The Charles Effect

This post was inspired by my cat using the kitty litter. Seriously.
My husband and I have two cats, and one of them - Charles - is a hot mess when I comes to tidy kitty behaviors. We adopted Charles from a local shelter, where a good samaritan had dropped him off.  I name this because his behaviors are largely informed by his early months as a stray.

When Charles uses the litter box, he obsessively tries to cover his mess. The thing is, he paws everything and anything in the area, most of which doesn't involve actually moving kitty litter.  He paws the sides of the box, he paws the litter box package if it's sitting close enough, he paws the air, he paws the wall….. and a good time later, he leaves; satisfied that he's completed his task.

The thing is, somehow Charles does end up adequately covering up his mess.
However, this only happens after a very long and unnecessarily excessive effort to do so.

What the heck does this have to do with teaching, data, and testing? I've heard or seen Charles engage in this act many times, but today the connection came to me and flooded my head…. when we test kids, we (often/hopefully) have a clear reason for doing so. We want to know if they've learned what we've taught, or we want to know if they can apply one understanding in a certain context, etc. Beyond the classroom, testing can be driven by a desire to ensure equitable access to instruction and content; or even to gauge what investments correlate to outcomes.

The thing is, having a purpose and then an outcome doesn't necessarily mean that everything that happens in between is necessary, efficient, or worthwhile.

Charles' goal is to cover up his mess, and he does so… but at the cost of excessive time and effort. All of his motions are the same; he paws at every surface with the same goal, and yet only some of those swipes actually translate to the desired outcome.

On this random morning, this has become the perfect analogy (in my head), for what we're doing with testing in education right now. It's obviously oversimplified, but in some ways it's not. Anytime I administer an assessment to my students, I'm acutely aware of the limitations.

I'm actually not entirely against testing. And I love using data to inform my practice (and my students' understanding of their work). That said, I'm increasingly frustrated with the Charles effect, if you will… where we are expending a lot of energy (and limited resources) on testing, without meaningfully reflecting on whether we're measuring the right things in the most efficient ways.