USCOTS 2011 - Breakout Sessions

Breakout Session #1
Friday, 11:00 am - 12:20 pm

  1. Allan Rossman, Cal Poly (moderator):
    Panel Discussion on Randomization Methods with Robin Lock, St. Lawrence University; Rob Gould, UCLA; Roger Woodard, North Carolina State University; Jeff Hamrick, Rhodes College; Nathan Tintle, Hope College; and Beth Chance, Cal Poly
  2. Roxy Peck, Cal Poly:
    What is the Next BIG Thing and Will We be Ready for It?
    The theme of USCOTS 2011 is "The Next Big Thing." This session will begin by taking a look at the "last big things" that have already changed what we teach and how we teach. Then, in the spirit of the conference theme, the session will engage participants in a discussion of what the next BIG things will be and what we should be doing to prepare for them. If you choose this session, please come prepared to share your ideas. We will identify a few potential next big things, and then participants will have the opportunity to be part of an informal debate to argue for or against one of the potential next big things identified.
  3. Andy Zieffler & Bob delMas, University of Minnesota; Randy Groth, Salisbury University; and Nyaradzo Mvududu, Seattle Pacific University:
    Introduction to Qualitative Research Methods in Statistics Education Research
    The report titled Using Statistics Effectively in Mathematics Education Research by the Working Group on Statistics in Mathematics Education Research (2007) points out that "good quantitative studies generally require a qualitative rationale" (p. 43). Statistics education research is arguably at a young stage of research where generative studies based on qualitative studies are needed. While there are examples of qualitative statistics education research, studies based on quantitative or mixed methods are the most common. This lack of qualitative research may be due to a lack of training in qualitative methods among statisticians and statistics educators. This session will provide an overview of qualitative data collection methods and actively introduce the audience to some of these methods. Participants will work in small groups to create good questions and probes to gather student data via interviews or open ended questions. Methods of collecting data from classroom observation, student interviews, small group problem solving taks, and written responses will be compared and contrasted by participants in small and large group activities and discussions.
  4. Larry Lesser, UTEP and Martha Aliaga, ASA:
    Engaging Student Diversities in a BIG Way
  5. Daren Starnes, The Lawrenceville School:
    Cultivating Statistically Savvy High School Grads - AP Stats, Nspire CX, and Common Core

    From 1997 to 2011, over one million high school students took the AP Statistics exam. An unknown number of additional students completed the AP course but didn't take the test. Over that same time period, an increasing number of high schools offered semester or yearlong introductory statistics courses. With the recent adoption of the Common Core State Standards, many states have committed to including a substantial amount of statistics and probability in their secondary school mathematics curricula. There are even three standards devoted to simulation-based inference! Increased access to dynamic technology, like the newly released TI-Nspire CX, gives students an opportunity to engage statistical ideas in sophisticated ways without getting bogged down in computation. As the headline of the Lawrenceville School newspaper, The Lawrence, confirmed following a well-received talk from distinguished alumnus Dick DeVeaux, "Statistics is the New Sexy."

    With increased attention on statistics and probability in secondary schools, we should expect 21st Century high school graduates to be more statistically savvy. But what does that mean, exactly? What statistical understandings can we expect these students to bring with them to college? In this session, we'll search for some possible answers.

  6. Camille Fairborn, Utah State University and David Zeitler, Grand Valley State University:
    Making the Transition to Online Teaching
    Use of online or cloud technology is no longer unusual, it is rapidly becoming the norm even for technophobic statistics instructors (almost everyone uses a learning management system today). The pressure for this change is coming both from social pressures to provide flexibility for students as well as from the changing students themselves. In this session we will guide participants in exploring how they can transition to the new reality of teaching statistics using online social and collaborative technologies. We will consider what must change, what can be retained and what needs to be left behind.


Breakout Session #2
Friday, 3:00 - 4:20 pm

  1. Robert Gould, UCLA:
    Data in the Classroom - post-plenary discussion
  2. Mark Glickman, Boston University and Larry Lesser, UTEP:
    USCOTS Theater - show off your talents, try out your ideas and get feedback. Contributors also include Marc Issacson, Augsburg College; Nancy Pfenning, University of Pittsburgh; Milo Schield, Augsburg College; and Gunnar Stefansson, University of Iceland.

    Qualitative data is a backbone of education research. However, statisticians are rarely taught how to analyze qualitative data. In this session, we will share interview data from an ongoing study on student conceptions of sampling distributions. We will introduce two techniques for analyzing interview data: generative coding and framework-based coding. Participants will view video excerpts and transcriptions, develop coding categories and manuals, and discuss the difficulties involved with the research process.
  3. Stacey Hancock, Clark University; Jennifer Noll, Portland State University; Sean Simpson, Westchester Community College; Aaron Weinberg, Ithaca College; and Marsha Lovett, Carnegie Mellon Uniiversity:
    Making Sense of Qualitative Data

    Qualitative data is a backbone of education research. However, statisticians are rarely taught how to analyze qualitative data. In this session, we will share interview data from an ongoing study on student conceptions of sampling distributions. We will introduce two techniques for analyzing interview data: generative coding and framework-based coding. Participants will view video excerpts and transcriptions, develop coding categories and manuals, and discuss the difficulties involved with the research process.
  4. William Duckworth, Creighton University and Amy Froelich, Iowa State University:
    Using the new JMP Concept Discovery Module for Regression to Build Conceptual Understanding in Introductory Statistics

    Simulation is an important tool in teaching topics related to sampling distributions and inference in the introductory statistics class. Many of these simulations have been developed with Java applets and made available on the web. While these applets are easy to use and readily available to statistics instructors, they usually do not match classroom, laboratory or homework activities, are generally different than software used for data analysis, and are often limited to basic descriptive and inferential topics. As a result, students can struggle with the transition between classroom activities and assignments, data analysis and computer simulation activities. In this session, we will lead the participants through classroom lessons, homework or laboratory activities that showcase the descriptive and inferential capabilities of the new JMP Concept Discovery Module for Regression. The new module goes well beyond the offerings of current java applets providing a unique teaching and learning opportunity!
  5. Robin Lock and Patti Frazer Lock, St. Lawrence University:
    Using Bootstrapping and Randomization to Introduce Statistical Inference

    At previous USCOTS, George Cobb argued for moving away from using normal and t-based procedures as the core for teaching statistical inference and towards the use of computer-intensive methods such as bootstrapping and randomization tests. This raises a number of important questions.

  • What are these methods and can they be taught in an introductory course?
  • How can I integrate these ideas into my existing curricula (or do I need a complete overhaul)?
  • What activities help engage students and build understanding of these methods?
  • What technologies can make these methods accessible to my students?
  • To what extent should the simulation methods be tied to how the data were collected?
  • How do I go about assessing student understanding of ideas taught with simulation methods?
  • What are some challenges, pitfalls, and benefits from using these methods in practice?
  • Come share your opinions and get new ideas!
  1. Audbjorg Bjornsdottir, University of Minnesota and Ellen Gundlach, Purdue University:
    Rethinking Assessment in the Digital Age

    One concern that is often voiced about the online course format is that students may cheat in an unmonitored setting. There are many ways instructors can handle the issue of possible academic dishonesty in the online environment, but perhaps we should be re-thinking just what it means to assess our students now that so many of them might be working outside the confines of the traditional classroom? We believe this session would be of interest to anyone who might teach online or who might teach a traditional course with online components (like online homework systems or online exams).


Breakout Session #3
Saturday, 11:00 am - 12:20 pm

  1. Dick Scheaffer, University of Florida and Christine Franklin, University of Georgia:
    High School Changing Curriculum's Impact on Collegiate Introductory Statistics Courses

    The Common Core State Standards Initiative is a state-led effort coordinated by the National Governors Association Center for Best Practices (NGA Center) and the Council of Chief State School Officers (CCSSO). These standards have been adopted by 41 states (March 31, 2011). Within those standards is a strand on probability and statistics. By the time high school graduates arrive in college, they will have been exposed to basic descriptive statistics, the difference between observational studies and experiments, probability and how to use likelihood to make decisions, understand generalizing to population and an intuitive idea of margin of error. Presenters will introduce details about the Core Curriculum and lead a discussion on their impact on the Introductory College Statistics course.
  2. Nathan Tintle, Hope College and Paul Roback, St. Olaf College:
    Is your first course preparing students for the next BIG thing? Strategies for a second course in applied statistics.

    What opportunities exist in your department for the fast growing number of students who take a college introductory statistics course or Advanced Placement statistics? Are you finding that statistical methods used by client disciplines or in discipline-specific journals are containing more sophisticated methods not fully covered in a first course in statistics? Numeric growth combined with changing methodology used by client disciplines are two of the strongest cases for implementing a second course in statistics at your institution. In this session you will have the opportunity to (1) hear from individuals who are successfully implementing a second course at their institutions, (2) see and experience second course activities, and (3) discuss issues related to the second course with likeminded individuals in round-table discussions focused on topics of particular interest to you. The goals of this breakout session are to make you aware of developments and issues involved with implementing a second course at your institution, expose you to materials and approaches that you can implement immediately, and connect you with like-minded individuals to discuss specific issues related to the second course.
  3. Sterling Hilton, BYU; Hollylynne Stohl Lee, NCSU; and Felicity Enders, Mayo Clinic:
    A Conceptual Framework for Statistics Education Reseach

    Combining the strengths of several academic disciplines, statistics education research has emerged as a field of study concerned with issues relating to the teaching and learning of statistics at all levels and in a variety of contexts. In this session we will propose a conceptual framework for statistics education to enable researchers to see the whole field while focusing on a particular area. Participants will contribute their areas of interest in statistics education research and major focus of research. Taking these ideas, session leaders will categorize them and dynamically build a connected framework to illustrate how the areas of interest fit together. Given some broad categories, we will highlight progress in the field of statistics education research in these areas and discuss the inter-related nature of these areas. Participants will then engage in small groups to reflect on their own contexts and experiences and work together to identify research domains that can drive future endeavors in statistics education research.
  4. Wayne Stewart, Unviersity of Auckland, New Zealand:
    A Tactile approach to Markov Chain Monte Carlo
  5. Michelle Everson, University of Minnesota and Jackie Miller, The Ohio State University:
    Using Social Media and Technology to Engage the Next Generation of Statistics Students

    Many students now have Facebook or Twitter accounts, and most students know something about YouTube. Given the popularity of social media, how can we best use this--along with other newer technologies (e.g., clickers , "Poll Everywhere," Prezi presentations, etc.)--to engage and motivate our students? This session will focus on innovative uses of social media and technology. Participants in this session will learn about different tools and have opportunities to brainstorm ways of using such tools for educational purposes. They will also have opportunities to use different types of social media and technology in creative ways DURING the session.
  6. Kari Lock, Harvard University; Eric Lock, University of North Carolina; Dennis Lock, Iowa State University; Robin Lock and Patti Frazer Lock, St. Lawrence University:
    Technology for Teaching Bootstraps and Randomizations

    One of the key issues in teaching with computer-intensive methods is to find appropriate technology to make the ideas and techniques accessible to students. This session provides opportunities to get hands-on experience with a number of different software environments for creating and using bootstrap and randomization distributions. These include traditional statistical software packages (such as Minitab, R, JMP, and SAS), teaching-centered packages (such as Fathom), general purpose software (such as Excel) and online applets. We choose a couple of specific examples for doing a bootstrap confidence interval or randomization test and provide documentation, written for students, to accomplish the required simulations in various software environments (along with computer files where needed). Participants will be able to work individually or in small groups on the implementations that are most appropriate for their current teaching situations.

    Bring your own laptop to try the examples with familiar software - or try out a new environment using one of the computers provided for this session.