Wednesday, August 20, 2014

Cohorts and Critical Mass

Yesterday I attended a statewide meeting of community college and career center people to discuss various projects we’ve been working on to help displaced workers find their way through our systems to employable credentials and actual jobs.  It was heartening in some ways and frustrating in others, which is to be expected; the goals are good and the successes real, but if it were easy, it would have been done years ago.  

(For the record, I remain convinced that the phone lines between the Department of Labor and the Department of Education don’t work.  If they did, I would think someone would have realized that “stackability” presumes stopouts, which make time-to-completion numbers look bad.  We can be flexible or we can be tidy, but we can’t be both.  Sometimes, all good things don’t fit together.)

The discussion that hit home the hardest for me was about ‘cohort’ programs and the problem of critical mass.  A ‘cohort’ program is built to move a given set of students through together.  The idea is that the students will bond with each other and provide the informal support that often makes the difference between persevering and quitting.  That’s especially important for adult students, since they’re often under greater external demands than eighteen-year-olds, and they’re often looking for a different kind of bonding.

When everything aligns, cohort programs offer great appeal.  They’re easy to schedule, since everyone moves in lockstep.  Bonding is nearly automatic.  In best cases, faculty can even collaborate on assignments across classes, helping students see connections.  (We call that a “linked course” model.)  

But cohorts have a way of falling apart.  In yesterday’s discussion, I was consoled to see that everyone else who runs cohorts has seen the same issues I have.  Among the major causes of entropy are:

- Transfer credit.  Adults sometimes show up without credits, but they frequently bring anywhere from a few scattered courses to a previous degree.  If the cohort is built on the assumption that, say, everyone will take English Comp in the Spring, but half of the students have previous credit for English Comp, then you have a serious scheduling problem.

- Failures/Drops.  Between academic rigor and the realities of adult lives, sometimes people just can’t stay on the path.  Finances intrude, someone gets sick, or a long-dormant math phobia returns with a vengeance.  Suddenly, someone is off-track, and the next cohort doesn’t come along for another year.  

- Remediation.  Some need it, some don’t.  

- Jobs/Families.  With working adults, you have to assume some competing demands for time, attention, and money.  It comes with the territory.

Let’s assume an 80 percent completion rate per semester, no non-academic drops, and a four-semester program.  Start with 20 students (and round, where necessary).  After the first semester, you have 16.  After the second, 13.  After the third, 10.  And you graduate 8 on time.  

So what, you ask?  Small sections pose a financial problem for the college.  Even assuming no non-academic motives for dropping out and no transfer credits, you’re already committing to running a full slate of small classes in the second year.  Assuming the usual distribution of transfer credits, knock another twenty percent or so off each section.  And lose another one or two to the random stuff of life.  Now you’re looking at a fourth-semester cohort of maybe five or six people.  The economics of running a full slate of classes that size are often prohibitive.  

The same thing happens with sequenced courses in the larger college, though to a lesser degree.  The fourth semester of Spanish gets far fewer students than the first semester.  But there, the starting number is so much larger that it’s still often possible to have decent section sizes at the upper levels, if you schedule carefully.  In a self-contained cohort program, you’re starting with a much smaller number, so each new round of attrition hurts that much more.  And they aren’t offsetting the one small class with several other larger ones, as students in fourth-semester Spanish typically do.  In the cohort, every fourth-semester class is small.

The economics of attrition create a nasty chicken-and-egg problem for colleges.  You don’t want to run a cohort program without critical mass, but you’re unlikely to get critical mass until the world knows the program is there.  In practice, that often means eating some severe losses in the first few rounds of later-sequence courses, until the program either takes off or doesn’t.  When your fiscal climate is austere already, that’s a tall order.

Cohort programs are much beloved by the think-tank types who insist on “pipeline” metaphors.  They look clean, on paper.  But they assume a disembodied, disconnected student with no competing demands and no history.  That’s simply not how students -- especially adult students -- are.  Students are stubbornly embodied, with complicated needs.  

Nobody had an easy and elegant solution, but I was glad to see many of us asking the right questions.  Wise and worldly readers, have you found an elegant way around the issue of cohorts and critical mass?