We have tested boot camp participants immediately after their experience and then again after another 1.5 years.  The tests are "quiz" based, with problems designed to evaluate boot camp concepts.  As might be expected, we find the biggest learning gains in those students who came in with the most limited knowledge.  Importantly, gains are retained in the long term (although this may additionally be related to subsequent instruction and experience).  It is a good idea to address impact on PhD theses - we will have to think about how to do that.  The other broad outcomes we measure are self-assessed confidence in, and familiarity with, quantitative approaches, and the extent of student interactions with other students outside their home program.

Question 2: My last response above begins to respond to your second question.  We are very intentional about team building.  We engineer teams such that no two students are from the same graduate program.  We start the week with team building ice breaker games, and we ask the students to evaluate and discuss their "styles" as collaborators.  They also establish team compacts on he first evening. We use daily "challenges" is to encourage a fun rivalry and a team spirit, this is fostered further with team temporary "tattoos" (that students like to swap), course tea-shirts, and prizes for the team that does best in the daily challenges, but also for the team that shows the best team spirit.  We evaluate whether this works by asking the students about their network in their annual reports - we are still evaluating that data - and we have also had our consultant evaluator interview their program chairs.  The interviews have revealed that the student's programs are convinced that they have formed a supportive and cohesive network.

Question 3: How we will sustain the program? This is still a work in progress.  We hope to obtain University and Divisional support, but are also interested in philanthropy and industry sponsorship.  The number of our graduates going into data science careers is now quite significant (11% of graduates in the last 5 years) suggesting that we are developing a useful pipeline to industry that may be considered worthy of support.

Question 4: I have talked about the program at the AAMC Graduate Education and Teaching (GREAT) Annual Meeting, and have been invited to speak at the upcoming “Transforming the Learning Environment in Biomedical Education” symposium at the University of Nebraska Medical Center (UNMC).  We also plan to publish our findings next year, once the funding comes to a close.  All our students attend the boot camp, and we are pleased to have a diverse class that this year includes 16 students from under represented backgrounds.  We have also developed a new mechanism to ensure underrepresented students are encouraged to return to the MBL for educational courses, and that they have access to funds to help support such training.

Thanks for the great questions!

Our 18 graduate programs are extremely broad (see https://biosciences.uchicago.edu/programs).  They span from immunology to computational neuroscience, from ecology to molecular biophysics, from developmental biology to public health sciences.  Because of this incredible breadth we insist that all boot campers are just "biologists" for a week, and we usually analyze 3 diverse data sets, one of which may be field based.  Our emphasis is on the commonalities of analysis problems, not on the details of the data itself.   While it would be fun to work with data we collect at MBL there is just not time to make that happen for nearly 100 students in a one week period, so instead we use previously collected dat sets from our workshop instructors who are UChicago faculty, so the precise datasets follow the individual's research interests.  Let me know if this addresses your questions adequately.

We have used a boot camp as part of one of our summer research programs for students working as part of a large multi-university consortium to build a radio telescope in South Africa. The students work on different aspects of the project in the summer at different universities, so we start the summer with one week boot camp. At the camp, they basic skills in programming specific to the projects they will be working on, as well as basic information about the science (cosmology) and the observational technique the telescope employs (interferometry). There are also social events to further build the sense of cohort the students have before leaving to their respective research sites for the summer.

We would love for our material to benefit others and are very happy to share it.  However, my co-presenters are better equipped to respond to this question than me, so I will let them tackle this one,

Dear Patricia,
the materials we prepared are actually well-suited for junior or senior undergraduates. We have collected the latest rendition of all of our materials in this GitHub repository. For example, here you can find a pdf collating all of the lecture notes, while here you can find self-contained short tutorials introducing programming in R. I think that the "basic" and "advanced" tutorials would be especially useful for undergraduates who want to start analyze their data in R. All the materials are open source: we welcome anybody who wants to adapt them to their own needs. I have also prepared similar notes for my undergraduate teaching, and I would be happy to share them.

Thanks for your comments and questions.   We take 8 teaching assistants with us to MBL each year, who are more senior graduate students. They play a key role in the instruction and the students often feel more comfortable asking them naive questions than they might of the faculty.  

Regarding publication, we do track all papers across the division, however I know that there is significant variation discipline by discipline, and even lab by lab. While we could quantify number of papers and time of first publication, we do not have any 'control' students who did not attend the boot camp.  So the only comparison we can do is between the last 4 'boot camp' cohorts versus the preceding 4 years—unfortunately not a very well controlled experiment given the temporal offset.  

We had not previously considered the utility of our program or one like it for undergraduate faculty, but the questions and comments are showing us that there could be a lot of interest in such a program. I will also share these comments with MBL leadership.  Thank you for the suggestion.