Wednesday, September 26, 2012

Education policy - follow up

I wanted to continue some of the questions raised by Nels in his earlier posts here and here.  I think one of the biggest challenges to come to grips with in education reform is in figuring out what our goals are ... do we want to guarantee a certain baseline of knowledge or do we want to push the frontier as possible.  Probably both, but how do we weigh each of those goals?  This is a fundamental question in everything from early education tracking to availability of vocational schools to relative funding of public higher ed and community colleges vs. elementary and secondary school. 

There are two fundamental issues in my mind - equity and fairness of opportunity is one ... the more we weigh the idea that "everyone should get a fair shot at success in America," the more we might structure goals based on achieving minimum proficiency levels for everyone.  The other issue is: what is the benefit to society as a whole of educating our students?  This is a much broader question (of which the equity question is only a part) with more subtle implications.

For instance, let me posit a scenario for you:  Suppose one of the goals of society is to further our technology and allow creation of new innovations to improve all of our lives.  For instance: maybe we want to educate people so that we'll be able to do things like cure cancer someday or invent Internets.  OK, so if this is a goal what would an education policy look like to meet that particular goal (independent of the equity concerns)?  This is a question I've always been fascinated by ... is it better to structure education to identify a few of our best and brightest and push them to their utmost abilities?  Or is it best to try to get as many people as possible into a position where they can work on these problems and diversify our social portfolio ... essentially buying a lot of lottery tickets increasing the chances that one of them hits for us? 

I think history shows there is something to be said for both approaches.  Even if you look at the arena of sports and our recent Olympic games we see similar conflicts ... some countries and sports invest their resources in widespread youth programs to broaden the pool of possible athletes.  Others have very intense, specialized training programs where superstars are identified at an early age and are drilled night and day to push their abilities.

There may not be an obvious answer to this question, but it is these types of things that must be reconciled before we develop a metric to measure how we're doing.  Do we care more about performance on multiplication tables or performance in creative problem solving?  Do we want to promote expertise in a subject area or promote social skills and teamwork?  Again, we probably want to do all of these things, but our current testing mechanisms increasingly are very one dimensional and are ignoring many of these other equally (or perhaps more) important goals. 

This is partially a measurement issue - many of the things we're truly interested in are harder and more costly to measure so we give up and choose to just measure the easy stuff.  My current APSS 300 students recently have been thinking about how to operationalize and measure conceptual ideas and I think they're finding that the fact is that it's often really hard.  If what you really want a 3rd grade teacher to be imparting on her students is a multi-dimensional combination of (1) basic literacy skills (2) Teamwork and citizenship (3) Creativity and problem solving (4) Curiosity and a desire for learning, yet you only test and evaluate that teacher on point (1), then we are dramatically failing in our policy evaluation structure.  As a result, we are likely mis-using the skillsets of our teachers and misallocating our educational time and resources. 

My prescription for meaningful education reform would be to have a more serious national dialog about how much we value these other dimensions of education.  And secondly, we may need to undertake a commitment to evaluate the things that are hard to measure, remembering that quantitative standardized testing is only one of many measurement options.

One last thing to leave you with ... I had an elementary school teacher one year (3rd grade I think) who (much like one of my favorite Simpsons' episodes) taught us the food chain as follows: 
grass>rabbit>wolf>deer

I raised my hand and told her that I was pretty sure she had the last one wrong and that deer didn't eat wolves.  She disagreed because deer were bigger.  The class agreed with me and started yelling things out and exasperatedly she reversed the wolf and deer magnets on the board.  I raised my hand again saying deer didn't eat rabbits either.  That nearly got my recess privileges revoked.  The reason I tell this story is that all of us would have failed this question on a test as a result of her teaching and you might think she should have been fired.  However, this same teacher was AMAZING at getting students to work together, stifling bullying, and encouraging understanding and interaction between students with different backgrounds, races, genders, and disabilities - the two things I remember most from that year were the food chain debacle and what we learned in the "Everybody Counts" program. (and, I guess I also remember the challenges that came from having to go to school that year in classrooms constructed from cubicles in the concourse of a basketball stadium because our school levy didn't pass and we had no school ... but that's a topic for another day).  I got another shot to learn science better as I got older but I never had to relearn some of the other skills she imparted.  And really, which is the more important thing for 3rd graders to be mastering?

1 comment:

  1. Incidentally, for the nerdier, more mathematically inclined readers out there (if there are any) part of the reason I'm fascinated by the "how do you structure education to best cure cancer" question comes from the theoretical implications of different math choices you might make in setting up an optimization question. This is going to sound really technical, but if you're trying to model economic or technological growth as a function of human capital (ie: education levels) you have two basic structural options ... model total human capital as the sum of each individual's human capital, or model it in a way where there is some complemantarity amongst different skills. Essentially, (and I'm simplifying here) you can say Total Human Capital = person1's + person2's + person3's ... etc OR you can say Total Human Capital = person1's * person2's * person3's ... etc.

    The implication of choosing the first one is that it doesn't matter how it's distributed ... one person with a million PhD's is equivalent to a million people with 1st grade educations. In the second version, if any one individual's education level is zero, then the total is zero.

    At the honors colloquium the other night someone asked me whether economists (or maybe specifically me) believed that if we had a big enough, fast enough computer we could program it to make all of our policy decisions optimially for us. My answer was that maybe you could, but the problem is you'd have to know the goals to be able to program it. Things like this minor math issue of summation vs. multiplication would have dramatically different policy outcomes.

    /end math rant/

    ReplyDelete