D.C. schools’ new five-year plan says the 40 lowest-performing schools need to increase their collective math and reading proficiency rate from 23 percent in 2011 to 63 percent by 2017. It is hard to understand how intelligent adults could believe that that target has any connection to reality.

The District’s records show that those schools have made no significant improvement in the past five years. No regular D.C. public school has jumped that high that fast. The team that former chancellor Michelle Rhee brought in five years ago, including current chancellor Kaya Henderson, can’t explain what it will do differently in the next five years. Even worse, the new goals might inspire more of the corner-cutting and dishonesty that have tainted the effort to help kids.

Nalle Elementary School, for instance, has gone nowhere in the past five years. Twenty percent of its students tested proficient in math in 2008 and just 29 percent this year. Reading proficiency was 24 percent in 2008 and 23 percent this year. The five-year plan says the District will fix this by continuing “to invest in high quality instruction” and “increase investments” to reduce tardiness and truancy, an approach that hasn’t worked previously. Yet Nalle must shoot for 70.7 percent proficiency in both math and reading by 2017.

Aiton Elementary School had 58 percent of its students proficient in both math and reading in 2008. By 2012, proficiency rates had skidded to 21 percent in math and 18 percent in reading. The five-year plan doesn’t explain why that drop occurred. D.C. officials either don’t know or are afraid to say. Still, they have big goals for Aiton — 57 percent proficiency in math and 62 percent in reading by 2017.

Proficiency rates at Noyes Education Campus peaked at 63 percent in math and 84 percent in reading in 2009. By this year they had fallen to 33 percent in math and 32 percent in reading.

Test security experts such as Greg Cizek at the University of North Carolina at Chapel Hill say such drops often indicate that the high scores were caused by adults changing answers.

Noyes won teacher and principal bonuses and a National Blue Ribbon School designation for its high scores in 2009, when 81 percent of Noyes classrooms were flagged by the CTB/McGraw-Hill testing company for extraordinary numbers of wrong-to-right erasures on the annual exams. When investigations began and a new principal was hired, the scores dropped. D.C. officials have never explained the erasures, but their 2017 goals for the school — 68.3 percent proficiency in math and 71.9 percent in reading — could again open the door to potentially questionable practices.

Public schools with lasting gains in achievement usually don’t announce test-score targets. They prefer to support every student and let the statistics take care of themselves. Doug Lemov, a founding managing director of the Uncommon Schools charter network, said a 65 percent proficiency target “feels strange” because it means “we win if 35 percent of our kids fail to meet the standard. It’s just a perverse message.”

D.C. schools spokeswoman Melissa Salmanowitz said in defense of the targets, “If we want all of our students to succeed, we have to set clear, high expectations and support their work with excellent teachers, strong school leaders and a rigorous curriculum.” Veteran D.C. educators have heard that before. A teacher told me that the targets showed that District leaders were “completely divorced from reality.” A principal said that “there doesn’t seem to be any rhyme or reason” to the target numbers.

D.C. school leaders can be smart. Last week they offered teachers up to $1,000 in grants for innovative classroom projects, a terrific idea. A good follow-up would be junking Fantasyland test-score targets that only worsen schools’ chances of real improvement.

For previous Mathews columns, go to washingtonpost.com/blogs/