Assignment 12: Redesigning Seminar

Posted on

Considering the pros and cons of each of the three options for the course structure, I think that the last option, which is to expand the course into a larger one-credit course, is the best option if the instructor is to redesign this course. Even though with the current structure, students have a chance to work with the community partners throughout the whole semester, and have a lot of time to get to know them as well as their data visualization goals well, the class only meets for one hour per week, which somewhat limits the amount of material that can be covered. Moreover, since students started working with community partners from the beginning of the semester and did not have any prior knowledge about data visualization, for most of them, the first half of the semester was mostly about analyzing data and talking with community partners about prioritizing ideas. Although these are all very important factors in the process of creating data visualizations, they are not the main focuses of the course. I think students could have spent much more time on the main focus of the course, which is data visualization, if they already have sufficient knowledge on visualizing data by the time they start working with their community partners.

If the course is expanded into an one-credit course and divided into two parts, with more data visualization content and coding instruction during the first half, and students working on community partner projects during the second half, I think both the students and community partners will make use of their time much more efficiently. Students start their work with their partners after already being introduced to the concepts and techniques of data visualization, so they would have much better and clearer ideas when discussing with their partners on how to achieve their goals with the data visualization tools. Moreover, with more data visualization contents and coding instructions introduced, this course is also a great option for students who are interested in the course material to fulfill the Numerical & Symbolic reasoning requirement. The only downside I see in this structure is that since the time students and community partners have to work together is shortened, they must find ways to make the best use of it. Therefore, it might help a lot if before working with the students, community partners have already prioritized their goals and ideas, had the data ready to use, and done some preliminary analysis if possible.

The first option, which is to insert data visualization as a 3-week module into a topical seminar, does not sound like an appealing option for me. Even though it is very helpful for students who take a topical seminar and want to gain additional knowledge on data visualization, I think to be able to effectively use data visualization tools, a 3-week period is too short to introduce all of the materials necessary. Incorporating data visualization materials into a topical seminar is a great option to get some introduction on data visualization, but a more focused, intensive course like in the third option should be offered as well for students who want to go more in-depth into this subject.

Lying with Maps

Posted on

Both of the following maps are based on the same data of employment rates across all census tracts in Connecticut. However, due to the number of intervals that the data is divided into in each map, the two maps look very different from each other. In the first map, the data is divided into only 2 buckets, showing a clear distinction between the census tracts with high employment rates (64.9% – 79.6%) and the ones with low employment rates (less than 64.9%). Highly contrasted colors are also used to emphasize the difference between the 2 baskets.

On the other hand, the data is divided into 8 buckets in the second map. Colors of the same hue and slightly different shades are used in symbolizing the baskets. It seems like the employment rates across the census tracts in Connecticut are quite even, with only a few census tracts having slightly higher rates of employment than the others.

The above maps have demonstrated how mapmakers can lie with maps and give totally opposite impressions to viewers, just by changing some of their decisions on map elements in making maps.