This week, Florida announced it may change the way it calculates the grad rate, eliminating a formula that included students who passed the GED and state assessments as high school “graduates.”  The expected result, like Michigan, Florida may soon see a significant drop in the high school graduation rate overnight.

These are but two examples of the challenges facing states in high school improvement efforts.  Take a look at the longitudinal data on high school graduation, and the numbers are quite unsettling.  States like Michigan and Florida tell you one thing, while Education Week and its Graduation Counts effort tell a completely different story (and it is usually a far-scarier one).  Talk to an urban superintendent about his graduation numbers, and you’ll hear rates in the 70 or 80 percents.  Ask Jay Greene and the Manhattan Institute the same question about the same districts, and you will often get a number that is 30 to 40 percent less than the superintendent is offering.

Why the great variance?  How can intelligent people look at the same schools, the same students, and the same data, yet come up with results that hold no resemblance to one another?

Florida is the perfect example of that.  When we talk grad rates, we expect it to measure the percentage of kids who started ninth grade and then finished 12th grade four years later (or in some cases, five years later).  We don’t expect students who drop out to pursue GEDs to be included in the grad pool.  After all, those students did not graduate from high school.  They pursued an alternative education path, but they did not graduate.

We talk a lot about AYP and how to compare schools, districts, and states when it comes to academic achievement.  We question whether student reading proficiency in Mississippi is equal to student reading proficiency in Massachusetts.  So why is it so hard for us to wrap our hands around a singular, clear high school graduation rate?

Years ago, the National Governors Association got all 50 states to buy into a common graduation formula.  Take a look at how many kids start ninth grade.  Factor out the school transfers and similar considerations.  Then look at how many of those kids graduated four years later.  That’s the grad rate — how many students completed high school four years after beginning it.

Several states have adopted this formula (including Michigan, thus the change in its most recent numbers).  But many more still have yet to apply the common formula to their state’s data.  Some are holding off because they are fearful of announcing a significant drop in grad rate overnight.  Others are working on building the data collection systems they need to do the work effectively.  And still others are just trying to sort it all out, trying to fit this priority in with a growing list of state education needs.

No matter the reason, the time has come for all states to embrace a common longitudinal graduation rate.  There is simply too much at stake not to.  In virtually every state in the union, we talk about the need to prepare our students for the opportunities of the 21st century.  We talk about new skills and new jobs.  About working smarter.  We discuss that a high school diploma is no longer a sufficient terminal degree, and that postsecondary education is a necessary step for all.

Can we really get more kids into postsecondary education if we don’t know who is actually finishing high school?  How do we boost graduation rates if we don’t have an accurate baseline to build on?  How do we improve the high school experience if we don’t have good data on who finishes, who doesn’t, and why?

Michigan and Florida’s announcements are indications we are heading in the right direction.  The first step might be painful.  No one wants to see their grad rates significantly decline.  But it is the right thing to do.  And it is a necessary step if we are to improve our nation’s high schools, increasing the number of kids who graduate from high school and go on to college.