About the Theme
This year's conference theme "SHOW ME THE DATA" is intended to encompass the many aspects of teaching statistics, including but certainly not limited to:
- Helping students to recognize that data beat anecdotes and are essential for evidence-based decision-making;
- Emphasizing data visualization in statistics courses;
- Embracing ideas of data science in teaching statistics;
- Using data as educators to make informed decisions related to effective teaching and learning
Special Features of USCOTS 2017
- Plenary sessions by national leaders in Undergraduate Statistics Education on current and emerging trends in research, curriculum, pedagogy, and resources
- Hands-on breakout sessions to incorporate new ideas into your courses
- Posters & Beyond sessions exchanging your ideas on teaching and learning statistics
- Conference banquet
- Opportunities to meet other statistics teachers from a wide range of institutions and disciplines
- Pre-conference workshops to delve deeply into a specific statistics education innovation
- A fun, active atmosphere where everyone can be involved!
USCOTS 2017 - Workshops
Monday, May 15, 2017 - 1:00pm to Thursday, May 18, 2017 - 12:00pm
W01: Research On Statistics Attitudes
With Marjorie Bond (Monmouth College) and Alana Unfried (Cal State Monterey Bay)
This three-day workshop will continue the work started at a workshop in July 2016, which had as its main purpose to develop a strategic plan to answer the four research priorities set forth in the 2012 report title Connecting Research to Practice in a Culture of Assessment for Introductory College-level Statistics, Section 2, Affective Constructs (Pearl, et al., 2012).These four research properties are:
- How can affective constructs be accurately measured?
- How do affective constructs contribute to success in learning statistics, in either the short or long term?
- How do these affective constructs contribute to long-term engagement with statistics (e.g. statistically literate citizenship)?
- What are the important affective constructs to measure about teachers, and how do these influence teaching practice and ultimately impact student outcomes?
The term affective includes the broad areas of attitude, beliefs, emotions, motivation, etc.Researchers are building on the work of previous instruments measuring affective constructs in students and instructors to create at least two new instruments which are aligned with educational psychology theory such as Expectancy-Value Theory (EVT). During this workshop, research collected since the July 2016 workshop will be presented and discussed which will lead us to the potential items to measure the affective constructs. Plans will be made for evaluation of our items over the summer so that we have two pilot instruments ready for testing during the academic year 2017 – 2018. Additionally, we will need to create a plan for the data collection of the pilot instruments.Anyone who is interested in statistics attitude research is invited to attend the workshop whether they were at the July 2016 workshop or not; however, participants are expected to prepare for the workshop and to continue working on the project after the workshop. New researchers to the area, two-year community college instructors, graduate and undergraduate students are encouraged to join us.
Tuesday, May 16, 2017 - 1:00pm to Thursday, May 18, 2017 - 12:00pm
W02: Web Scraping and Data Visualization with Python and Tableau
With Silas Bergen and Todd Iverson (Winona State University)
In today’s data-driven society, two skills are especially important for successfully navigating the data cycle: the ability to work with and manipulate unstructured data, and the ability to effectively communicate data visually. A primary source of relevant unstructured data is the Internet, where often the data of interest are not available in a ready-to-use form. The Python programming language is used by many data scientists for web-scraping and data cleaning while Tableau is an industry standard for data visualization. In this workshop, we will provide introductory instruction in the use of both of these tools. We will illustrate use of these tools with a case-study, using Python to scrape data from a web site and prepare it for visualization in Tableau.
Wednesday, May 17, 2017 - 8:30am to Thursday, May 18, 2017 - 12:00pm
W03: A Fully Customizable Textbook for Introductory Statistics/Data Science Courses
With Chester Ismay (Reed College, Pacific University); and Albert Kim (Middlebury College)
This workshop will provide a guide to creating a user-adaptable electronic textbook incorporating data visualization, data science, and other relevant pedagogical concepts into your introductory statistics course. RStudio, Inc. has recently released the bookdown package for easily creating interactive textbook resources in a variety of formats using R Markdown. We present our own introductory statistics and data science textbook using bookdown that:
- Focuses on the entirety of the data/science pipeline
- Adheres as much as possible to Hadley Wickham’s tidyverse principles
- Blurs the line between lecture and lab with students writing in R and R Markdown for their coursework
- Uses freely available modern, rich, and complex data sources
- Leverages the mosaic package for statistical inference via resampling and simulation
- Most importantly, provides complete customizability to the instructor and reproducibility to the student
This workshop is intended for instructors with a background in R, RStudio, and R Markdown, and who are curious about incorporating the ggplot2 and dplyr packages for data visualization and manipulation into their introductory statistics courses.
W04: Teaching Introductory Statistics with Simulation-Based Inference
With Nathan Tintle (Dordt College); Beth Chance, Allan Rossman (Cal Poly - San Luis Obispo); Todd Swanson, and Jill VanderStoep (Hope College)
This workshop presents hands-on activities that introduce students to concepts of statistical inference using simulation-based methods. A related theme of the activities is emphasizing the process of statistical investigations throughout an introductory course. Topics include significance tests and confidence intervals about a single proportion, comparisons of two means and two proportions, and correlation/regression. The activities involve real data from genuine studies and make use of freely available applets. Presenters will also provide advice and lead discussions about implementing and assessing student learning with this approach.
Wednesday, May 17, 2017 - 1:00pm to Thursday, May 18, 2017 - 12:00pm
W05: Developing R Shiny Applications to Enhance Teaching and Learning
With Justin Post and Herle McGowan (North Carolina State University)
This workshop is intended to instruct on the creation R Shiny Apps - interactive web applications written in R - and on the best pedagogical practices for utilizing them in class. In particular we will focus on creating dynamic visualizations that can help students make informed decisions using data. The workshop will consist of two parts.During the first part the goals are to see current applications and activities currently being used to teach statistics, learn the basics of coding an R Shiny App, discuss design principles, and create basic applications.In the second part the goals are to create an application corresponding to a data set or topic provided by the participant, which they already use or would like to use in their classroom. We will also discuss best practices for writing corresponding activities.At the conclusion of the workshop participants will have a number of interactive example activities and their newly created application and activity. Participants should have a working knowledge of R.
W06: Visualization and Wrangling of Complex Datasets as an Entree to Statistics
With Danny Kaplan (Macalester College)
Visualization and modeling of today's data generally requires preliminary work to clean, reshape, and condense data and to bring together data from multiple sources. This workshop will present the techniques taught in Macalester College's no-prerequisite, short course: Data Computing Fundamentals. This will be very much a hands-on workshop where you will learn to use visualization software such as ggplot and data wrangling software such as dplyr. You'll also see how to introduce the topics to students. A reference for the material that will be introduced is "Data Computing" (2015) by Danny Kaplan. You'll be given a printed copy at the workshop.Pre-requisites for this workshop:
- working knowledge of R and experience with RStudio. In particular, you should know what a "data frame" is, and the use of functions.
- comfort with editing dot-R and dot-Rmd files in RStudio
- experience making simple statistical graphics in R
- a computer on which you have installed the most recent versions of RStudio and R. You will also need to install several R packages from CRAN before the workshop begins.
- You don't need to know data wrangling at any level.
- You don't need experience with *programming* in R. As you'll see, we'll have no use for loops, if statements, or even writing new functions.
W07: Challenging Introductory Statistics Students with Collaborative Data Visualization
With Lynette Hudiburgh and Lisa Werwinski (Miami University)
Data visualization is a perfect vehicle to teach students to summarize data graphically and numerically, as well as the importance of context in statistics. We believe that it is imperative to provide students in introductory statistics courses with the tools to create their own data visualizations. In so doing, we equip our students with the ability to make sense of data around them, to understand how displays can be misleading, and to become critical consumers of data and statistics. The data visualization group project, which includes the following six components: group contract, data set submission, data viz sketch/plan, rough draft, presentation, and final draft, is designed to expose students to this important field and, hopefully, to encourage them to pursue further coursework in statistics and/or data visualization.The goals for this project are:
- To promote active learning and equip students with a relevant and marketable skill: In today’s age of big data, the ability to visualize data and communicate their story is a crucial skill to possess.
- To build communication skills: Students must present and communicate their process and results.
- To facilitate team-building and group work: Students will be required to work within a team in future work environments.
- To develop critical thinking and statistical thinking; encourage students to think deeply about data and information.
- To avoid misleading visualizations: Communicate the story of the data in a truthful, insightful, and enlightening manner.
- To encourage creativity and engender appreciation for the application of statistical thinking to analyzing and exploring data.
- To engage students that may not traditionally be enthusiastic about taking a statistics course.
During the first part of the workshop, participants will be given a brief presentation outlining the underlying principles associated with data visualization and the project. Conference participants will form teams of three to four people to work through the project as students. Teams will be able to select from five data sets that we provide. Each group will share their data visualization project so that we can debrief the experience from a student perspective. The second part of the workshop will focus on pedagogical issues such as implementation, adaptations to fit different class types (face-to-face, online, etc…), scaffolding, and assessment. Past student projects will be shared. Participants will be provided with copies of all project related materials.
W15: Real world data and real world questions in the introductory statistics curriculum: a research focused, multidisciplinary project-based approach
With Lisa Dierker (Wesleyan University)
This one-day workshop will support instructors who teach an introductory statistics or quantitative research course in designing or redesigning any or all portions of their course to engage students in the rich, complicated, decision-making process of real statistical inquiry. Core features of this approach include providing opportunities for students to flexibly apply their statistical knowledge in the context of real data, the use of computing as a window to core statistical concepts, supporting students with varying levels of preparation, and attracting and inspiring students from underrepresented groups. The workshop will include very brief presentations focused on the nuts and bolts of supporting project-based experiences, followed by ample hands-on opportunities that will be supported by experienced faculty and students. Similar to the approach that will be presented; your experience in the workshop will be individualized to your own interests, background and needs. You are welcome to bring your preferred statistical software or may use the cloud-based SAS Studio, requiring only an internet browser. Supporting materials will be made available. Personal laptops are required. The workshop is intended for instructors at all levels, regardless of discipline-specific training (e.g. math, statistics, biology, political science, psychology, sociology, education, epidemiology, geology, etc.). This curricular approach has been used at many colleges and universities (e.g. Wesleyan University, University of New Mexico, Theil College, Ashesi University -Ghana, Southwestern Oklahoma State University, Virginia Tech, Appalachian State University, Concordia University - Texas, Emory College, Naugatuck Valley Community College, SUNY - Purchase) and has recently been introduced to students at the high school level at Scarsdale High School, NY and through the GEAR UP Program http://passiondrivenstatistics.com/2016/09/21/gear-up/. Supported by NSF DUE # 1323084
Thursday, May 18, 2017 - 1:00pm to 4:00pm
W08: Show Me How to Use Interactive Statistics Songs!
With Dennis Pearl (Penn State University); Larry Lesser (University of Texas at El Paso); and John Weber (Perimeter College at Georgia State University)
- Brief discussion of the use and value of fun items, and songs in particular, in teaching introductory statistics (informed by readings provided in advance).
- Build to progressively more interactive statistics songs that can be used in the classroom or online - and discover the new idea of Student-Made Interactive Learning with Educational Songs (SMILES) that uses content-relevant inputs from students to generate songs, loosely modeled after the popular word template game known as Mad Libs.
- Experience first-hand how a particular example of such a song could be used in the introductory statistics classroom.
- Whole group opportunity to discuss and critique the student experience and effectiveness for learning of an interactive song.
- Team-based opportunity to create a mini-lesson plan on different songs to share with the whole group. No one will be singled out to sing, but everyone will have the chance to have SMILES!
- Discuss initial implementation of SMILES in introductory statistics labs at Penn State and Georgia State and how to move forward given lessons learned.
This workshop supported by the NSF-funded (DUE EHR 1544426/1544237/1544243) Project SMILES for introductory statistics. Workshop participants will have the opportunity to partner (and be supported) in field trials or in developing new materials for this innovation.
W09: Implementing Specifications Grading in a Statistics Course
With Eric Reyes (Rose-Hulman Institute of Technology)
In her book “Specifications Grading: Restoring Rigor, Motivating Students, and Saving Faculty Time” Linda Nilson makes a case that the current grading systems used throughout academia are flawed and actually motivate students to produce unsatisfactory work. She then introduces what she calls specifications grading as a solution. The system grades students pass/fail on assessments which are tied directly to course objectives. The system promises to reduce faculty workloads and improve student mastery of the topics, but there are details in the execution that are not resolved fully. This workshop introduces specifications grading and tackles some of these unresolved issues with implementation in a statistics course. By the end of the workshop, participants will:
- be able to articulate the flaws in our current grading systems and the claimed benefits of specifications grading,
- be able to state the primary components of the framework,
- have written specifications for at least one assessment item and
- have made significant progress on constructing a syllabus for a course in which they hope to implement specifications grading.
In addition, it is my hope that participants will have constructed a network of colleagues for continuing the conversation regarding obstacles when applying this framework in our classrooms and approaches to overcoming them.
W10: Modules for Infusing Data Science into the Statistics Curriculum
With Adam Loy (Lawrence University)
This workshop is will explore ways to incorporate data-scientific topics into existing statistics courses, especially at the introductory level. The modules allow for the inclusion of introductory tutorials and lab activities into courses for students with limited technical backgrounds. Participants will work through class-tested tutorials and lab activities. Key topics will include data wrangling and statistical graphics. The workshop will close with a discussion related to how these materials have been used in existing courses and an overview of the supporting materials. Basic knowledge of R will be valuable, but not required.
W11: Teaching a Data-Driven Approach to Inference
With Patti Frazer Lock, Robin Lock (St. Lawrence University); and Kari Lock Morgan (Penn State University)
In many traditional approaches to teaching statistical inference, the only connection between the data and an inference procedure is through summary statistics in a computational formula that relies on some approximating distribution. These methods typically require a substantial amount of underlying background material and each parameter situation (mean, proportion, difference, correlation, etc.) needs its own specialized set of derivations. On the other hand, simulation-based inference methods are completely data-driven, relying only on the collected data to generate the distributions needed to do inference. These methods are quite intuitive, offer good visual connections to the key ideas, and can be easily adapted to different situations. They are reflected in the Common Core State Standards for mathematics, but many teachers have little background in some of the concepts, such as “develop a margin of error through the use of simulation models” or “use simulations to decide if differences between parameters are significant.” This is not surprising since these approaches are still fairly novel in undergraduate statistics courses. This workshop will discuss what these methods are, how they work, and how they can be easily incorporated into introductory classes. This session will be especially useful for current high school teachers, those involved with training future mathematics teachers, and those interested in offering intro stat courses that include some simulation-based inference.
W12: Critical Thinking with Data Visualization
With Leanna House (Virginia Tech)
Big Data are thrilling when we think about how information gained from the data may result in positive changes in industry, government, education, etc. However, raw data alone, do nothing for society. Datasets are just tables of numbers without humans to assess, process, discover, and communicate information in the data [Thomas and Cook, 2005]. That is, only when humans think critically with data can we capitalize on opportunities offered by data. Alas, the unfortunate truth is that teaching critical thinking is not necessarily a byproduct of teaching analytical methods [Jablonka, 2014]; masters of technical methods for summarizing data are not necessarily responsible, capable consumers of data. With this in mind, we at Virginia Tech bring our visual analytics research to the classroom. With the software that we developed, we synchronize teaching visual exploratory data analyses (EDA) with critical thinking. We refer to the software as Andromeda and it offers a way for analysts (e.g., students) to explore data visually and dynamically - in response to personal curiosities or feedback about the data - using multiple linear projections of high-dimensional data based on Weighted Multidimensional Scaling (WMDS) [Kruskal and Wish, 1978]. Crucially, students do not need to master WMDS to learn from high-dimensional data and create new visualizations. Rather, Andromeda enables Visual to Parametric Interaction (V2PI) [Leman et al., 2015; House et al., 2015] which means that when students interact directly with data visualizations (e.g., re-locate observations in a projection), Andromeda has the technology to interpret the interactions parametrically and update visualizations accordingly. Depending on the level of the course, students may or may not learn the details of WMDS. But, regardless of level, students have repeated opportunities to make conjectures, discover information in data, and process any implications of these discoveries. In this workshop, we will have open discussions about critical thinking, the role EDA and visualizations can play in developing critical thinkers, and the advantages interactive data visualizations may have for practicing critical thinking. We will also discuss methods for assessing 1) student critical thinking and 2) the approaches we take to foster critical thinking within EDA. This discussion will be inspired by the activities, lesson plans and recitation assignments we share, as well as the results we obtained from controlled and observational studies. Throughout the workshop, participants of the session will also use Andromeda for short exercises and we will take time to reflect on personal experiences with Andromeda.Requirement: Please bring a laptop with Chrome downloaded or pair with someone who does. Andromeda is a web-based software and we will use it during the breakout session.
W13: Training Statistics TAs to Teach for Conceptual Understanding and Foster Active Learning
With Jennifer Kaplan, Kristen Roland (University of Georgia); Roger Woodard (North Carolina State University); and Vickie Weber (Meredith College)
This workshop is designed for faculty who work with and train graduate student teaching assistants (TAs) in statistics. The overall goal of the workshop is to develop a set of evidence-based guidelines for best practices in training statistics TAs to facilitate active learning and teach for conceptual understanding. There are three main sections of the workshop:
- Discussion of the aspects of using active learning to develop conceptual understanding that are challenging for statistics TAs and the complexities of training novice instructors with subject matter knowledge typical for statistics graduate students.
- Illustration of ways to address the challenges and complexities evident in our project through the discussion of an activity that stresses conceptual understanding and videos of TA training and execution of the activity.
- Creation of a template for designing effective training sessions for statistics TAs that address the challenges and complexities identified by the workshop participants.
In first section of the workshop, the session leaders will invite participants to contribute ideas from their own experience in addition to presenting the issues that arose in our work. The second section of the workshop will be grounded in an activity developed for our project that provided rich data about the struggles statistics TAs face in teaching for conceptual understanding. We will use the activity and student responses from an open ended question to illustrate the ways we addressed the challenges and complexities we faced. In the last section of the workshop, the workshop leaders and participants will synthesize the ideas discussed in the first two sections of the workshop, leading to the production of a training template that participants can use at their own institution.
W14: Adapting and Adopting High Impact, Little Time (HILT) Activities to Clarify the Meanings of Key Words Used in Statistics
With Neal Rogness, Jackson Fox, Lori Hahn (Grand Valley State University); and Jennifer Kaplan (University of Georgia)
Participants in this workshop will join a growing community of statistics instructors who are creating, adapting, and using activities developed to help student better understand how the meanings of certain key words as used in statistics differ from their everyday meaning. These activities, known as HILT activities, are designed to have a High Impact on student learning, but require Little Time in class to implement. To date, eight HILT activities have been developed for words which include random, normal, parameter, and skew. The workshop will introduce the idea of lexical ambiguity of statistical words and acquaint attendees with HILT activities that have been developed and classroom-tested by six statistics educators whose primary teaching focus is in the introductory course. The workshop will also discuss evidence of the effectiveness for these activities using data collected from students who were exposed to the activities and data from students who were not exposed to the activities. It is hoped that, by the end of the workshop, participants will (1) identify one or more existing HILT activities to adapt for use in their fall semester introductory statistics classes; and (2) generate ideas for additional HILT activities that could be developed for other words of interest. Participants will be connected via email with the authors of the activities selected for use and the formation of an online Faculty Learning Community among the participants will be explored. Such an FLC could regularly meet via web conference to discuss usage of the HILT activities and to support one another in the development of new HILT activities.The workshop is based on the results of an NSF-funded project (NSF DUE 1504013). The importance of language in statistics and the positive effects of joining the group can best be summed up by one of the HILT Instructors involved in the project: Prior to this study, I assumed students had an understanding of the words used in statistics. This project provided evidence that I was wrong. Bringing to light the understanding of words has assisted me in being a better teacher. I am aware of the confusion words can cause because of their everyday meaning. Providing visual illustrations of the differences in the colloquial and statistical meanings has helped solidify their meanings for my students. The HILT project has not only made me a better teacher, it has sparked an interest in the words we use and how confusion arises because of the multiple definitions.