Wednesday, July 29, 2015

How Do You Measure Success?

During last night's #edchat, there was a short discussion about whether those on social media truly implement the ideas they discuss during chats.

I've been thinking about that, and I've also been thinking about how we measure the ideas we employ. What do you do to truly measure the effect of practice you deem profitable?

Not only does measurement help us to identify success or lack of success, but it also helps us to revise our programs for better effect, and it can serve to develop our confidence too. Rather than just subjective response about efforts, good measurement provides us with more accurate response.

This year as I teach math and STEAM as the main subjects of my work, I'll employ the following measurements.

General Math Knowledge
GMADE is a system-wide test we give at the beginning of the year to assess students' general math knowledge and skill. This year not only will I give the test, but I'll also teach the content included in the test as the year progresses. In year's past we gave the test, but didn't pay specific attention to the content, then last year we assessed the test, categorized the information we not included in the grade 5 common core, and made a teaching list for the year ahead. I'll see how students do on the test when they take it again in late winter. Also, I want to look more closely at the validity of this test as some students who score very high at the start of the year aren't able to demonstrate much growth since the test does not extend that much higher than the highest students' skill level. I want to study that with greater depth.

Math Problem Solving
SRSD problem solving. I'll give students an early year assessment and assess their work using the created rubric. I'll continue to assess and coach as the year goes along and use the assessments to inform teaching and practice. I'll assess growth by students early year to late year improvement.

Common Core Math
Common core math will be assessed in a large number of ways. I'll use formative assessments throughout units and a summative assessment at the end of each unit. I'll also give the system-wide end-of-year test and a practice standardized test (PARCC or MCAS dependent on what the State chooses as this year's test venue).

Math Facts
I'll give early year facts test and then set up a responsive learning menu coordinated in part with our RTI efforts, at-home practice, and in-class study. I'll assess student growth regularly in this regard in an attempt to help every child reach mastery. I consider mastery as 100 facts with three or less errors in 5 minutes or less. For fifth grade I expect students to reach this with addition, subtraction, mixed add/subtraction, multiplication, division, mixed multiplication/division, and mixed add/subtract/multiply/divide, and for enrichment I add simple algebraic expressions and other math skill facts.

STEAM
I'll take a look at existing STEAM assessments. I'll use that information to create a STEAM assessment. My main focus in this area is that students are able to collaborate, think/act critically, create, and communicate with STEAM problems. I also want to make sure they are learning the core information embedded in the STEAM activity. I need to think about this assessment pattern.

Too often the way we report on student learning may not be accurate. Even when we use assessments like the ones above, we may not be accurate as there are so many factors that come to play with regard to student learning and effort.

Yet, we know it's important for students to learn well and gain the knowledge that will help them to succeed in tasks as they move forward. My own son relayed a STEAM-like, real-world learning challenge he had at a course lately. I had just read that his course was trying out non-cognitive tasks and I bet this was one of them (it wasn't exactly non-cognitive, but I believe they meant it wasn't purely academic, but more hands-on and collaborative). My son had trouble with the task, and I noted that we're trying to employ more learning related to that kind of task with our STEAM labs.

Back to the point--to do good work, we need to assess well. Our assessments won't be perfect, but in many ways, they'll be better than subjective response alone as subjective response often represents just a few voices and those voices tend to want to show success whether it occurred or not.