By Jordan Bienstock, CNN
(CNN) - It began on May 7 with Chemistry and Environmental Science, and ended on May 18 with Human Geography and Spanish Literature. During the two weeks in between, millions of U.S. students pored over questions and essays on more than 30 Advanced Placement exams.
Now, all they can do is wait.
Advanced Placement, or AP, courses provide high school students the opportunity to earn college credit. They’re overseen by the College Board, the same organization that administers the SAT college admission test.
The battery of exams takes place in early May, but students won’t find out how they did until July, when scores are revealed.
Even then, students won’t know which questions they got correct or what individual mistakes they may have made on essays. All they receive is a number, 1 through 5, with a 3 or higher being a passing score.
How does the College Board arrive at those scores during the two months between the exams and the results?
It’s a multi-stage process, much like the AP tests themselves.
Nearly all Advanced Placement exams are comprised of multiple-choice and free-response sections.
The multiple-choice sections are scored by computer. The number of questions a student gets right equals his or her multiple-choice score. Simple enough.
The free-response sections consist of essays and problem solving. To evaluate this part of the exams, the College Board gathers teams in early June for the annual AP Readings.
“We believe it’s the largest scoring experience in the world in terms of the number of student responses that are scored,” says Trevor Packer, the College Board’s senior vice president for advanced placement and college readiness.
The AP Readings take place at four sites around the country with an all-star team of graders.
You know that scene in an action movie, when the team of heroes has been assembled and walks in slow motion together while an electric guitar wails on the soundtrack? Replace the actors with academics, and you’ve got the general idea.
According to Mark Cavone, executive director for AP program operations and finance, the College Board flies in 11,000 graders, split between high school teachers and college faculty.
“I’m amazed we can get people to give up a week of their summer to sit in a windowless convention center and score exams for eight hours a day,” says Packer.
That’s eight hours a day for seven straight days, if you’re a reader. If you’re part of the leadership structure at an AP Reading, you arrive four days ahead of time.
At a Reading site, teachers and professors are divided into tables to grade tests in their subject areas. Each table is staffed by eight readers and a Table Leader, who is responsible for managing the readers. The entire AP Reading is overseen by a Chief Reader.
Before the readers arrive, leaders look at sample student responses to establish scoring standards. Each question is assigned multiple score points, and leaders identify what level of student work meets each score level.
“When the readers show up, the leaders train them on each question so they understand clearly the standards around what a 1 is, what a 2 is, what a 3 is,” says Cavone. “We spend two full days making sure that they are understanding how to grade each of the questions, and that they’re clear on how to set standards and how to range the score scale.”
Throughout the AP Reading, Table Leaders are also responsible for what’s called back reading: re-reading a large number of exams at their assigned tables to make sure that what they see as a 7 on a scoring scale is what the reader saw as a 7.
“A Table Leader’s sole job is to make sure readers grade tests fairly and according to an appropriate standard,” Cavone says.
Once the scores are totaled for every question in the free-response section, those numbers are combined with the multiple-choice results to create a student’s composite score.
Some might question why the College Board doesn’t automate the free-response scoring, or move the AP Readings into an online community. After all, it’s the single-biggest expense in the organization’s budget.
“There are so few opportunities for higher and secondary education to rub shoulders,” says Packer. “I think it’s important to support bringing people together to share perspectives.
“The level of trust that is built is really healthy and vital in American education right now.”
The last two weeks of June are spent on the final step in the AP scoring process: grade setting. This is how a composite score is translated into the 1-5 number that students ultimately receive.
Packer says that when an AP exam is changed in a meaningful way, or when a new test is created, new standards are set.
Since AP courses offer college credit, the standards are established by professors testing college students on disguised versions of the AP exam questions.
An A-level performance from college students “becomes the threshold for what AP students need to earn to get a 5,” says Packer.
A 4 on the AP exam is equivalent to an A-, B+ or B from a college student, and the scoring scale continues from there.
In other years, instead of setting new standards, the College Board compares student performance on this year’s exams to previous years, based on questions that appear on both tests.
The range for what constitutes a 1-5 varies from year to year and from exam to exam. You can see a sample chart here.
Two weeks of testing, two weeks of preparation, two weeks of AP Readings, and two weeks of grade setting. The final AP scores may be simple, but it’s far from a simple process that dictates whether all that hard work in Advanced Placement courses will pay off in college.