MELISSA BLOCK, Host:
American schools spend billions each year on educational software. Now, as NPR's Larry Abramson reports, a new government study finds that many of the best-known educational programs do not improve reading and math scores.
LARRY ABRAMSON: Three out of four kids at Bell Multicultural High School in Washington, D.C. are immigrants, still struggling with English. But federal testing standards say they all must learn to read at grade level, so kids like Jamie Arias(ph) get an hour and a half of intensive reading instruction every day with a computer program called Read 180.
JAMIE ARIAS: (Reading) That drugs were first tested on animals. And many of them were hurt. Is testing - is animal testing evil(ph)?
ABRAMSON: Kristin De Vivo is with Scholastic, the company that designed Read 180.
KRISTIN DE VIVO: The teacher can then listen to the recording. And over the process of months, she can compare the recordings of the reading fluency from early on to later.
ABRAMSON: Administrators here turn to Read 180 because they needed the 180- degree turnaround in test scores that the program promises. Susan Frost works for the nonprofit group that helped raised the money so the school could buy the program.
SUSAN FROST: We did not expect the jump that we saw. However, we went from a 13 percent proficiency to 40.
ABRAMSON: So Russ Whitehurst, director of the Federal Institute for Educational Sciences, put together a scientific yearlong study. His results proved just the opposite.
RUSS WHITEHURST: In a word, on average the software products did not produce enhanced learning in classrooms.
ABRAMSON: Whitehurst says the results are particularly striking when compared to the very rosy claims made by software companies. This promotional video for one of the products tested in the federal study is typical. Principal Toni Trice (Unintelligible) lyrical about how the Waterford Early Learning Program boosted test scores at her school, Francis Scott Key Elementary in Indianapolis.
TONI TRICE: So it's like an aide in the room that doesn't talk back and will keep working even through lunch and no coffee breaks. And who benefits? The kids. Who looks good? The teachers.
ABRAMSON: Now the federal government says, no, it doesn't work. Toni Trice says, in her school, it does. No federal study can replicate the challenges she faces in her school, she says. For example, the State of Indiana does not offer full-day kindergarten and doesn't require kids to come to school till they are seven.
TRICE: So we can get a kid at seven who's been looking at "Young and The Restless," you know, for the first seven years of his life. And so then it becomes the thing of how do we catch him up.
ABRAMSON: Software manufactures claim the materials didn't work well in the federal study because those teachers didn't follow the directions to the letter. Scholastic, for example, has very specific guidelines for Read 180.
RUBY NUEWAN: Would those information go into the comparison box or would they go into the different contract box?
ABRAMSON: At Bell Multicultural High School, teacher Ruby Nuewan(ph) talks with students in a discussion session that Scholastic says is critical for success.
NUEWAN: Unidentified Students: (Unintelligible).
ABRAMSON: Students are only supposed to spend about 25 minutes in front of the computer - 25 minutes discussing, and then 25 minutes reading independently. Ruby Nuewan says it's not easy to stay on schedule.
NUEWAN: It's very hard because sometimes you get carried away, and that you'd see that some groups finish a lot earlier. And some groups, you know, take a little bit longer time to finish and I have to juggle with that.
ABRAMSON: The researchers in the federal studies say in the real world, teachers often have to change the schedule. Software that promises results should work even if conditions aren't perfect. Russ Whitehurst of the Institute for Educational Sciences says his study shows software-based materials just has not reached that level of reliability yet.
WHITEHURST: I think one message from the study is that we should not look for a magic bullet. Clearly, the kind of software products we were looking at here cannot be looked to by districts as providing the solution.
ABRAMSON: Larry Abramson, NPR News, Washington.
NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.