Two years ago, first graders at Charles Barrett Elementary School in Alexandria ranked in the 39th percentile on a national test of reading skills. The following year, when those same students were tested again, the scores had shot up to the 68th percentile.

Three years ago, achievement test scores at Barrett showed that the first graders were average readers; that is, they scored in the 50th percentile. Last year, the same students, then in the third grade, were scoring in the top quarter nationally on reading tests.

What's happening? Have the teachers at Barrett discovered some fountain of instant knowledge?

Not quite. Instead, what they're saying in the halls in this school in the upper northeast corner of the city is that a bit of computer wizardry can sometimes go a long way.

"It may be all an illusion . . .," says James Akin, who brought a $500 computer with him two years ago when he became principal at Barrett. "But somehow the combined effect of both teachers and students knowing that I am watching and can evaluate every test score and every skill seems to keep the scores improving.

"And, what is more important, is the fact that I know that every day when I see the test results I can see if a student has a problem, and we can catch it then. Not when it's too late."

Indeed, the results are impressive. On a recent morning visit Akin produced charts showing standardized test score improvement at nearly every one of the six grade levels over the last two years -- about the same period Akin and his computer have been at Barrett. Akin, who says he is the only elementary principal to consistently use a computer to monitor scores, this year added a $1,000 printer to his system. All the computer equipment has been paid for with school funds.

Teachers at Barrett follow the same basic reading curriculum prescribed for most of the city's elementary schools. The difference, Akin and teachers say, is The Computer. All students must pass a standardized test before they can progress to the next learning unit and most students are tested every two weeks. The scores, graded by a reading specialist, are immediately sent to Akin for analysis and to be recorded in the computer. If a student fails an important section of the test, he or she cannot er. proceed to the next unit until remedial work is completed.

Few days pas when Akin can't be found tapping furiously on the computer keyboard on his desk, recording the latest batch of test scores. If he sees a poor test result or an unusual dip in a particular skill for a student, Akin usually consults with the student's teacher and the reading specialist. Each student's name is entered into the system, with various skills to be learned. Sometimes if a problem cannot resolved, the student might be called in for consultation.

That way, Akin insists, "we catch most students before they fall through the cracks."

Akin says he is not too concerned about teaching style if the results show students are learning. Classrooms range from structured and tradititional to open and relaxed atmospheres.

"Each teacher has a different style and method, that doesn't concern me," Akin says. "Every once in a while I or the reading specialist might suggest another approach if there seems to be a problem. But usually the teacher is allowed to teach in whatever way they feel comfortable.

"I'm not concerned about strategy . . .What I'm concerned about is that fellow down there in the first grade who missed a few important items on his test. I want to know what he's doing wrong and correct it.

"There is a large enough supply of educators who will philosophize all day about teaching. They can do that all they want. Our business is to make sure Johnny can read."