I like to listen to NPR in the mornings and afternoons. I sometimes find the news to be refreshing, but always interesting. Today’s feature was on the way Nielsen is falling behind in the way it calculates viewership for television shows. The move to watching TV on Tivo, Hulu, and other media outlets might just be making Nielsen’s ratings obsolete. Currently, Nielsen is not doing much, if anything, to calculate the eyeballs viewing a show in these formats.
But Tivo and Hulu can.
They can tell viewership based on time watched, commercials watched and fast-forwarded, and more. To do this, they have to look at a show over time. People don’t watch these shows on the nights they air. Viewers have found a way to watch when they want to, as long as they want to, in as many snippets as they want to, and they fast-forward through parts that are boring. What we often forget is that everything we do on Tivo and Hulu is being watched by someone. Every log-in is tracked by our IP address, and that tells them (if nothing else) where we live.
We shouldn’t be surprised that we are watched so closely over time. Groceries stores and big box stores have done this for years. Did you really think they gave you that discount card to save you money? Think again. With a discount card they have your demographic data and track every purchase your make whether you have a credit card or not. Using these cards stores determine what products are selling. They determine which aisles are working. They determine which shelf is going gangbusters. And, like advertisers, they charge their suppliers to get their products in those aisles and on those shelves.
So, in education, we do the same right? We track student scores over time. We look at how they behave in a certain class. We look at interactions between students and teachers. We calculate time on task every period. We assess how one student relates to another. We look closely at whether they are late or early to class or school. If they are absent, we tabulate the reasons. We factor in how many minutes or days they spend in In-School Suspension. We look at whether they completed all of their homework, half, or none. We’ve even looked at whether their parents are involved in their schooling or not. And once we’ve looked at all these data points throughout the school year, we determine whether the student is working or not, learning or not, behaving or not.
What’s that you say? We don’t do this? Well, pray tell, what do we do?
We look at one test on one day for an entire year.
Let’s flip that around for a minute.
Let’s say that Nielsen decided to measure a show’s impact on one night only rather than an entire season. Let’s hypothesize that Tivo and Hulu will do the same. Or that grocery stores will determine the impact of their store layout by calculating customer purchases on one day out of the entire year.
Kind of silly isn’t it?
Don’t misunderstand me. I’m not saying to throw away standardized tests. They tell us a lot. They can help us determine what we should teach or reteach. They can show us if a student “gets it.” We are doing this in the classroom with tests, quizzes, warm-ups, and more. We are collecting many, many data points and using them to fine tune our classroom instructions. We need these tests, but we need more as well.
If we truly want authentic assessment that measures not only the learning of children but the effectiveness of teachers, we are going to need a lot more information than what a standardized test can give us.
What is your opinion? What data points would you gather to assess student learning? Teacher effectiveness? Leave me a comment.
