By Kristin
The last three posts on this blog have responded to the Federal Government's warning that unless Washington uses test data as part of a teacher's evaluations, we no longer meet the waiver requirements for the Elementary and Secondary Education Act (ESEA or No Child Left Behind).
In no moment of my professional or academic life has one test been used to measure my growth. The biggest assessments I've sat in my life – the SAT, the GRE, my National Board tests – were used to measure whether or not I was ready for the next step of my career.
Why are some groups still insisting this is the best way to measure a teacher's impact on student growth? It is so misguided.
This beneath-the-radar presentation does a beautiful job of explaining the strengths and weakness of two most commonly used strategies to measure student growth with one sit-down test. Read it. It's short, and those of you who are teachers will appreciate getting information from a power point without having some presenter read every word aloud to you.
The final slide is too good to not copy:
Balance Test-Based and Non-Test-Based Measures in
Accountability
• Test-based measures serve an important signaling function
and provide information about a key outcome of schooling
• Non-test measures can facilitate better decision making by
reducing overemphasis on a narrow set of standardized test scores
• Determining how to combine and weigh various measures
requires both empirical and judgmental considerations
Those who are making policy decisions 30,000 feet above the classroom and its students are failing to understand the last two bullet points.
It's good we have the data to reveal that our students who live in poverty, or whose families are affected by generations of racism, are not being as well served by our schools. It's like a satellite image of Brazil that reveals catastrophic deforestation. But what policy makers have failed to do is set the satellite photo down and get to ground level. Instead, they're saying things like, "Brazilian farmers are bad farmers and don't know what to do, so let's take another photo next year and see how they've done at reducing deforestation. Let's measure forested land and then either reward them or punish them."
When I measure the growth of my students, I look at evidence through time. I lay work they did in September next to work they did in May and June. I measure my impact by looking at the evidence between September and June, because there's nothing I can do about the previous June, July, or August. Using only the state test to measure my impact means I'm held accountable for months I didn't see the child. It's not accurate.
And there's growth I'm responsible for that OSPI and the Department of Education don't seem to care about. I've helped thousands of children learn to enjoy reading. I've helped children have faith in their intellect, and confidence in their abilities. I've helped children learn conflict resolution, how to interact with adults, and how to advocate for themselves in a confusing system. I've helped them find their voice.
The Department of Education and some policy makers need to quit fooling themselves into thinking a one-test snapshot shows a child's growth in a way that can be attributed to one teacher. It's critical to the health of public education that teachers be expected to do great work but also trusted to do great work. We can gather evidence to measure and demonstrate student growth, and if it's worth doing, it's worth doing right.
I hope Washington policy makers can advocate for the children in their state by making good choices in terms of evaluating teachers and measuring student growth.
RTTT grants are a little different. With those things, you have to actually spend the money on the stuff you promise you’re going to spend it on.
Federal education funding is essential, especially for folks like me who teach in a Title 1 school. That money is going to what we already do; it’s already budgeted. Losing that money is simply not on the table.
18% huh? That would be a tough sell. In a semi-related situation I was supportive of teachers who were seen as sabotaging our chances for RTTT money because they resisted evaluation via test score “growth”. With dollar signs dangling from Arne Duncan, the typical public response was we should stop the grown-up arguments and do what was best for the kids by pursuing the grant. When the total influx of money was going to be less than a 1% increase, and half of it destined for administrivia anyways, it was easier to resist. I don’t know that I could stand by my go-ahead-and-crash-it rant if the budget cut went that deep.
I agree with David. Washington may have to forego our waiver and bide our time until NCLB is redefined. Or, a new state-level testing system needs to be designed, with a test at the beginning, middle, and end of the school year, for all grade levels and for all subject areas. But who is going to pay for that?
Touché
@ Tom… except when they don’t.
I’m with David. NCLB was well intentioned, but creating a bigger bureaucracy has not resulted in more student learning and it has created a whole industry that sucks scarce resources from children. I am in favor of high standards, local decision making and common sense.
That would be fun, David. Sort of like taking a parked bulldozer out for a joy ride. (yes, I have, but only once) Unfortunately, we’d lose 18% of our school funding as a consequence, and ultimately we’d end up capitulating.
I think I see how this ends: we do what we think is right and trust that our test scores somehow validate our work as teachers. And generally, they will.
I’ll read your rants any day, David, because at least I know they are coming from someone who knows teaching, schools, and students. I have to admit, the whole natural-consequences approach is appealing, but we all know who’d really be suffering in such a scenario. But, maybe that’s the only way…
Another issue is the acceptance of last year’s test score as a valid “pretest” with which to compare this year’s score.
I teach fourth grade. I have also taught third grade. The third grade math test covers different skills. You can’t just compare the two because they’re both “math tests.”
This whole issue just infuriates me!
Rant coming. Here’s a idea that, while it seems a bit irresponsible to say it, probably has as good a chance as any in advancing our work in schools. I think we need to give up on the federal government – let NCLB crash and burn big time. No waivers, just a whole huge mess of stupid federal policies that are doing nothing for anyone. Let them bring the consequences down on everyone everywhere and see if that really convinces anyone that the federal government has the first clue what it’s doing. Then, fight the good fight on the more local level, and (as much as I’d prefer it were otherwise), maybe we make ed. policy into us vs. them, state vs. federal, and rally our communities to recognize and resist the problems from Washington. A costly approach? Perhaps. Painful? Perhaps. How does it compare to what we’re going through now?
So how do we either (1) get the feds to see that our way is the right way OR (2) move on without the feds’ blessing? Not hypothetical… I really want to know the answers to these.
Keeping it close to the classroom has got to be one of the best ways to look at growth that might be associated with a classroom teacher, and you’ve described it so well here, “I lay work they did in September next to work they did in May and June. I measure my impact by looking at the evidence between September and June, because there’s nothing I can do about the previous June, July, or August.”
As I go back to school this year, I’m looking forward to seeing the growth my students make over the year in my classroom! (I just wish the US Department of Education was more concerned with this as well!)