## Tuesday, April 28, 2009

### The Limits of Educational Software

So if you control for the variables you know help students achieve, your study of (fill in the blank) shows no significant gain due to (fill in the blank). In this case the blank is filled by the darling of grant funders everywhere, technology, specifically education delivered by computers.

For the second year in a row, a controversial \$14.4 million federal study testing the effectiveness of reading and math software programs has found few significant learning differences between students who used the technology and those taught using other methods.
...snip...
“These studies are intended to wash out all the variation in school environments, teacher quality, resources­—all the things that we, in fact, know make a difference when it comes to student learning,” said Margaret A. Honey, a technology expert who is the president of the New York Hall of Science.

In other words, students who do well with the educational software were doing well anyway probably because of the school environment or teacher quality or other known predictor of student achievement.

Technology is way over-rated. Consider the mathematical concept of place value. Teachers can design an activity that involves students boxing objects into groups of ten, and then packing ten boxes into a case, then stacking ten cases on a crate. I have such an activity with pinto beans. Students keep a columnar tally as they fill each “box” with beans. One rule of the game is that there can be no partially filled boxes, cases, or crates.

Subtraction is modeled with the same manipulatives. Students must often unpack a crate or a case or a box do complete the subtraction. If, for example, students must open a case in order to subtract boxes, they must empty the case entirely and stack all ten boxes with however many full boxes they already have, all the while keeping a columnar tally. If students have 2 crates, 4 cases, 3 boxes and 6 loose beans and want to “fill an order” (subtract) 4 boxes, they must empty a case. Now they have 2 crates, 3 cases, 13 boxes and 6 loose beans. We do not worry about conforming to the standard algorithm when we model on paper the actual mathematics of the task.

I once had a group of education students do the same activity on computer using a Java applet. Among other features, the applet used a rope tool to surround 10 loose objects and a box tool to pack into full boxes. I asked the students to compare the educational soundness of each activity. They quickly observed the “magical” aspect of the computer version. The concrete activity was real. Students could easily see how boxes were filled because they physically filled the boxes. The computer converted a lassoed group of objects to a box by some mystical means. At least, it may seem mystical to a child of the target age group.

Computers compromise sensory experience. No matter how 3D the graphics, the display is essentially two dimensional relying almost exclusively on the visual. Brain scientists might say the concrete activity forms more neural pathways by utilizing more of the five senses.

Paradoxically, some computer animation looks amazingly real. I often wonder how unhealthy a reliance on computers might be. At least in the days of Captain Kangaroo, small children could easily distinguish the real from the unreal. At a age when children are known to confuse reality and fantasy, can it really be a good idea to deliberately smudge the line between the two? Could computer animation undermine the development of analytical ability when the child's own senses cannot be trusted? When painted pictures of squirrels on cardboard placard danced around on Captain Kangaroo, no child was led to conclude that squirrels actually do hip-hop. The cardboard squirrel was obviously unreal. Can the same be said for the squirrel in the famous commercial doing a fist pump after causing a car accident? Will our children think less critically and be more vulnerable to scams?