Statistics Narrative

In the time between the 12 May and 17 May meetings, Linda asked the group to contemplate a narrative about what the experience would be like for our individual fields. I attempted to draft a narrative, but it only yielded a silly tale which leaped around and between class functionality to my/our own preparation. Linda shared her narrative at the 17 May meeting. It appeals to me because much of the process is very similar to my own preparation process. I am eager to see a graphic of her narrative for future visual reference and recollection. From the perspective of the statistics partner, the only preparatory consideration I must mention are the projects. Before the experience begins, the statistics partner must be familiar with the projects, their individual goals, the individual data, and similarities/differences in and between the various data. Ideally, much of the data will have already been collected or will be collected in the first few weeks of the term.

The statistics learning objectives will be best experienced in the second term, an ideal term to support the collaborative effort. The projects will form the basis for student engagement. Having statistics in the second term will allow time for the projects to mature to include data collections. It is possible that not every project will have traditional data or practical data analysis needs to be successful. However, this may result in students collaborating on multiple projects (perhaps good), projects which are too complex, and/or not ideally aligned with the statistics learning objectives.

When the term begins, each student will be given a syllabus for their respective course, which includes the learning objectives and a tentative outline of when learning objectives and/or topics need to be attained. I will show all of the students all of the learning objectives from each course, so that they can see the overlap in learning objectives and topics. I will identify learning objectives which are similar and capable of attainment through projects collectively, learning objectives which cannot be taught through projects, and learning objectives which are specific to their course. Learning objectives in the last two categories are likely to be attained through supplemental online tools, suggested readings, short lectures, and small group discussions as needed.

There are no impossibilities about the structure of student contact time. This could range from traditional/structured lectures, to days/times devoted to small group interactions specific to individual projects/groups. I envision a hybrid approach where all students are present. Students will arrive having done some independent learning, which may not entirely be connected to their project. The independent learning tools will have self-assessment exercises and, perhaps, a formal graded quiz. Students will have to think through and develop questions related to their project. The students will bring their data and questions to the 'class room' which ought to be a computer lab (eg the statistics studio 02-206).

We begin with a real question: 'student X wants to use his/her data to answer question X1.' Connections across the projects will further develop, if not already, when student Y recognizes similarities and offers: 'student X's question X1 is just like my question Y1.' Then, we will talk about about what statistics are appropriate, what are inappropriate, a little theory, some assumptions, software use, and examples in the context of the student projects. The contact time should utilize the statistics partner as a coach or guide, devoting ample time to students working in teams to analyze project data.

I have written this narrative with lots of possibilities, but there are some conceptual and practical impossibilities to consider.

1. Projects

The projects must be well organized, and have a minimal set of diverse data types (qualitative and quantitative variables). Students with projects that do not have this will need to 'partner' with other groups. As a consequence, students may feel like they are working on multiple projects - which isn't necessarily a negative. Alternatively, related data could be identified and utilized.

In general, students will have to accept that concepts learned though projects will occur gradually, somewhat sequentially, and over time. Students will not obtain an ideal, global data analysis rapidly or at the beginning of the experience.

2. Grading

To have support from the senior statistics faculty, traditional grading will have to be used to determine official university grades. This may be some combination of online quizzes, project related homework, a project report which demonstrates mastery of the statistics learning objectives (likely part of a larger report), at least one midterm, and a final.