��ƷSM����ӰƬ

The problem that I wish to address concerns trying to encourage a substantive conversation with a large class regarding the problems and issues associated with data analysis. Although I’ve been successful in creating active learning assignments where students work on analyzing a problem, I’ve been less successful in collecting their work and using it to spark discussion in the class. I want to learn how to use technology to help generate a more fruitful conversation.

The title of the course is “Introduction to Research Methods”. It is an introductory statistics course with approximately 180 students (3 TAs). The class is required of all Political Science majors. Out of 180 students, approximately 100 are PSCI majors, 60 are business majors, and the rest come from ECON, SOCI, and IAFS.

The implication of not addressing the problem is a huge wasted opportunity for the students to learn important concepts from the assignments they’re given.

The goal is threefold. First, I want to be able to hold the students accountable for the work they’re doing in class. Although the TAs and and float throughout the class as they’re working on an assignment, our eyes can’t be everywhere at once. Second, I hope by getting students involved in a discussion, the level of interest in the material will climb. My sense is that if they’ve invested in the work and have made an interesting discovery, sharing that discovery with the class will excite, motivate, and reinforce the work they’re doing. Finally, I want the discussion to turn toward issues that are central to the class. Wrestling with those problems in an assignment and sharing that with others will help provide concrete examples of how these problems come up when doing real data analysis.

In each assignment there are certain skills and concepts they have to master in order to complete in the assignment. In one sense there’s simply a test of their ability to execute code in a statistical package. In another sense, to complete the assignment satisfactorily, the student must understand the concept that is underlying the exercise, whether it is data description, regression analysis, regression diagnostics, or logistic regression.

The evidence that I’ve succeeded in sparking a discussion will be whether students become engaged in the process, actively participate and argue with one another. The other marker will
be that the deeper issues involved in the exercise will come front and center.

I wanted to get a large class on data analysis engaged in a discussion as students went through various projects that ask them to do data analysis. The class is an introductory statistics course with roughly 180 people enrolled. I have students in groups (around 40 groups). The students are asked each week to tackle a different problem. I wanted a good way for students to communicate with each other and me regarding their findings. So, I needed a means for them to present their findings in a quick and easy way that would allow me to then easily show the rest of the class.

My larger goal is to use this exercise to get students to critically think about the data and see how it relates to arguments they make all the time with little evidence.

I used a program called ‘Padlet’ for students to post to instead of the clunky D2L discussion list. I was worried that they would fill up the screen with their entries and that it would be difficult for me to spot the different contributions. I was also worried that Padlet would not really allow me to show (blow up) things nicely and that it would not really provide a good way to spark conversations. I would set up a discussion board before class and insert into D2L where they could drag and drop their work, identifying the group and including some analysis. Padlet helped tremendously as it provided a very easy way for each group to see what others were doing and to compare their work with others. In terms of the interface, Padlet was extremely easy to use and organize (it really required little if any preparation on my part and on the students’ as well).

In short, Padlet allowed there to be an electronic conversation or presentation of everyone’s work. Showing how different groups attacked the problem. It set the stage for me to be able to discuss people’s work and perhaps to identify important problems what were introduced in the assignment. In terms of learning goals, it helped lay a foundation for discussing the problem and getting at the various challenges students confronted in the assignment each week. Perhaps the only pitfall of the program was that it didn’t automatically spark conversation. While it did make it easier for students to instantaneously share their work with the rest of the class and for us to identify it in class, I realized two important things: 1) I still needed additional time to analyze what the students had done and 2) I needed to do a better job of framing the assignments so that students would be cognizant of the deeper issues that they were dealing with like measurement, causation, etc.