Smarter Tests Make for Smarter Schools

Our current standardized, high stakes tests do a disservice to our students.  They don't measure student growth and they penalize students who develop at a rate other than what society deems to be the norm.  We should get rid of those tests and replace them with a testing system that actually measures student progress from year to year rather than measuring if schools can meet basic minimum thresholds.

A testing system that measures progress  would help to better identify teachers and practices that have value.  It would also help students see how their learning builds from year to year.  It would provide relevant and meaningful feedback for teachers.  In short, it would provide better data from which students, teachers, administrators, parents, and the greater community could make better decisions.     

I like one particular option that creates a testing system similar to that of the GRE.  It would be computerized and would work to locate the level of the students knowledge by adjusting the difficulty of questions and concepts based on student responses.  To do this the test would need to take the skills necessary for success in a discipline and break them down into their component parts.  Those parts would then have to be categorized in terms of difficulty from the first testing year up until the last testing year.  For example, all the math concepts taught from say second grade through twelfth grade would be included.  Multiple question types would need to be created for each concept and then all of them would be compiled into the program.  Students would take the test from the same program every year.

Right now, the current tests don't measure student growth for smart students very accurately so we don't have a picture of how much they actually learned from their teacher.  Because each test given each year is different, you can not measure student growth from year to year with much accuracy.  By giving the same test, year after year, you start to get a picture of students learning growth.

I am looking forward to the day that education starts getting smarter about how it collects data.             

 

Teaching is about more than Tests

There is a lot of talk about moving part or all of teacher evaluations towards a value added measurement system in the education world. These value added scores would be calculated using the state exams, which here in Virginia and in many other states are almost entirely multiple choice exams. I really hate that we evaluate the knowledge of students based on their ability to pick between the letters "A", "B", "C", "D".

My issue with evaluating teachers in this way is that it fails to take other factors into account.
Justin Baeder summed this idea up best today in his blog post called, Collecting the Wrong Data: Fundamental Attribution Error in Teaching Quality,

"We don't collect data on how many subjects someone teaches, how little prep time they have, how often we interrupt this prep time with meetings and bus duty, and the myriad other non-instructional responsibilities that characterize work in US schools. We analyze test scores and jump to conclusions in an effort to be data-driven, while ignoring perhaps the most important data of all: data on the contexts in which teaching and learning take place."

When the final determination is made on how to evaluate teachers, I hope they make sure to look at the whole package, test scores included.