APS' voila math on graduation statistics.

Imagine the you have a group of X number of ninth graders.
Four, maybe five or more years later, you count heads of "graduates".

You divide the small number by the large, multiply times 100, and you have the percentage of APS ninth graders who "graduate".

That is the essence of the calculation.

Example, using imaginary numbers;

Let's say that APS has a group of 1000 ninth graders, and
at graduation there are only 480. APS' graduation rate would
be 48%, as reported.

Now if you ask that same group of 1000 ninth graders to raise
their hand if they have already flunked the ninth grade,
238.1 of them will raise their hands and then, apparently,
be asked to leave the calculation of graduation percentage.

The Journal reports that Rose-Ann McKernan, Executive Director of Instruction and Accountability at APS said,

"The change in the graduation rate does not reflect a significant difference in the number of graduating students, but rather in the overall number of students tracked.
... many students should not have been counted because they were not first-time ninth-graders."

Four years later, you still have 480 graduates, but you get to divide by 761.9 students instead of 1000.

Voila, your graduation rate is now "actually" 63%.

Makes perfect sense except for the part where they get to take the kids least likely to graduate out of their calculation of ninth graders who will likely graduate.

Voila?