Design Problem 2: Brainstorming Options

The goal this time around is to brainstorm ideas that seems promising.  We’re not going to worry yet if they fail one of the design criteria from before; we’re more focused right now on thinking of ideas that meet at least one criterion than throwing out ideas that don’t meet all of them. 

To organize the brainstorming, let’s categorize efforts by age group.  Based on class readings like Whatever it Takes, the Paul Tough book about the Harlem Children’s Zone all the TFs read earlier this year, by the time a student in a bad district is in high school, on average they’re already several grade levels behind.  This suggests we target broad catch-everyone programs at younger ages where that achievement gap is narrower and hasn’t had a chance to fester, and focus our initial efforts at the high school level on the few odds-beaters who have shown promise and achievement in spite of their environment.  The goal should be to build a good base of participants at the younger age group and follow them as they get older. 

We’ll start with some ideas (bullet-pointed) for program-based earlier interventions.  Based on our limited resources, a good approach may be to come up with existing program infrastructures where we can lend technical expertise.  Ideally this setup allows us to limit planning efforts and other start-up costs and get the ball rolling immediately in a short timeframe.

1. FIRST Robotics
   • FIRST is a program that exposes kids to building robots to meet a design challenge.  The challenge is different each year.  http://www.usfirst.org/.
   • Pros:
     • Direct exposure to hands-on technical activities
       • Design, build, test ties in especially well to engineering
       • Opportunities to enrich hands-on work and designs with basic math to solve small sub-problems – relevant application of math
     • Comes bracketed into different age-appropriate segments
       • Can start targeted to younger segments and expand into later years with same students
   • Cons:
     • Expensive
2. MathCounts
   • MathCounts is a middle school extracurricular program that revolves around preparation for a spring multi-school competition in mathematics.  Schools can field multiple “teams” of 4;  judging is based on individual and team problem-solving.    Local individual and team winners can proceed to state and national level competitions. 
   • Pros:
     • Junior high age group, more receptive
     • Teaches high-school level math in a one-off problem format, emphasizing creativity over formulaic skills
     • Can be treated as a contest or game as motivation, but without adverse consequences for failure
   • Cons:
     • Probably better geared to undergrads (or even high school students?) as mentors than PhD’s (maybe a PhD as overall admin to lead the undergrad / HS student effort?)

Self-started lower age-gr0up options: I’m not coming up with any good ones here yet.  Opinions?  Leave a comment.

High School Options:

4.   Semester-long Seminar Electives (Seminars)

• Science electives on topics of interest to grad students who will serve as primary teachers
  • plasma physics
  • biomedical topics – artificial joints, etc.
• courses open to top-performing, interested seniors and juniors (by application?)
• classes taught by the grad student as a full GSI-supported position
• Classes heavily focused on labs and demos
• Include a design component to incorporate what the students learn
• Pros:
  • great opportunity to continue with lessons learned from current TFs
  • Focuses on likely future UM applicants
• Cons
  • Very heavily dependent on the skills and emphasis of the PhD who is teaching.  Probably best to do a team-teach scenario to share the load, have a partner to keep each other motivated / prevent being discouraged.
  • Likely teacher backlash for “stealing” their best students.  Wah.  This is why it needs to be marketed as an elective. 

5.   Magnet Science Courses with Heavy Math Emphasis

• As opposed to the above “advanced elective”, this is an equivalent course for, say, earth science, bio, chem, phys, the basics
• Still heavily demo and lab-based.  Bring in a seismometer, learn logs while learning the Richter Scale.  Do Brinell hardness testing with real MSE equipment while learning rock types.  Do DNA resequencing in Bio I with sweet equipment from a lab here.  Bring cutting edge to the class, and then present it well.
• Still require top performing and application for admission 
• Also GSI-supported
• Pros:
  • Earth science was boring.  Bio was a lot of rote memorization.  Chemistry was a little better, physics better still.  I think not coincidentally, that’s also about the order of increasing hands-on learning in those classes. 
  • Including labs takes expertise, and that’s what UM can provide. 
• Cons:
  • This is a stopgap measure to prevent losing the odds-beaters who are still engaged in 9th-10th grade before the college application process kicks in.  This is also a task that would likely require 2 PhD’s per class.
  • Teacher backlash, teacher backlash, teacher backlash.  See above, except this time it’s not an elective, it’s their turf. 
  • You’re going to need a full-time teacher present, to address legal liability / trained teacher present issues.  You’re also going to need to make this teacher the backup, secondary to primary instruction by the PhD’s with demos.  Depending on the teacher, this may be a very tough (pride issues) or very easy (less stress teaching) issues.
  • You’re also going to need that teacher doing the grading and administrative to keep the PhD’s under whatever half-appointment tuition time commitment level you’re looking at.
  • The quality of the PhD is again critical here.  It’s better not to do it at all than to do it poorly. 

Other ideas:  This one doesn’t fit into an age bracket, because it’s not an in-school idea.  It’s one that can be done safely from anywhere you’re in front of a computer, though it would likely benefit from insights gained from the TF program. 

1. Data Analysis for Student, Teacher Performance Evaluation Metrics
   • Fact is, UM involvement doesn’t necessarily mean we need to be on the front lines with the kids.  A heckuva lot of PhD and professorial types are terrible with kids, but great with numbers.  Leverage that – create better tools to keep track of the numbers.
   • Get a couple of PhD’s (maybe School of Ed. plus an EECS coding monkey) to partner on writing an open-source software total monitoring system that can keep a teacher abreast of student issues at a glance, and keep the student appraised of their performance as well.
   • Aggressively support and direct these students in applying for external fellowship funding so they don’t have to cut into your $40k
   • Pros
     • Schools pay good money to get all their grades, attendance, data taken care of.  Dead tree grade books are out.  But those systems kind of suck, and I’d bet a dime to a dollar that the data isn’t effectively crunched.  Build the system, give it away and you have built-in access to the (anonymized) data.  It’s the Google model.
     • Your product has immediate applicability anywhere, not just in Podunk School District.  Broad impact.
     • If you can monetize the product without subjecting your customer (the school districts) to high deployment costs, you have the potential to spin this off as an enterprise that can fund itself and expand.  The issue here is figuring out a revenue model that, if we follow the Google example and guess ads are involved, remains tasteful and refrains from projecting a crass commercialistic veneer over a fundamentally beneficial endeavor.
       • You don’t want text messages automatically sent out to failing students advertising fly-by-night overpriced tutoring services.  This is the nightmare scenario.
   • Cons
     • This requires substantial creativity, vision and focus, and it does not offer any of the warm, fuzzy feel-good aspect that direct involvement with the students does.
     • The district can always tell you to get lost, they don’t want your system
     • You need really good intel on the school’s desires to gear your tools to be appealing to them.  You also need good insight into what they ACTUALLY need, not their own perceptions of same, to add value to the product.

For some good info on why better performance metrics are a really good idea, see these links:

• http://www.gatesfoundation.org/united-states/Documents/empowering-effective-teachers-empowering-strategy.pdf
• http://www.gatesfoundation.org/united-states/Documents/empowering-effective-teachers-readiness-for-reform.pdf

The two reports share an identical intro section, but they are different once you get about 7-8 pages in. More on this in a subsequent post.