USCOTS 2013 - Breakout Session #3

The title of this breakout session is borrowed from a learning assessment retreat I am hosting on May 7, 2013, in my role as Dean of the Graduate School of Arts and Sciences at Harvard University. The retreat itself has five breakout sessions, organized by five questions:

• What Do I Want My Students to Learn?
• What's the Evidence that My Teaching is Effective?
• What Can We Learn from Student Evaluations?
• How Can We Evaluate 40,000+ students?
• How Can I Know that my Teaching Has Lasting Impacts?

These questions were designed to engage faculty, especially those who wonder whether learning assessment is merely a bureaucratic accreditation tool, an academic exercise by those who do education research for a living, or something that actually can inform and reform their pedagogical practices. This session invites the audience to discuss, debate, and delineate these five questions and their answers---and more questions---generated from the Harvard retreat.

While online resources have been gradually transforming post-secondary education, the recent advent of massive open online courses (MOOCs) has the capacity to completely and quickly transform who we teach, how we teach, and even whether we teach. I will describe how I got involved in teaching an introductory statistics MOOC and share my observations on the experience. I'll also talk about our plans for using the resources created for the MOOC in a flipped classroom model that is designed to better meet our students' interests. Participants in the session will get a taste of what it is like to be a student in our MOOC with a demonstration of the content, the assessment, and the interface. We will also take a look behind the curtain to experience what it is like to be an instructor for the course by planning part of a lecture and by engaging with the students in an online forum. And we'll brainstorm about ways MOOCs might be used to help the students we teach at our home institutions.

Popular methods of assessing readiness to teach K-12 mathematics and statistics include using measures of undergraduate pre-service teacher content knowledge, such as course grades and Praxis exams, and the completion of some student teaching experience. However, other methods of assessment exist that speak to the readiness of undergraduate pre-service teachers in terms of their ability to develop statistical thinking in their students. This session will present methods of assessing undergraduate pre-service K-12 mathematics/statistics teachers for statistical content knowledge (such as the Comprehensive Assessment of Outcomes in a First Statistics course, and the Data, Probability, and Statistics module for the Learning Mathematics for Teaching project), teacher attitudes, efficacy and beliefs (such as the Survey of Attitudes Towards Statistics, Self- Efficacy to Learn Statistics, Current Statistics Self-Efficacy, Self-Efficacy to Teach Statistics, and the Statistics Teaching Inventory), and classroom assessments in teacher preparation courses (such as the Japanese Lesson study and using writing prompts). These assessment tools will be discussed in context of changing assessment in individual undergraduate pre-service teacher education courses as well as at the program level. This session is the last of a three-part session on Preparing to Teach K-12 Statistics.

Deliberate practice and team-based learning are described in a recent article:

"...deliberate practice, which inverts the traditional university model. Class time is spent on problem-solving, discussion and group work, while the absorption of facts and formulae is left for homework. Students read assignments before class. In class students work in small groups, discussing specific problems, with the teacher roaming between groups to offer advice and respond to questions." (The Economist, "Applying Science to the Teaching of Science," May 12th, 2011)

Instead of the traditional lecture where facts are delivered to students in class, in this approach students learn the facts ahead of time (via readings, videos, exercises, etc.), and class time is devoted to applying content knowledge via team exercises involving real data and interdisciplinary examples.

During the session, participants will experience first-hand the flipped classroom and team-based learning. There will be no "lecture," but rather a sequence of examples used in our classes to demonstrate this pedagogical approach. Our goal is to convey why and how teaching introductory statistics courses using this method is beneficial in building students' skills in data analysis, modeling, and analytic thinking as well as addressing differences in learning styles. In addition to providing examples, we will also discuss tips for designing activities that intellectually engage students from a wide variety of backgrounds and preparedness levels, forming effective teams where the most prepared and the least prepared students equally benefit from the experience, preparing materials to help students with self study, and promote effective team dynamics. We will also share positive and constructive feedback from students. Materials required for implementing these activities (data sets, handouts, etc.), as well as sample syllabi, will be made available in order to help interested participants implement this pedagogy as well as the specific activities in their classes.

The use of R in statistics teaching and learning offers many opportunities. It can help empower students for future use of statistics by providing a free and extensible platform for statistical computing. It can support a randomization-based curriculum for which many traditional packages were not designed. RStudio and R Markdown can help students communicate their statistical insights.

However, adopting R may be an intimidating change for both faculty and students. In this session we will invite participants to share thoughts, experiences and concerns about using R in an introductory course, identify key questions that need to be addressed when considering and managing a change to R, and identify particular R language challenges and ways that these could be addressed. Participants will engage in a supportive network of fellow educators and take away key points to promote discussions at their home institutions to help make change happen.

The movement to incorporate active learning and real world data within the statistics classroom has dramatically improved many courses. However, when active learning involves a student's psychological investment, research has shown it deepens the level of student engagement. Students feel a part of the learning process which promotes higher understanding and a more enjoyable experience.

We will provide resources and examples that motivate students to become psychologically invested by taking ownership of their data. Ownership comes when students are allowed to make their own decisions about what is important in a study, take action based upon those decisions, and then defend their decisions against their peers. When students have input into the research process and the outcome is not known a priori to either the students or the instructors, the study becomes real to the students in very new ways. This greater level of investment encourages greater student learning and a passion for knowing how to find a solution to their study.

Participants will see 1-2 day activities as well as semester long projects that can be easily integrated into a first or second algebra-based statistics course. These activities bridge the gap from smaller, focused textbook problems questions towards the open-ended nature of messy, real-world problems. Conceptual understanding is emphasized through guided assignments which take the student through the entire process of a statistical study. This includes transitioning from a research question to a statistical model, properly collecting and cleaning data, appropriate model building and assessment, as well as effectively communicating their results.

The importance of statistics and increasing demand for a statistically literate citizenry have prompted researchers to study students' struggles with the topic. However, researchers have largely overlooked the knowledge of those teaching introductory statistics courses. Many instructors at two-year college are faced with the challenge to teach a subject outside of their realm of expertise: only 2% of full-time instructors and 2% of part-time instructors hold a degree in statistics. Therefore, this proposed section aims to create an open discussion on the topic of variability and will employ a set of questions that are known to produce different types of responses as a springboard. This section will be based on hands on activities, modeling new methods of approaching the concept of variability that instructors can implement in their classroom. Discussion will be carried out in an atmosphere where participants will feel safe making mistakes and asking questions without feeling judged or inadequate. Participants will receive detailed handouts regarding the activities that will be performed during the section so that they can replicate their experience with their students. The variety of questions will help participants see variability in several contexts making them aware of the essential role of variability in statistics.

Regardless of how much we may want to change, we often get stuck in the process and our great ideas and good intentions fall short. This presentation approaches change as a step-by-step process and illustrates how statistics course changes can be planned, implemented, and evaluated using this change process. Statistics courses will be used as examples, but the focus of the presentation will be making changes. Participants will be guided through the process of change - come with your ideas and we will help you make them happen.