Except for isolated studies, little is known about the effects of computers on children's learning rates.

So say educators and computer specialists who only now are starting to study the effects of computers on learning. "That is the question that everyone is avoiding answering," said Henry Jay Becker, a researcher at The Johns Hopkins University's Center for Social Organization of Schools.

It is known that computers can teach children new analytical skills and give them practice in applying those skills, said Carolyn Stauffer, manager of teacher training for Apple Computer Inc.

"The computer can be used to teach kids things they have never been taught before such as problem solving, communications, interaction of information, access of information, analysis and presentation of information," she said.

But while research has been done in lab settings, said Becker, many schools do not have enough computers for the lab findings to be validated in real-life learning situations.

"The other thing is people expect a little too much," said Becker. "Most programs are good at one thing, like having kids practice skills they understand conceptually but need to get down fast, but they don't substitute for having kids understand what they are supposed to do."

So-called "drill and practice" software is effective for learning math but not for reading, said Marlaine Lockheed, senior research scientist at the Educational Testing Service (ETS) of Princeton, N.J., who directs research conducted for the testing service by the Harvard Education Technology Center.

"Drill and practice is good for certain things where memory is important but not good for high-level inquiry thinking," she said. Another software line, "Writing to Read" is "very good for making kids write and read. I don't know whether they think more analytically or thoughtfully because it hasn't been tested."

However, word processors help to develop writing fluency, she said. "Kids who use word processors write longer sentences than kids who don't use them -- we are talking about a gray area and it is very important that more reserach be conducted on this."

Lastly, working with a computer is exciting to a child and seems to stimulate learning -- for example, by taking the sting out of getting wrong answers when the computer politely urges the student to try again. Significantly, the exchange between the computer screen and the student is private -- not in front of a class.

But educators are just beginning to understand the effects of these and other computerized education programs, said Lockheed.

Likewise, the effects of "simulation" software that exposes children to time-bound situations involving complex thought also are just beginning to attract research interest.

The difficulties in studying the effect of computer aided instruction on children are twofold, according to Roger Kershaw, ETS' director of technology research. Effects of learning "must be measured over time and it's still too early" to know what they are, he said; lack of research funds is a factor too.

"The important thing is teaching children how to think," said Kershaw. "Computers can be used that way but haven't gotten there yet. Teachers are reluctant to go for it because there are no studies that say computers can provide this kind of effectiveness."