I do not use many charts and graphs in this column. They can be confusing, and I often don't understand them anyway.
But I have just found in a new book a chart that is so interesting and powerful that I cannot resist. The book is "Do What Works: How Proven Practices Can Improve America's Public Schools," by Tom Luce and Lee Thompson (available at www.communitiesjust4kids.org). Luce is a lawyer who has become one of the heroes of the improvement-of-education effort in Texas, first with his work with Ross Perot, toughening state standards, and then as chairman of the National Center for Educational Accountability and founder of Just for the Kids, an organization that is greatly improving the way schools use student testing data. Lee Thompson, also a lawyer, is deputy director of the O'Donnell Foundation, which is encouraging Texas students to take more challenging courses in high school.
_____About the Author_____
Jay Mathews, a Washington Post education reporter, writes a weekly Class Struggle column exclusively for washingtonpost.com. He also covers school issues in a quarterly column for The Post Magazine. He can be reached via e-mail at firstname.lastname@example.org.
Here is the chart (to the right), which appears as Figure 17 on page 143 of the book:
Okay, it does look boring at first glance. It has too many numbers and percentages, and no explanation of what it means. But to someone like me who has, along with many high school Advanced Placement (AP) teachers, sought statistical proof of the power of their work with disadvantaged students, this is like finding that big flat-screen TV I always wanted under the Christmas tree.
In exactly two weeks I celebrate the anniversary of what was, other than my wedding day, the births of my children and the day this newspaper hired me, the most important day of my life, Dec. 7, 1982. On that morning I drove over to Garfield High School in East Los Angeles because of a strange little news story I had seen in the Los Angeles Times and introduced myself to a fat, balding math teacher with a heavy Bolivian accent named Jaime Alfonso Escalante.
Six years later Escalante became famous when two young filmmakers, Ramon Menendez and Tom Musca, persuaded Edward James Olmos to play the rebel calculus teacher in a feature film, "Stand and Deliver." But on that December morning in 1982 hardly anyone had ever heard of the man. The story the Times had published that morning did little to advance his reputation because it focused on cheating accusations against his students and did not explain how he had managed to produce 18 students who, everyone eventually agreed, legitimately passed the Advanced Placement calculus exam in a school so disadvantaged that it was a surprise to find any students even thinking of taking one of the most difficult tests in American secondary education.
I spent the next five years trying to answer that question: how did he do it? And while I was hanging around the school, Escalante's numbers just kept going up. In 1987, 129 Garfield students took AP calculus tests, with 66 percent passing, which was above the national average. That was more students taking AP calculus than all but four high schools, private and public, in the entire country. That could not have happened, people told me, because kids from low-income Mexican American homes, with parents who dropped out of grade school, cannot learn at that level.
But I had seen it happen, and have spent the past 22 years of my life trying to explain to readers why those kids were able to learn so much, and why most American high schools are so frustratingly blind to the lessons of Escalante and Garfield.
It wasn't magic. Escalante and the many other great AP teachers at that school did not triumph because they were classroom geniuses. Then and now, they successfully prepared low-income students for college level tests by encouraging them to believe they were capable and by making sure they had enough time to prepare.
I learned one more thing at Garfield that is still so contrary to popular opinion that one parent in a wealthy suburban Chicago high school said in a letter to her local newspaper that it "defied common sense." Students who struggle in an AP course with its college-sized reading list and flunk the college-level, three-hour final exam, I learned, are still much better off than if they had been denied a chance to take the course and the test. They have just played 72 holes with the academic equivalent of Tiger Woods, and although Tiger has beaten them, they have gained from the experience a visceral appreciation of what they are going to have to do to survive in college. That taste of academic trauma stays with them and helps them work hard enough to get their bachelor's degree.
Which takes us back to Figure 17 on page 143 of "Do What Works." The chart is statistical proof of what the lady in Winnetka, Ill., who wrote that letter thought was so unbelievable.
The left column in the chart, under "Passed an AP exam," is the easiest to understand. Those students showed some academic talent in high school and have good degree completion rates, above the national average for five years after high school graduation. I thought it was odd that the overall degree completion portion of 57 percent was identical to the percent for Anglos. Since the other ethnic groups percentages were lower, I thought the overall percentage had to be lower than 57 percent. But Luce explained to me that that average was pulled up by the higher college completion rates of other ethnicities, particularly Asian American students, who were not listed on the chart but had a 62 percent graduation rate in the "Passed and AP exam" column.
The exciting parts of the chart for me are the middle and right columns, under "Took, But Did Not Pass" an AP exam and "Did Not Take" an AP exam. Students who did not take AP in high school showed little success in college. That was not very startling. But look at the college completion percentages of students who took and failed an AP exam.
Theirs was a strange kind of failure. They were beaten by the equivalent of 30 or 40 strokes by this Tiger Woods of exams, but they still substantially increased their chances of college success. Anglos who flunked an AP exam were twice as likely to get their degrees as Anglos who never took one. Hispanics, African American and low-income students were three times as likely to get their degrees if they at least tried AP.
These results confirm the work of Clifford Adelman, a U.S. Education Department senior researcher who in 1999 reported similar results from an analysis of cohort data. But the group he studied included only about 8,700 students. Luce's sample, the result of his work to attach a student identity number to every important test in Texas so that individual improvement can be monitored, is nearly 10 times as large as Adelman's, and even more powerful.
Every AP teacher in the country should copy that chart, blow it up to three-by-four foot size and tape it to the wall of his or her classroom. International Baccalaureate teachers should do the same, because their college-level courses and exams are just as challenging. And every parent and every student in every high school that restricts access to AP and IB courses -- this includes, sadly, the vast majority of American high schools -- should wave this chart in the faces of the principals, school superintendents, school board members and other thick-headed people who are refusing to open the doors to AP for everyone.
An advocate for gifted education has been exchanging e-mails with me lately. She worries that letting too many B and C students into AP and IB courses will slow them down, and hurt the A students. I have seen no evidence of that, but if it is true, I know what Jaime Escalante would do about it.
He often made fun of the gifted and talented program at Garfield. He thought it was silly to give special instruction to students just because they had done well on an IQ test in the second grade. The few gifted and talented students in his classes often did not do as well on the AP exam as students who were not so designated, and he sometimes succumbed to the temptation to point that out to them.
So I am sure he would be happy to accept an idea I once suggested at an AP convention. Let the fast kids have their own AP courses. Let everyone else who wants to take AP, and experience the power of the data in Figure 17, enroll in their own, lower-level AP courses. They can call it AP for dummies, and see just how well they do when compared to the smart kids in their own little enclave. They will all be taking the same difficult exams, and that kind of inspiration can sometimes do wonders. Just look at what happened to the Red Sox this year.
At least those kids who want to succeed in college will be allowed into AP, which is the important thing. Any educator who has seen Tom Luce and Lee Thompson's chart, and read their book, must acknowledge that restricting access to our best high school courses is a form of educational malpractice. Show them the numbers, and tell them if they don't understand them, they might want to try another line of work. They don't really qualify as educators any more.