NSF Awards: 1245841
This video introduces interested teachers/educators to BRIDGES, a software infrastructure for programming assignments in data structures and algorithms courses. BRIDGES provides two key capabilities: (1) easy to use interface to real world datasets spanning social networks, entertainment (movies on IMDB, Song Lyrics), scientific data (real-time USGIS Earthquake Data), civic issues (crime data), and literature (books); and (2) a visualization of the acquired data can be used in assignments by students to populate their implemented data structures, including the capability to bring out attributes of the dataset. The visualizations are displayed on the BRIDGES website and are easily shared (with family, friends, peers, etc) via a web link. The presentation will showcase a variety of example projects and assignments used in data structures and algorithms courses, with new extensions to lower level Computer Science courses, that can be of interest to a wider and more diverse audience, including high school
students and educators.
Allie Beckman
Undergraduate Research Assistant
Hello and thank you for your interest in the BRIDGES API. This year we are working toward expanding BRIDGES to the CS curriculum beyond data structures/algorithms courses as well as targeting CS in middle/high school. We are focused on making the BRIDGES API even easier to use and adapting the assignments to different levels of the curriculum to meet targeted learning outcomes.
Students appreciate having a product to interact with, to show friends when they reach a new milestone. With BRIDGES, each time they master a new concept, they have a visualization which helps them share it with pride.
To start the discussion here are some questions the BRIDGES team would like to ask you:
Thank you again for your interest in BRIDGES. We look forward to hearing from you!
Gillian Puttick
Senior Scientist
The idea of giving immediate visual feedback to programming input is engaging!
Could you say a little more about how you've engaged teachers in BRIDGES?
Kalpathi Subramanian
Associate Professor
Hi. Thanks for the post.
So, over the past 4-5 years (and continuing) we have deployed BRIDGES in Data Structures and Algorithms courses across nearly 10 US institutions through instructors expressing interest in the software, attending our workshops, etc. These have collected feedback from their students that we analyze and understand. All of these interventions have been in 4 year teaching colleges, community colleges and universities.
We hope to have a more appropriate version of BRIDGES for high schools by this fall. We have 2 school teachers that have expressed interest. We will also have BRIDGES Game API that would be useful for building simple 2D board style games, that we hope to have ready by fall.
-- krs
Ellis Bell
In the age of "big data" in biology and chemistry have you considered partnering with intro courses in biology and or chemistry to expose a broader range of students to the power of algorithmic thinking?
Erik Saule
Assistant Professor
Hello Ellis and thanks for your message.
We have thought of looking into bioinformatics problem, but we have not found yet the right approach. For instance, we investigated looking at some classic bio informatics problems, such as smith-waterman's algorithms; or looking at some protein-protein interaction network. But we haven't found a problem that was highly compelling and easy to explain to computer scientists in early courses. We are still looking into that.
Engaging biology and chemistry student is not something we directly considered. But if you have ideas on how to approach these student population, we would be glad to hear your ideas.
Any thoughts?
Matt Fisher
Professor
Two questions came to mind as I watched the video:
1) It sounds like the project has been running long enough (around 4 years?) that it would be possible to collect some longitudinal data on student persistence and career paths. If you have collected any data like this, are you seeing any differences with the students that participated in BRIDGES?
2) The idea of creating a visualization that a student can manipulate to change its appearance intrigues me. Has any thought been given to collecting evidence of the potential impact of BRIDGES on visual thinking?
Kalpathi Subramanian
Associate Professor
Thanks, Matt. A great question.
In fact, we are working towards a journal publication to include both quantitative, qualitative and longitudinal data from the project.
We have been tracking 3 cohorts of students over the years to look at persistence/retention. We found the following:
1. We measured student performance who underwent BRIDGES intervention vs those in the remaining sections in 4 follow on core courses. In 3 out of 4 courses, they outperformend the students in the non-BRIDGES sections.Fourth course didnt have enough data to measure significance
2. Retention has been difficult to measure. This requires a carefully controlled study and with the explosion in our majors in the past 5 years and difficulties of coordination across the whole course, the results have been inconclusive and hard to quantify. We collected a lot of survey data from the students in each semester and did a qualitative evaluation. The results were overwhelmingly positive (they were coded positive or negative by 3 independent coders).
Your point (2) -- no, we have definitely not measured impact on visual thinking. As conceived originally, the visualizations were used as an engagement tool. A few students have used BRIDGES as a means to debug their results by ensuring the visuals are doing what the program is supposed to.
Dave Miller
Interesting project, and thanks for sharing in the showcase. I'm wondering if your team has had any thoughts about applicability and transfer to K-12 secondary curriculum. Thanks for any insights! - Dave
Kalpathi Subramanian
Associate Professor
Great question. Absolutely. We actually presented a BRIDGES Games API this spring at ACM SIGCSE, described briefly at:
https://dl.acm.org/citation.cfm?id=3293758
We are hoping this API can be used to engage K-12 students with simple game based projects and reinforce basic concepts in CS (loops, control statements) while keeping them interested and 'hooked'. We have had a few school teachers express interest in using BRIDGES and we are working on adapting our assignments (and expanding the list) so that they are at a reasonable level for K-12 students.
We hope to deploy an initial version this fall. We shall see! Lots to do.
-- krs
Peg Cagle
math teacher & math department chair
The project appears to capitalize on a critical component of sustaining student excitement-namely, providing opportunities for early success. As the world revolves increasingly around Big Data, engaging students from the beginning of their CS studies in creating displays to both communicate and interpret large data sets seems relevant to both their potential careers and their future lives. I am curious if you have considered what a version of this might look like at the secondary level.
Erik Saule
Assistant Professor
Hello Peg,
Provided the kind of benefits that we see with college students to high school students is something we are looking into. Currently we are investigating providing support for AP CS A and AP CS Principles courses. The hard part is to provide support for what these classes teach so that we can integrate the material and provide engagement without breaking the flow of the course. If you go to instructor and tell them "Change everything you do and teach that way instead", it is not going to work.
The strategy we plan on following is understand what the instructors really need, map learning outcomes and topics of their courses and adjust our framework/assignments/datasets accordingly. We believe that many of the things that we do (e.g., algorithm on images, using the simple game API, simple text analysis) are going to map well to CS A and CS Principles courses and provide engagement benefits. The work we have to do is really to connect with people teaching these courses and providing clear instructions and scaffolding for the students. We are interested in doing that within the next year or so. If you know someone that could be interested in working with us, please connect us!
There is an other strategy that is potentially interesting and that @EllisBell is referring to. Using BRIDGES data and visualization to support non-CS courses. To give an example, you could imagine talking about literature by analyzing classic text with algorithmic techniques. Or teaching DNA mutation in biology by looking at the output of DNA alignment algorithms. This is probably a lot harder to do well in practice, but the benefits we would collectively reap would be much higher.
Further posting is closed as the showcase has ended.