Lake Woebegon and grad rates

Astute readers looking at the graduation rates for Charlotte-Mecklenburg's 28 high schools may notice something odd:  24 of them list rates above the district average of 81 percent, including 13 above 90 percent.

That may call to mind Garrison Keillor's mythical Lake Woebegon, where all the children are above average.  Worse, it may raise questions about whether the numbers are valid.

In this case,  the seeming impossibility is due to the way school and district rates are calculated.  It's not a CMS thing;  I'm willing to bet you'll find the same pattern in virtually every district.

North Carolina,  like other states,  tracks the students who start high school and calculates the on-time graduation rate by the number who get a diploma four years later  (Superintendent Heath Morrison says our state is unusual in not including summer-school grads).  But students who switch schools after falling behind don't count toward school calculations.

Take,  for instance,  a student who starts ninth grade at School A but doesn't get enough credits to be promoted.  Sometime in his second year of high school he moves to School B.  School A doesn't get blame for his failure to graduate  (or credit if he does)  because he left.  But School B also isn't held responsible because he was behind when he got there.  As long as that student stays within one district,  though,  he still counts toward the district rate.

As you can imagine,  students who fall behind and move around are at special risk for failing to graduate.  So you see those students bringing down the district and state numbers without dragging down schools.

Morrison and his crew are well aware that plenty of people,  including me,  scrutinize CMS numbers closely  --  and with good reason.  The district has gotten black eyes for bad data,  including flawed graduation rates reported in 2006,  when the current tracking system debuted,  and error-filled school progress reports posted last year.  The skepticism started flowing as soon as we posted a story on this year's gains.

John King of Harrisburg quickly emailed me questioning why CMS couldn't quantify the impact of a change in graduation requirements,  from 28 credits in previous years to 24 starting with the Class of 2013.

"Given any level of competent statistical tracking,  it should require very little time or effort to produce dual graduation numbers, one assuming 28 required credits and one assuming 24,"  King wrote.  "I do something similar almost every day in my job!  It’s a key step in evaluating the effectiveness of the decision to make the change.  The failure to do so simply proves that there is more interest in managing the perception than the result and that there is no more transparency under this Superintendent than under the last!"

I requested exactly that calculation Tuesday afternoon,  when I heard that graduation rates would be released Thursday.  I agree with King that it's valuable information,  and I'm disappointed it wasn't available right away.  But Morrison and Chief Accountability Officer Frank Barnes have repeatedly said they're going to check and recheck anything they release,  rather than take another credibility hit by giving out something they have to retract.  They say they'll give me the numbers when they're confident of them,  and I'll report on that information when I get it.