Unconference Session notes:

The project will involve compiling information on CS programs from data sources like IPEDS, College Scorecard, the Common Data Set, and other such sources, followed by conducting a preliminary sample of institutions where we tag of 100-200 institutions and identify common course structures and features to keep in mind. Then, using information from this sample, we formalize tags and then tag courses, faculty information, major requirements, and other relevant information noted during the preliminary stage. Then, a front end would be created in order to access and search the database in a robust and useful way for researchers and educators.

After reviewing the tentative plan, we had a back and forth about many topics, which is summarized below:

Subject 1: Motivation

Since understanding about what programs at institutions look like is poor in computer science, this database can help us:

• Identify research directions
• Help when writing motivation for research
• Help identify curricular revisions
• Help convince administration of necessity of revisions (when seeking funding)

Subject 2: Things to keep in mind

The following are things that were discussed as things the project should do/keep in mind

• Information sources should be recorded and archived
• The data is only valid for the academic year data was gathered for (and should be available as an archive if updated in the future)
• The front end needs to be accessible rather than just another archaic spreadsheet
• There are many weird edge cases to consider:
  • What about institutions that are accredited and state approved to award degrees but don’t report to IPEDS?
  • What about faculty that are in multiple computing departments included in the survey?
    • eg: a professor who is both an assistant professor of computer science and assistant professor of data science
    • Another related issue: combined departments, like Math & CS, and how to distinguish faculty
• The importance of distinguishing between information obtained via the institution itself, vs information obtained by a separate source
  • By default, queries will only involve “institution sources” in queries, but option to use external sources can be manually toggled if the user desires
• Ways to structure the project as multi-year
  • 1.0: 4 year institutions with basic information
  • 2.0: Community colleges + graduate only programs
  • 3.0: Programs with only a minor
  • 4.0: Identify computing related sources taught in external departments
    • Incredibly extensive, comparable to 1.0 in workload

Subject 3: How to contribute

We discussed various ways collaborators can contribute:

• Helping tag faculty numbers and course information
• Helping design a website for the database
• Helping devise interesting questions that can be answered by the database and presented in the initial publication

Matt Ferland

Assistant Professor of Computer Science

Dickinson College