View Only

Statistics and data science courses have often been described as full of student engagement opportunities. As the information age advances, data, graphs, and statistics have become an everyday staple on every Internet and social media platform available. Our students need to be able to understand and discuss this statistical information in appropriate ways so they can make informed decisions and develop informed opinions about various issues. I believe our grand goal is always to develop more statistically literate citizens.

*IMPACT* (2018) guides us in the implementation of student-centered learning environments that engage students:

“Students should interact with each other often through meaningful discourse and collaborative activities for the purpose of sharing and refining ideas.

- Students should develop as mathematical thinkers by engaging in inquiry-based learning through exploration, conjecturing, questioning, sense making, and seeking alternate solution paths.
- Students should be provided opportunities to make mistakes and collectively learn from them.
- Students should work in a physical setting that promotes teamwork, builds respect for one another’s ideas, and critique the thinking of others.
- Students should work with appropriate tools to expedite computations and symbolic manipulations, but also to formulate hypotheses, test conclusions, and validate their thinking” (p. 46).

I would like to briefly share my own thoughts, with the goal of sparking some dialog around these strategies in statistics and data science courses.

**“Students should develop as [statistical] thinkers by engaging in inquiry-based learning…”**

In statistics and data science, it is possible (and some say necessary) to emphasize for our students the investigative process when working with data (see GAISE, 2016). Asking questions, seeking information, evaluating that information, and presenting solutions are valuable opportunities for students to engage: not only with the course content, but with each other and with their instructor as well. Data analysis projects often involve teams of individuals, and creating this collaborative environment in our courses is an important consideration.

*Question: How can we help instructors find the best scenarios for inquiry-based learning?*

**“Students should be provided opportunities to make mistakes …”**

Exploration, by nature, involves choosing a path and seeing where it leads. Sometimes, it turns out to be the wrong path. Instructors should offer students an opportunity to explore data, explore statistical information (e.g., a graph shared on the Internet), and explore their own (and others’) thinking about statistics. This means mistakes are going to happen. How can we promote a learning environment that allows for students to make mistakes and learn from them without being penalized? This question leads into a whole other area – assessment.

*Question: What assessment opportunities can we use to encourage making mistakes while also promoting appropriate interpretations and computations?*

**“Students should work in a physical setting that promotes teamwork …”**

This idea ties in nicely with the first – creating a collaborate environment that promotes teamwork and learning from and with others can lead to powerful results. All of this takes time, however, and very often instructors struggle to make large, sweeping changes to their course structure all in one term.

*Question: How can we help instructors find ways to build in collaboration and teamwork in small steps?*

**“Students should work with appropriate tools to expedite computations and symbolic manipulations, but also to formulate hypotheses, test conclusions, and validate their thinking”**

Here, we remind ourselves that there are procedures and mechanics involved in statistics and data science courses; we cannot escape that fact. However, how do we appropriately emphasize those mechanics, as well as generating correct answers, while also emphasizing what the results actually tell us about the data in question? This is an important dichotomy to explore further – “doing statistics” vs. “thinking about statistics”.

*Question: When in our statistics and data science courses can we shift the computations to technology, and when do we (if at all) need to give students the opportunity to see how the computations work?*

#ST-Statistics

#Engagement

4 comments

37 views

Mark Earley

03-03-2021 19:13:40

I have heard good things about Kaggle Jerome. Thanks for the resource!!

I think that the shift to technology is very important. I think that for an Elementary Statistics course, conceptual understanding is the key. If it were an Intermediate Statistics course or a Mathematical Statistics course, then the computations would be more important.

I think it's also important to note who is the class targeted for? Our Elementary Statistics course is not taken by math nor Computer Science majors, so typically, this is the final course that most students complete in their mathematics career.

Keisha Brown

Assistant Professor of Mathematics

Georgia State University - Perimeter College

If you are looking for activities for students, visit www.kaggle.com . There are lots of fun datasets to play with, and you can see what other users have done with them.

For the more adventurous, there are Kaggle competitions to solve a particular data science problem. After years of viewing other people's work, I took the plunge and entered my first Kaggle competition. It was humbling! My ranking is just above the median, although on a percentage basis my raw score is not that far off from the leader.

Happy data sciencing!

Jerry Tuttle

Online adjunct instructor

Rocky Mountain College of Art & Design, Denver CO