan to attend, please RSVP by signing up to bring food or beverage to share. And please spread the word among any colleagues that may be interested in the night’s topics.

Our next StatChat is Wednesday October 30, 2019 from 6:00–8:00pm at Macalester College (Olin-Rice 250). The agenda for the meeting will include:

**6:00–6:30 Settle In & Eat****6:30–7:45 Guest Speaker:**Kris Gorman from the*Center for Educational Innovation*at the University of Minnesota will discuss the nebulous balance between breadth and depth in statistics education.*Abstract:*In order to help students retain information long term and transfer that knowledge to new situations, there’s research to suggest that rather than cover everything (i.e. provide full breadth), it’s better to cover some things more thoroughly (i.e. provide depth). Kris Gorman will share findings about depth vs breadth and a few examples of how instructors at UMN have altered courses to incorporate more depth. There will be time for discussion about what this might mean for choices in our own teaching.**7:45–8:00:**Discussion, wrap-up, and next StatChat (We need to pick place to have our first ever StatChat social.)

We hope to see you there! If you plan to attend, please RSVP by signing up to bring food or beverage to share. And please spread the word among any colleagues that may be interested in the night’s topics.

**Resources**

- Luckie, D. B., Aubry, J. R., Marengo, B. J., Rivkin, A. M., Foos, L. A., & Maleszewski, J. J. (2012). Less teaching, more learning: 10-yr study supports increasing student learning through less coverage and more inquiry.
*CAdvances in Physiology Education, 36*(4), 325–335. doi:10.1152/advan.00017.2012 - Schwartz, M. S., Sadler, P. M., Sonnerts, G., & Tai, R. H. (2008). Depth versus breadth: How content coverage in high school science courses relates to later success in college science coursework.
*Science Education, 93*(5), 798–826. doi:10.1002/sce.20328

As we haven’t met in quite some time, this meeting will be quite informal—meeting new friends, catching up, and ad hoc sharing of what is happening in your statistics/data science classroom. We will also use the time to plan the dates and agendas for future StatChats. If you can, please bring a snack or beverage to share.

**Location:**Olin-Rice Hall (Room 243) on the Macalester Campus [map]**Day:**Wednesday November 14, 2018**Time:**6:00–8:00pm**Parking:**There is a free parking lot (off of Snelling Avenue) adjacent to Olin-Rice.

Our next **StatChat has been rescheduled** due to the upcoming polar vortex. The new date will be Wedesday February 13, 2019 from 6:00–8:00pm at Macalester College (Olin-Rice 243). The agenda for the meeting will include:

**6:00–6:30 Settle In & Eat****6:30–7:15 Material Potluck:**Paul Roback from St. Olaf will share some teaching materials they use in their data science course with the group. The materials can be accessed from Paul’s github site (here).**7:15–8:00 Book Club Discussion:**Danny Kaplan and Milo Schield will lead a discussion of*The Book of Why*(see below for more information). During this StatChat we will discuss:- Introduction (pp. 1–21);
- Chapters 1 (pp. 23–51); and
- Chapter 2 (pp. 52–91)

**Resources**

- Paul’s materials on discrimination in Mississippi jury selection
- Danny’s slides on the
*Book of Why*

**Book Club 2019**

During the 2019 spring semester, StatChat will be reading and discussing *The Book of Why: The New Science of Cause and Effect* by Judea Pearl and Dana Mackenzie.

“Correlation is not causation.” This mantra, chanted by scientists for more than a century, has led to a virtual prohibition on causal talk. Today, that taboo is dead. The causal revolution, instigated by Judea Pearl and his colleagues, has cut through a century of confusion and established causality–the study of cause and effect–on a firm scientific basis. His work explains how we can know easy things, like whether it was rain or a sprinkler that made a sidewalk wet; and how to answer hard questions, like whether a drug cured an illness. Pearl’s work enables us to know not just whether one thing causes another: it lets us explore the world that is and the worlds that could have been. It shows us the essence of human thought and key to artificial intelligence. Anyone who wants to understand either needs The Book of Why.

Below are the links for ordering the book from Amazon and Barnes and Noble:

Also, there is an unedited working copy of the introduction and first two chapters available online at:

Our next StatChat is Wedesday March 27, 2019 from 6:00–8:00pm at Macalester College (Olin-Rice 243). The agenda for the meeting will include:

**6:00–6:30 Settle In & Eat****6:30–7:15 Material Potluck:**Ann Brearley and Laura Le will present on the updated “The Islands”, a virtual world developed by Dr. Michael Bulmer at the University of Queensland. They will bring a guided activity for you to explore “The Islands” and also describe the various ways “The Islands” have been used by us and others at the University of Minnesota, Biostatistics Division.*Be sure to bring your computer so you travel to the “The Islands” from the comforts of Macalester College.*

You may also want to read the following article for some background on “The Islands”:- Bulmer, M., & Haladyn, J. K. (2011). Life on an island: A virtual population to support student projects in statistics.
*Technology Innovations in Statistics Education, 5*(1).

- Bulmer, M., & Haladyn, J. K. (2011). Life on an island: A virtual population to support student projects in statistics.
**7:15–8:00:**TBA

The April StatChat will be Wedesday April 24, 2019 from 6:00–8:00pm at Macalester College (Olin-Rice 243). The agenda for the meeting will include:

**6:00–6:30 Settle In & Eat****6:30–7:15 Material Potluck:**Amelia McNamara (St. Thomas) will present on her “Handmade Data Viz” activity, which she has used in several settings; once with introductory-level undergraduates in a*Communicating with Data*course, and once at a conference for journalists. Amelia will show how she scaffolded the activity with inspiring handmade visualizations from people like Mona Chalabi, Giorgia Lupi, Stefanie Posavec, and others. Then she will show some work created by her students. Time permitting, we can try a little visualization ourselves! (Amelia even offered to bring craft supplies.)**7:15–8:00 Book Club Discussion:**TBA (Book of Why)

**Resources**

Here are some resources related to Amelia’s talk that you may find of interest.

- Dear Data book: Giorgia Lupi and Stefanie Posavec’s book collects a year’s worth of hand drawn data visualizations that the two authors would send to each other via postcards.
- Dear Data Website
- Observe, Collect, Draw!: A Visual Journal: Giorgia and Stefanie also recently published a second collaboration which is described as a guided journal of self-examination while simultaneously functioning as a mini-course in information design:
- Mona Chalabi’s Instagram page

**Speaker:**Xiao-Li Meng (Harvard University)**Topic:**Statistical Education and Educating Statisticians: Producing Wine Connoisseurs and Master Winemakers

Xiao-Li will talk about his innovative course in intro stats as well as his program to prepare excellent teachers of statistics. Following the talk, you are invited to stay on and visit with Xiao-Li as part of an informal social hour with our statistics education faculty and graduate students.

- Stat 105 at Harvard

**Speaker:**Charles Guyer (University of Minnesota, School of Statistics)**Topic:**What Statistics 101 Doesn’t Teach, But Should

Sponsored jointly by StatChat and Project MOSAIC

**Speaker:**Michael Bulmer (University of Queensland, Australia)**Topic:**The Island: Letting Students Experiment and Collect Data

Access the Webinar Recording here.

**Dessert Activity:**The Omnipresence of Coincidence (Milo Schield; Augsburg University)

Coincidence is much more likely than expected, leading many people to conclude that there is something more going on than “mere” coincidence. Educators often see this differently, and ponder how to lead students to a more accurate idea of “expected.” I’ll describe some spreadsheets to make the unseen more visible and help students challenge and develop their notion of “expected”. The spreadsheets demonstrate runs with coins, linear and non-linear clusters in a two-dimensional grid, and the Birthday problem.

**Journal Club:**Arnold, P., Pfannkuch, M., Wild, C., Regan, M., Budgett, S. (2011) Enhancing atudents’ inferential reasoning: From hands-on to “movies”.*Journal of Statistics Education, 19*(2).**Speaker:**Rob Gould (UCLA)**Topic:**Educating Citizen Statisticians

What do we want out students to learn in an introductory statistics course? Historically, the answers have ranged from “How to compute …” to “How to read the newspaper” to “How to analyze data” to “It depends on who the students are.” Most of these discussions took place in a context in which data were hard to come by and statistical analysis tools were expensive. But today we live in a world where data are ubiquitous and anyone with an internet connection can analyze data. In this new data-driven world, I will argue that there is a core curriculum needed by all students, regardless of major and that the purpose of this core is to teach them to be Citizen Statisticians. We’ll discuss how the new textbook I wrote with Colleen Ryan, Introductory Statistics: Exploring The World Through Data, was written to provide a core statistics education to all students regardless of background or mathematical preparation.

**Speaker:**Allan Rossman (Cal-Poly)

Joint meeting of StatChat and the Twin Cities chapter of the American Statistical Association.

**Guest Speaker:**Daniel Kaplan (Macalester College)**Topic:**Recent U.S. Supreme Court Finding on Statistical Significance

“Supreme Court decisions famously reflect political divides, but they also are a product of the multiple simultaneous objectives of the legal system: creating incentives for honest behavior, distributing risk and cost among competing parties, and maintaining predictability, among others. Statistical theory was created for a completely different set of objectives, of which a major one is supporting an idealized scientific process of nonpartisan, disinterested investigation. The Matrixx case inhabits a domain where both sets of objectives apply: the desire for informed scientific judgment and the exigencies and conflicts of civic life. Whenever multiple objectives are involved, it’s likely that not all of the objectives will be fully realized. Or, to quote Dickens’s Mr. Bumble,”The law is an ass." The question is whether statistics is likely to be any less so when dealing with complex matters of allocation and decision-making. I’ll discuss in particular the extent to which the canonical set of topics covered in university-level statistics education is oriented toward supporting quantitatively sophisticated, effective decision making in the civic domain.

**Journal Club:**- Wieman, C. (2009). Why not try a scientific approach to science education? [Blog].
- Wieman, C. (2009). A scientific approach to science education—Research on learning. [Blog]. Wieman, C. (2009). A scientific approach to science education—Reducing cognitive load. [Blog]. Wieman, C. (2009). A scientific approach to science education—Beliefs, guided thinking, and technology. [Blog]. Wieman, C. (2009). A scientific approach to science education—Technology and institutional change. [Blog].

**Speaker:**Chad Topaz (Macalester College)**Topic:**Teaching with Blogs

Personal response system (PRS) clickers are small, handheld electronic devices that students use to respond to instructor-posed questions in class. Blogs are online diaries that students can use to chronicle their learning outside of the classroom. I will discuss some pedagogical and technological aspects of using blogs and clickers to support student learning in quantitatively-focused courses. I will also connect clicker and blog pedagogy to two learning science frameworks, namely the four “centrisms” of Bransford et al., and the revision of Bloom’s taxonomy by Anderson and Krathwohl.

- Higdon, J., & Topaz, C. (2009) Blogs and wikis as instructional tools.
*College Teaching, 57*(2), 105–109.

**Journal Club:**Kuiper, S., & Collins, L. (2009) Guided labs that introduce statistical techniques used in research from multiple disciplines.*The American Statistician, 63*(4), 343–347.**Stat-o-Lanterns:**Presentation of this year’s Stat-o-Lanterns! Bring your own statistical squash, likelihood lantern, jacknife jack-o-lantern, or probability pumpkin!

*Image credit:* Photos of previous years’ Stat-o-Lanterns thanks to Michael Huberty.

**Speaker:**Chad Topaz (Macalester College)**Topic:**Teaching with Clickers

Personal response system (PRS) clickers are small, handheld electronic devices that students use to respond to instructor-posed questions in class. I will discuss some pedagogical and technological aspects of using clickers to support student learning in quantitatively-focused courses.

**Speaker:**Bob delMas (University of Minnesota, Educational Psychology)**Topic:**Applets in R

You can create small, interactive programs in R. I’ll show two of these. Rates Beat Humans simulates a guessing game from a psychology study. Another, which I’ll call With or Without Replacement? That is the Question, is designed to help students understand the distinction between sampling with and without replacement, as well as when they are used. I will also demonstrate two GUIs (graphical user interfaces): one for conducting randomization tests and the other for estimating bootstrap confidence intervals.

**Speaker:**Josh Paulson (RStudio)**Topic:**R-Studio: Delivering R through a Web Browser

R-Studio lets you interact with R through a standard web browser. It allows you to move easily from one computer to another without losing your place, and provides other user-friendly features. The beta version is being used at several colleges and universities across the country. I’ll talk about the applications to teaching to making R more accessible.

**Speaker:**Danny Kaplan (Macalester College)**Topic:**Teaching Calculus with R

It’s likely that you have two immediate reactions to the title. (1) R is for statistics, not calculus. If you’re going to use software at all for calculus — and it’s not clear that you should — you should use a symbolic algebra package like Mathematica or Maple. (2) I don’t teach (or study or use) calculus, so why should I care? In fact, it is more or less a historical accident that calculus is taught through algebra. In the late 1600s through 1900, that was the only technology that was available. The basic techniques of calculus — differentiation and integration — have no intrinsic relationship to the algebraic techniques. It’s perfectly reasonable to present them entirely as simple software operations. And why you should care? Many of your mathematical ideas were formed by the calculus-oriented high-school curriculum that you went through. In those high-school years, you were drilled with algebraic and symbolic techniques. So you’re good with algebra, even though you may rarely use it. Seeing how to represent and perform those operations with software can help you to develop general computational concepts and skills.

**Discussion:**Why have an R Users’ Group? What can it do for me?

**Journal Club:**Seife, C. (2010).*Proofiness: The dark art of mathematical deception.*New York: Viking.**Speaker:**Katie St Clair (St. Olaf)**Topic:**Team-Based Learning

Team-based learning (TBL) is a pedagogical strategy that involves groups of students working together in teams to learn and apply the course concepts. I will give an overview of TBL principles and talk about how we’ve used TBL in our stat literacy class at Carleton.

**Journal Club:**Cobb, G. (2007). The introductory statistics course: A Ptolemaic curriculum?*Technology Innovations in Statistics Education, 1*(1).**Speaker:**Joan Garfield, Robert delMas, Laura Le, Rebekah Isaak, Laura Ziegler, and Andy Zieffler (University of Minnesota, Educational Pyschology)**Topic:**The CATALST Course

CATALST is an NSF-funded project that has created a curriculum designed to develop students’ statistical thinking and appreciation of statistics through a focus on modeling, simulation and inference. The CATALST curriculum is currently being taught in several sections of undergraduate-level statistics at the University of Minnesota and at NC State. We will share an overview of the curriculum, as well as a sample of class activities used in the course. Specifically, an activity used to introduce the randomization test for group inferences will be shared. The software TinkerPlots, which is used by students in the course to conduct the modeling and simulation, will also be demonstrated.

**Journal Club:**Bellos, A. (2010). Situation normal. In,*Here’s looking at Euclid: From counting ants to games of chance - An awe-inspiring journey through the world of numbers*(pp. 246–265). New York: Free Press.**Speaker:**Robin Lock (St. Lawrence University)**Topic:**Starting Inference with Bootstraps and Randomizations

Computer-intensive methods such as bootstrapping and randomization tests provide a way to introduce students to fundamental ideas of statistical inference that require relatively few prerequisites. I’ll discuss an ongoing project to revamp an introductory course to use such methods as the starting point for inference.

**Journal Club:**The Math-Stats Course. Some questions: What should be the goal of the math stats course? What topics are essential?- TOC from Larsen and Marx, An Introduction to Mathematical Statistics
- TOC from Chihara and Hesterberg, Mathematical Statistics with Resampling and R

**Speaker:**Milo Schield (Augsburg University)**Topic:**Models and Assumptions: Statistics and Assembly

Checking assumptions is a critical activity in modeling. Often times the assumptions used in analyzing data influence – if not determine – the results. I’ll discuss an ISI draft paper that argues that statistics have the same status as models – they involve choices in assembly that influence – if not determine – the results. One of the five elements of the AACU 2009 Quantitative Literacy Rubric was assumptions: “Ability to make and evaluate important assumptions in estimation, modeling, and data analysis.” The ISI paper extends this focus on assumptions to include the formation of categories, measures and summary statistics. While statistics may have little to say about which assumptions are best, statistical literacy can highlight how choices in how groups are defined or how quantities are measured can influence the results.

**Speaker:**Marc Isaacson (Augsburg University)**Topic:**USCOTS Theater-Teaching Activity: Multiple choice Olympic Success

This activity incorporates audience involvement to evaluate data from Olympic Competition results through the use of rankings. While rankings are commonly encountered by students in their everyday life, they are rarely discussed in an introductory statistics class. Audience members will be presented a handout, a single question and then asked to respond via text message or smartphone. Results will be collected via the internet and discussed instantaneously with an interesting twist.

**Speaker:**Milo Schield (Augsburg University)**Topic:**USCOTS Theater-Teaching Activity: Where do Statistics come from?

This activity incorporates audience involvement to compare related rates or percentages. Central question: Where do statistics come from; how are they assembled; what difference do their definitions make in their size or in the size of their association?

**Data Dessert:**Internet flavor (Julie Legler)**Journal Club:**Moore, D. (1995) The craft of teaching.*MAA Focus, 15*(2), 5–8.**Speaker:**Andy Zieffler (University of Minnesota, Educational Psychology)**Talk:**A Brief Illustrated History of Statistics Education

**Journal Club:**Gould, R. (2004). Variability: One statistician’s view.*Statistics Education Research Journal, 3*(2), 7–16.**Speaker:**George Cobb (Mount Holyoke)**Talk:**Drug Deals and Jury Wheels: Probability and Statistics Go to Court

In the case of US v. Shine et al., federal prosecutors in Burlington, Vermont wanted to send to jail a dozen accused drug dealers, all of them members of racial minorities. The defense team claimed that the citizens chosen to serve on the grand juries that had indicted the dozen accused dealers had been chosen using a jury selection method that was racially biased. If true, the legal rights of the accused under the US Constitution would have been violated, which would mean that they should be set free. The case involved a variety of issues, all statistical or mathematical, and so I was asked by the prosecutors to testify about these issues. In my talk I’ll describe one particularly simple mathematical problem that figured prominently in the case, and illustrate two different ways to think about it. For me, these two ways exemplify two very general strategies for solving mathematical problems.

**Short Report:**Project SENCER (Cindy Kaus; Metro State)**Journal Club:**Brown, E., & Kass, R. (2009). What is statistics?.*The American Statistician 63*(2), 105–110.**Speaker:**Victor Addona (Macalester College)**Talk:**A Course on Statistical Analysis of Sports and Games

Victor will describe the course he is teaching this semester to first-year students. Here is the course description:

In this course, we learn about the core descriptive, probabilistic, and inferential methods used in statistics. These topics will be motivated by using real data from a variety of sports. Quantitative analysis of sports data has become a serious research field. Baseball was the first North American sport to be studied by statisticians. The term coined by Bill James for this field, “sabermetrics”, is widely recognized in the statistical community, and many professional sports teams now employ academic statisticians to help them gain an advantage over the competition. The acceptance of sabermetrics (aided by the popularity of books like Moneyball) has led to quantitative research in other sports, as objective decision making continues to replace the haphazard “gut feelings” used in the past. The plethora of examples that are sports related are not restricted to “traditional” sports data. For example, we will discuss data on the graduation rates, and birthdays(!), of athletes. As we will see, at the core of this course is a desire to answer questions of interest in impartial and meaningful ways. Ultimately, this is a statistics course, so we learn about different methods used to analyze data. But we will not lose sight of the fact that, in statistics, the final analysis is only one aspect of the process used to answer questions. For example, measurement issues are a crucial, but often overlooked, piece of the puzzle: What metric(s) should we use to decide which of two players is better? How do we assess strategies in a game?

**Resources on the Web:**Statistics Online Computational Resource (Bob delMas; University of Minnesota, Educational Psychology)**Journal Club:**Cobb, G. W., & Moore, D. S. (1997). Mathematics, statistics, and teaching.*American Mathematical Monthly, 104*(9), 801–823.**Speaker:**Danny Kaplan (Macalester College)**Talk:**Hypothesis Testing: What’s the Alternative?

I know, you’re thinking this is going to be a session about Bayesian inference. But it’s just good ol’ hypothesis testing. In looking through standard statistics texts and comparing what they say to what I do in my introductory classes, I’ve been struck by the very limited role that the alternative hypothesis plays. The alternative seems to be entirely about whether to do a one-sided or two-sided test, and is written in very abstract terms, e.g., Ha : p > 0.5. In my classes, I have the students work with very specific alternative hypotheses, e.g., p = 0.70. This has several advantages. It makes it clearer to the student that both the null and the alternative are ‘hypotheticals’, statements about world that might or might not be true. It also relates statistical practice more closely to scientific practice, where a person designing a study has to think about how big is the effect he or she is looking for. This leads directly to calculations of sample size and power. I’ll show some settings in which I have students do these calculations.

**Journal Club:**Lovett, M., & Greenhouse, J. (2000). Applying cognitive theory to statistics instruction.*The American Statistician, 54*(3), 1–11.**Speaker:**Sandra J. Moen (3M)**Talk:**Teaching Experimental Design with a Paper Helicopter

To paraphrase George Box, “All toys are fun; some toys are useful.” The speaker “played” with helicopters in her graduate-school days when she worked for Prof. Box as a graduate assistant one year, and since then she has used them in statistics classes at 3M with good success with both technically-minded and not-so-technically-minded students. Drawing on Box’s article from 1991, Teaching Engineers Experimental Design with a Paper Helicopter, there will be a brief introduction to the problem, presentation of a few ideas of how to incorporate it as an exercise in class to illustrate different types of statistical tools and questions, and then it’s flight time! Try it out for yourself.

**Journal Club:**Sotos, A. E. C., Vanhoof, S., Van Den Noortgate, W., & Onghene, P. (2009). The transitivity misconception of Pearson’s correlation coefficient.*Statistics Education Research Journal, 8*(2), 33–55.**Speaker:**Paul Alper (University of St. Thomas)**Talk:**Hypothesis Testing: What’s the (Other) Alternative

Drawing on a concrete example from the paper by Michael Lavine, What is Bayesian Statistics and Why Everything Else is Wrong, Paul will review the areas of conflict and alignment between the Bayesian and hypothesis-testing paradigms.

**Journal Club:**- Pfaff, T., & Weinberg, A. (2009). Do hands-on activities increase student understanding?: A case study.
*Journal of Statistics Education, 17*(3). - Weltman, D., & Whiteside, M. (2010). Comparing the effectiveness of traditional and active learning methods in business statistics: Convergence to the mean.
*Journal of Statistics Education, 18*(1).

- Pfaff, T., & Weinberg, A. (2009). Do hands-on activities increase student understanding?: A case study.
**Internet Resources:**Life on an Island - A Simulated Population for Learning Statistical Reasoning (Michael Bulmer; University of Queensland, Australia)**Speaker:**Michelle Everson (University of Minnesota, Educational Psychology)**Talk:**Creating Web Applets from In-Class Activities to Foster Active Learning in the Online Statistics Classroom

In a classroom setting, students can engage in hands-on activities in order to better understand certain concepts and ideas. Replicating hands-on activities in an online environment, however, can be a challenge for instructors. Recently, at the University of Minnesota, we created two web applets to incorporate into our undergraduate online introductory statistics courses, and these applets are meant to provide opportunities for students to engage in the kinds of activities commonly used in comparable classroom sections. One applet goes along with an activity where students use Post-it Notes in an effort to develop a more conceptual understanding of the mean and the median, and the other applet is meant to replicate aspects of the famous “Gummy Bears in Space” activity (presented in Schaeffer, Gnanadesikan, Watkins & Witmer, 1996). This talk will focus on the development of these applets and the ways in which the applets are currently used in our online courses.