Post has attachment
This year at SIGCSE 2017 Google sponsored a panel and roundtables session to share results from Year 2 from of Google’s Computer Science Capacity Awards program. You can access the panel slides here. Our focus now is on finding and working with other university CS programs interested in implementing these interventions at their institutions.

Post has attachment
We are preparing for the new academic year at Mount Holyoke College, and with it, an expansion of the MaGE (Megas and Gigas Educate) program. This year, MaGE mentors are integrated with CS2 as well as CS1. The mentors have created new active learning modules to pair with the CS2 curriculum. Current cohorts of mentors and mentors-in-training are twice as large as last year. This summer we shared the MaGE Training Curriculum broadly and are eager to hear from you if you have interest in adapting parts of it for peer mentor training at your institution. We also recognize that for peer mentors, learning and personal growth does not end at the end of the Training course. This year, we are developing and refining the curriculum of the MaGE Practicum (the course that active mentors enroll in) to better tie together with the Training curriculum, the learning goals of Mount Holyoke, and our goal of training computer science students to educate, mentor, and support others in inclusive ways.

We are developing an evidence-based peer teaching fellow (i.e, undergraduate teaching assistant) training course.

We invited several experts in running large enrollment introductory courses to share their knowledge with the our project staff at a one day meeting on July 27, 2015. The attendees included faculty from University of Washington (Stuart Reges), Stanford University (Mehran Sahami), Harvey Mudd College (Colleen Lewis), Harvard (Daven Farnham), and University of Virginia (Mark Sherriff). The representative(s) from each institution with large-enrollment introductory courses then provided an overview of their teaching assistant program. We focused on the main ideas behind recruiting, training, and management.

We are piloting the training course at NC State, UNC, and Duke in Fall 2016. At each institution, students will learn about logistical aspects of TAing (tools and evaluation of student work); student interactions and concerns (including office hours and diversity training); and educational theory (active learning; creation of assessments, evaluation of teaching; and effective education). Students will be expected to complete a project, teaching sample, and reflection. Upon successful completion of the training course, a student will be able to:

1. describe and execute successful strategies for guiding students to solve problems in their coursework
2. evaluate student deliverables efficiently while providing formative and summative feedback
3. support course instructors on common administrative items
4. effectively use course tooling to support student learning
5. evaluate teaching - of themselves and others
6. apply strategies for student interactions that support inclusion and diversity
7. identify strategies for handling student issues and students of concern

We were very happy that not one, but TWO of Mason's team members won George Mason University's highest teaching honor, the Teaching Excellence Award. Kinga Dobolyi is our lead classroom instructor for our novel self-paced classes and Mark Snyder provides invaluable advice and feedback to the project. Mason's TEA is extremely competitive with over 50 nominees per year, around 15 finalists, and only a handful of winners.

Equally exciting is that three of our undergraduate teaching assistants (UTAs, or "yellow sashes" in the project) were recognized as CS department Outstanding UTAs for last year.

In September, we will present the SPARC teaching model to Mason's 2016 Innovative Teaching & Learning conference ( Kinga will lead a "virtual class," with audience members as students and other team members playing the part of UTAs.

Our website for the project is:

At Carnegie Mellon University, I am developing support materials for a CS2 course on data structures and algorithms.  The support materials include short videos covering the usual topics of a CS2 course.  Each video has a set of questions to assess the videos.  The support materials have been made available to highs schools in the Pittsburgh area (7 of them are developing courses this summer), and to community colleges (9 across the US so far).   The video recording is completely done and currently I am working on the questions to evaluate the videos.

As part of the Megas and Gigas Educate (MaGE) program at Mount Holyoke College, the MaGE Training Course prepares students for the task of educating, mentoring, and supporting others in inclusive ways. This training course raises awareness of the role of social identity in learning, emphasizes active learning within computer science, and provides preparation for being technical peer mentors. We are excited to share our curriculum materials with the community, in hopes that other educators and students might consider adopting similar peer mentor training at their institutions. You can find the MaGE Training materials at:

Post has attachment
Happy summer everyone! It's hard to believe that the summer is almost half gone, and there's still so much to do to prepare for the coming year. As I've described in my past post, our focus on scalability this year will be on developing and trialing tools. The first of which will be deploying and modifying CMU's autolab ( as an both an autograder and an analysis tool for our students. We are investigating ways of incorporating some of the big ideas in CS education in the tool (e.g. worked examples, peer learning/grading) and researching how we can automate the scaling of some of these ideas.

The autograder we've been using is Webcat, and it works well for sure. We are migrating to autolab with the hope that it will be easier to expand and maintain over time. If you are looking for an autograder to use, I don't think you could go wrong with either technology, and I'd love to hear how such solutions are working for anyone else.

Our early results of studying one-on-one tutoring for computer science show that when we study fine-grained dialogue, we can gain insight into the ways in which students and tutors interact. Some of the patterns we see are not surprising, yet they can serve as catalysts to improve our processes. For example, students with low self-efficacy tend to interact in very different ways (posing more questions, not making changes to their code without asking the TA if it makes sense) compared to students with high self-efficacy. Personality is also important: extraverts may be ok with more back-and-forth from the TA compared to introverts. All of these factors exist within the overarching framework of the importance of adapting feedback to students' levels of knowledge: we need to think deeply about the individual student in front of us before deciding what kind of support to provide. And much more research is needed on cultural adaptation in one-on-one support.

At the University of Florida (as part of the Peer Teaching Fellows project in collaboration with Duke, UNC Chapel Hill, and NC State University) we are researching how people use dialogue for teaching and learning in the second computer science course for majors. We facilitate students' interaction with teaching assistants through an Eclipse plug-in called Ascend that synchronizes the view of students' code and allows instant messaging style chat. The data that we are collecting is giving us insights into how teaching assistants tend to support students, and in the near future will yield evidence-based strategies for supporting students even more effectively. 

At Mount Holyoke College, a liberal arts college for women, we are developing a new technical peer mentoring program: Megas and Gigas Educate (MaGE). The goals of MaGE are to grow enrollment capacity in introductory CS courses while maintaining close interaction and quality feedback, to increase enrollment and retention for women and other underrepresented groups, and to train students to educate, mentor, and support others in inclusive ways.

We piloted MaGE in our CS1 course last year; we're expanding to CS1 and CS2 this coming year. CS1 students bring varying interests, including Art, History, Biochemistry, Economics and Engineering; most students have no prior programming experience. Each CS1 student is assigned a peer mentor in a 9:1 ratio. Prior to becoming a peer mentor, students first participate in a rigorous training course that raises awareness of the role of social identity in learning, emphasizes active learning within computer science, and provides preparation for being technical peer mentors.

Initial findings, based on rating forms completed by CS1 students, show that students consistently rated the peer mentors as highly knowledgeable, approachable, and creative/flexible in their approaches. Students also credited the peer mentor’s feedback with improving their own self-efficacy and understanding of the material.
Wait while more posts are being loaded