Hi Irene, thanks for your question. We ask the teachers to define a broad problem that contributes to an actual learning objective in their class. We figure that the teachers would know the problem area better than we ever could and are in a position to hep the students scope the specific problem they are working on. 

We also held monthly meetings (Disciplinary Commons-style) over the year so teachers can share their lessons with the cohort and get feedback, and also ask for help with technologies that are not familiar to them.

To follow up on this, I'm wondering whether you have a sense that teachers from some subjects have a harder time conceptualizing their lessons in this problem-based format than others. Have you found this to be the case or does it typically come easy to all of the teachers? I'm also wondering if there are particularly striking examples you could report of teachers approaching this in a particularly novel way.

What are different methods you have seen from teachers in different subject areas to engage students in problem based CS thinking? Our team is creating curriculum that infuses computer science into middle school classrooms, across subject areas. However, in making these lessons, we have found that certain subject lend themselves more naturally to algorithmic thinking. I am interested to know how are teachers, in these harder to reach, subject areas are engaging in problem based thinking in their classrooms.

Hi Alex and Allison, thank you for your questions! Indeed there are some subjects where algorithmic-based problem solving is not obvious. For example, in Music, creativity in composing music is important, but the music teachers found it hard to tie it to a problem that is relevant in the student's community. Instead, the teachers reframed the problem to focus on what they can do to improve music literacy in the community. Students then were able to come up with ideas for various games such as identifying the correct note on sheet music, guessing the pitch, etc.

Hi Katie, thanks for your question. We ask teachers to identify a broad problem that contributes to a actual learning objective in their class. The students come up with a more specific problem that can be solved computationally. Most of these can be boiled down to a few types: storytelling, simple games and quizzes, simulations, simple mobile apps, etc. Over the course of the summer PD, we showed teachers different examples of these and they in turn made suggestions to their students based on the specific problem they worked on.

Hi Nirit, thank you for your question. We could not collect student data directly because we don't have the approval of the school districts. From the teachers, we are able to glean some insights on student response. According to the teachers, most students do wish to continue learning more about CS. Interestingly, most teachers report also that they had better engagement from students who would otherwise not pay attention in their classes!

Hi Allison, our PD approach is essentially, "I do, we do, you do." We modeled a problem solving approach for the teachers, walking through a problem from definition to implementation. We then organized a week-long student summer camp (called Student Academy) and worked with the teachers to help them prepare lessons and engage the students in solving a problem. The teachers then prepared a problem-based lesson for their classroom and teach it themselves at some point in the academic year.

Throughout the Student Academy and academic year, we held meetings with the teachers to give them a chance to reflect and share experiences. While we did not explicitly ask about misconceptions and frustrations, teachers do share a few of these that arise from the lesson, such as some students not really understanding what is expected of them. It is a good idea for us to explicitly prompt the teachers about these -- we will add this to the teacher reflection sessions. Thanks!