I attended two of Idan Gazit’s talks about design at last year’s PyCon, and those talks have continued to influence my thinking in a number of ways. I teach at a small high school in southeast Alaska, and this time of year we have to go through each student’s transcript if they are planning to graduate in the spring. I always dread this task because our official transcripts are ugly documents, which are difficult to read and difficult to communicate to students and their families. When I started to go through this year’s transcripts, I was reminded of Idan’s talk about extracting meaning from data, and presenting that meaning. I decided to apply those ideas to our transcripts.
The Current Transcript: Nothing but Text
The current transcript contains important information, but it is an ugly document that contains nothing but textual information. Let’s look at a fictional student’s transcript:
The transcript lists student information in a header at the top, and then lists every class the student has taken, chronologically by subject area. The subject areas themselves are listed alphabetically, rather than by any semantic meaning. The transcript closes with information about the student’s test results, and some summary information about the student’s overall academic performance.
The Problem: Analyzing a Transcript
There are two main questions at the center of any conversation with a student about their transcript:
- How much progress has this student made in each of the required academic areas?
- Which classes does this student need to take in order to stay on track for graduation?
These are pretty simple questions, but they are not easy to answer from the current transcript. To understand how difficult this conversation is, consider the overall graduation requirements:
- 4 credits of English
- 2 credits of Math
- 3 credits of Science, including 1 credit of Physical Science and 1 credit of Life Science
- 3 credits of Social Studies, including:
- 1 credit of Government
- 1 credit of US History
- 0.5 credit of Alaska Studies
- 0.5 credit of Global Issues
- 2 credits of Physical Education, including 0.5 credits of Health
- 2 credits of Vocational
- 23 credits overall
- Pass the High School Graduation Qualifying Exam in Reading, Writing, and Math
- Pass a Swim Competency test
All of this information is contained on the official transcript, but it is not easy to extract. We end up making notes all over the transcript to summarize where students are at. For example, we find ourselves writing out all of the Social Studies requirements in a little table near that section on the transcript, and checking off which requirements have been met. This is error-prone because it is very easy to miss a quarter-credit, especially when students have taken the same class multiple times and earned zero or partial credit in the class.
Even when we as teachers have an accurate understanding of a student’s progress, many students walk away from these conversations without really understanding their current academic standing. They just trust us to keep telling them which classes they need to take, and that they will graduate if they take those classes.
Rethinking the Transcript: A Visualization Exercise
Idan’s talk about Data and Meaning encouraged us to focus on extracting the most meaningful information from our data, and removing anything that distracts readers from that meaning. In redesigning our transcript, I had the following goals in mind:
- Emphasize student progress towards academic requirements, rather than deficiencies in the transcript;
- Show the academic requirements visually;
- Let the transcript calculate every number that we need for us, such as how many credits of a certain subject area a student still needs.
With these goals in mind, it was fairly straightforward to come up with a better presentation of the transcript’s information. Here is what I came up with:
The top section combines the academic requirements, and the progress students have made toward meeting those requirements. There are a lot of specific requirements, but we can now see visually how close a student is to meeting all of the requirements. Looking at Social Studies, for example, we can see how much easier our conversation with Arthur will be: ”You need a half credit of US History, and a quarter credit of Global Issues.” Arthur can probably interpret these credit needs himself now, and the conversation can focus more on how to meet those requirements.
The smaller graph at the lower right summarizes testing information. We can easily see that Arthur has taken all three qualifying exams, but still needs to pass the Writing and Math sections. Arthur has passed his Swim Competency test already.
The table at the lower left could use some visual cleaning up, but all of this information is actually vital to these conversations. The visualization makes it easy to see student progress, but we still need to give them hard numbers. This table lists every number we might need to share about what credits are needed to graduate.
One strength of this presentation is the de-emphasis on Elective credit. Many students believe it makes no sense to take more core academic classes once they have met that subject area’s minimum requirements. They think they need to fill out their credits with Electives, failing to realize that an additional half-credit of Science helps them meet their 23-credit requirement just as much as an Elective class does. The rephrasing of “Required Content” and “Free Choice”, along with the bars on the visualization at the top, helps communicate that students can increase their credits in any area to meet the 23-credit requirement.
I fully expect this visualization to streamline our conversations with students, and ease the workload of staff. We will start using these visualizations with students this week. I intend to play around with the overall design of the document, but even in its current form I expect it to:
- Ease the burden on staff of accurately interpreting student transcripts;
- Shift the focus of academic advising conversations from what credit is needed, to how students can earn that credit;
- Increase the number of core academic classes students take to meet the overall 23-credit requirement;
- Increase the confidence students have that they are on track to graduate.
If you have thoughts on how to further improve our transcripts, please feel free to share them in the comments or to get in touch with me directly.
I had been intending to write a Python script to create a visual transcript for a long time, but never got around to finishing it. Then I read Nate Silver’s AMA on reddit, and I was surprised to learn that he uses Excel to make most of his graphs. That made me look at whether I could use Excel to make a quick version of a visual transcript. I was happy to find it was fairly straightforward to make something meaningful using Excel.
I am tempted to go back and finish the Python script at some point, but I don’t think I will. I like the idea of school structures being sustainable, and no one else at my school is going to dig into Python code to maintain visual transcripts if I move on to another school. I like the idea of students maintaining their own visual transcripts, and I like the notion that a technically-minded teacher could update this spreadsheet if they need to at some point. The Excel sheet is on scribd if you’d like to play around with it.