All but three states are using the SAT less.

The College Board aims to start a national crusade for college access with a revamp of its SAT admission test to debut in 2016.

But the nonprofit organization faces a major hurdle in its quest: Use of the SAT has shrunk in huge swaths of the country since the test’s last makeover nine years ago.

In 29 states, a Washington Post analysis found, there were fewer SAT test-takers in the high school class of 2013 than there were in the class of 2006. By comparison, usage of the rival ACT admission test fell in three states: Idaho, Maine and South Dakota.

The class of 2006 was the first to take the SAT after the test was expanded in 2005 to a 2400-point format with a required essay. In the wide-ranging revision the College Board announced March 5, which takes effect when today’s freshmen are college-bound juniors, the test will revert to its familiar 1600-point maximum score and its essay will become optional.

To look at full results of the analysis of ACT and SAT test-takers in a sortable table, click here. The state data are drawn from these Web sites: class of 2006 ACT; class of 2013 ACT; class of 2006 SAT; and class of 2013 SAT.

Examine the number of students who took college admission tests in the 50 states and the District of Columbia in the high school classes of 2006 and 2013.

Over seven years, the declines in SAT test-takers exceeded 20 percent in 19 states, including drops of 59 percent in Michigan, 46 percent in Illinois, 37 percent in Ohio and 25 percent in Tennessee. Those are all states where the ACT test is more widely used.

But the SAT also had declining usage in some states where its presence is greater than the ACT’s. In Pennsylvania, SAT usage declined 2 percent. In New Hampshire it was down 8 percent and in Vermont 14 percent.

The College Board declined to comment on The Post’s findings.

Both exams are accepted by selective colleges that require admission test scores.

The SAT, launched in 1926 and overseen by New York-based College Board, has a greater presence on the West Coast, in the Northeast and in the Washington region. The ACT, launched in 1959 and overseen by a nonprofit organization based in Iowa, has long been dominant in the center of the country and much of the South.

The older test was originally conceived as a measure of “aptitude,” the younger as a measure of what students have learned. But the College Board has abandoned the “aptitude” rhetoric. It now says the SAT “tests your knowledge of reading, writing and math — subjects that are taught every day in high school classrooms.”

Overall, the ACT surpassed the SAT in total usage in the high school class of 2012. But a closer look at the state-by-state numbers shows the depth of the challenges faced by the SAT and offers clues to why the College Board felt compelled to change course after the 2005 revision.

The College Board is changing the SAT again. The Post's Nick Anderson explains some things you might not know about the college admissions test and how it has evolved—starting with the meaning behind the letters "SAT." (Davin Coburn and Kate M. Tobey/The Washington Post)

Here are takeaways from The Post’s analysis of the 2006-2013 trend in test participation.

**The SAT’s edge is eroding in several of its core states.

Look at the Pacific states. In California, the most populous state in the country, the SAT grew 22 percent. But the ACT grew 92 percent in the Golden State. In Oregon, the SAT was flat while the ACT nearly tripled. In Washington state, the SAT grew 18 percent and the ACT 48 percent. In Hawaii the trend was the same: SAT up 10 percent, ACT up 137 percent. In Alaska, SAT participation actually fell slightly.

And now the Northeast. In New York, home to the College Board, the SAT grew 3 percent. But the ACT grew 78 percent. SAT numbers edged up slightly in New Jersey and Massachusetts. ACT numbers boomed, rising 80 percent in Massachusetts and tripling in New Jersey.

Previously The Post has reported on the ACT’s gains in the District, Maryland and Virginia.

**Competition is intense in Texas and Florida.

In Texas, the SAT led the ACT in participation in 2006 by a huge margin. The older test had about 130,000 participants in that class, the younger about 74,000. Over seven years, the SAT grow a robust 33 percent. The ACT grew even more--49 percent. Both now serve more than 100,000 college-bound Texans a year, but the SAT retains an edge.

In Florida, the SAT led the ACT in the class of 2006. The breakdown in participation for that class was roughly 95,000 to 66,000. Now the ACT has overtaken the SAT, with 124,000 test-takers compared to 113,000.

Two other “SAT states” in 2006 have flipped to the ACT: Arizona and North Carolina.

**The ACT has widened its advantage in most of its core states.

In Illinois, the ACT had about 137,000 participants in the class of 2006 and the SAT about 13,000 — an edge of more than 124,000 students. Now the edge is more than 153,000. The same pattern, with a widening ACT edge, holds in 23 states.

The SAT cut into the ACT’s numerical lead in just two states: South Dakota and Idaho. The difference is trivial in South Dakota, where the ACT still dominates. In Idaho, the SAT captured the market lead from the ACT through a contract the state started with the College Board in the 2011-12 school year to deliver the older test to students.

But a look at the big picture shows true market dominance for the ACT in many states. These are the 19 states where the ACT holds a lead of at least 10 to 1 in test usage: Alabama, Arkansas, Illinois, Iowa, Kansas, Kentucky, Louisiana, Michigan, Minnesota, Mississippi, Missouri, Nebraska, North Dakota, Oklahoma, South Dakota, Tennessee, Utah, Wisconsin and Wyoming.

In 10 of those, the ACT’s lead is at least 20 to 1.

By contrast, here is the one state where the SAT has an edge of 10 to 1: Maine.

If the College Board hopes for the revised SAT to have any influence in Chicago, Milwaukee, New Orleans, Detroit, Memphis, Little Rock, St. Louis and other cities in the ACT-dominated sphere, it faces a steep challenge.

**Bottom line: In the 50 states and the District of Columbia, the ACT had about this many participants in the class of 2013: 1.8 million.

The SAT had about this many: 1.5 million.

**How did this happen? A few factors are in play. First, the size of high school graduating classes has declined recently in many states. That demographic trend ripples through testing markets in unpredictable ways.

Second, the ACT has secured testing contracts with 13 states since 2001. The SAT apparently has had two such contracts — in Idaho and in Maine.

If you subtract states with testing contracts, the SAT leads the ACT in usage. That underscores that a great many students still choose to take the SAT.

But the contracts don’t fully explain the trend. First, in many states with ACT contracts, a substantial number of students probably would have chosen the ACT on their own anyway. Second, the ACT has had huge growth in many states without contracts. (See New Jersey and Pennsylvania, above.)

A third factor is the content of the tests. A major question is whether the SAT’s 2005 revision damaged its market position. In all likelihood, it did in many places. How much is unknown.

Fourth: The word is out among students that either test is acceptable for college applications. That’s a big change from previous generations, when the SAT was perceived in many quarters as the premier --and therefore “must-do”--test.

**Can the SAT rebound? Here is a statistic that could work in the College Board’s favor: Seven out of 10 high school graduates take the Preliminary SAT/National Merit Scholarship Qualifying Test. This test, 2 hours and 10 minutes long, is given to prepare students for admissions tests and qualify them to seek certain scholarships and academic laurels. That gives the College Board built-in marketing leverage to grow SAT participation.

The question is whether the revisions to the SAT — and other elements of the College Board’s access campaign — will succeed.