Using summative labs to demonstrate science skills

As our year draws to a close and I reflect on my math and science curriculum for next year, I can proudly say that one of my best achievements this year is designing some good summative lab assessments. Summative labs are more authentic, or performance based assessments that are more demonstrative of science skills under the strand "Scientific Inquiry" or "Nature of Science" rather than content knowledge that is specific to life, physical or earth science. I've toyed around with them over the last couple years, but as I've unpacked my curriculum I feel like I've been able to design some interesting assessments that target the learning standards in the nature of science. Some examples are 
  • Frame and refine questions that can be investigated scientifically, and generate testable hypotheses. 
  • Design and conduct investigations with controlled variables to test hypotheses. 
  • Form explanations based on accurate and logical analysis of evidence. Revise the explanation using alternative descriptions, predictions, models and knowledge from other sources as well as results of further investigation.
Just to name a few. What I love about standards in the nature of science, is that they're common throughout all units of study and they're really the behaviors that we want students to be able to demonstrate to say that we are "building scientists" and not merely regurgitaters of facts. The problem that I had in the past was that I felt that I merely gave these standards lip service or when I assessed it, it was not very accurate and honest about a students meeting that specific target. Take this formative lab on measuring soil porosity:

It definitely served it purpose of developing understandings of how sediment size affects drainage, but the indicators were difficult to ascertain. As students work in groups, they often share the same understandings and misunderstandings. Often groups have the same answers for certain prompts so I don't know what they really learned as a result of it. The real learning came from debriefing the assignment later.

In my unit "Into to Chemistry", the students engineered a distillation apparatus to purify water and try to solve the world's UN millennium problem of providing potable water for impoverished people. This was our end of the unit project, and by then, students had a better grasp on the nature of materials, matter, and the scientific method. The website that sponsers this is CIESE and they have a ton of great collaborative STEM projects. I've done this one, "The International Boiling Point Project" for the last two years:

This was a good step on the road towards mastery, but it was still a dubious indicator of learning as it spanned several classes as students built a prototype, learned how it held up under stress and through trial and error, learned how the nature of materials determined their use.

I've read about summative labs and done a ton of reading about them, but I hadn't seen many tangible examples of what they look like. Usually, they are open-ended and give students the freedom to show their understanding through a number of ways. I wanted students to have the freedom to be able to choose a question that related to the science standards. I decided that I'd offer three questions, differentiated on difficulty that students could design an experiment around on their own. They were:

Does food coloring spread faster in salt water
or fresh water?
Does the temperature of a rubber band affect
its elasticity (stretching ability)?
Does stirring speed affect the amount of salt
that can dissolve in water?
Which material (newspaper, plastic wrap, or tin
foil) is the best heat insulator?

I did not tell students what the questions were before they came into class as this "lab" was a test of their science skills. Because of this, they could not receive any help from outside tutors or friends. They were on their own. I told them that there were many different ways of investigating these questions, and we'd celebrate our different approaches afterwards. After entering class, the students selected a question that they felt they could investigate confidently and this was the scene: 

If you noticed, it was very quiet. Students had the ability to roam around and select equipment for their investigation, but most students were minding their own business. They knew other students may have different variables, different questions, so it was silly to try and copy someone else's work.
I developed a rubric and had a generic cover sheet that outlined the steps of the scientific method I thought best demonstrated the standards. Many students tell me that summative labs are quite fun because they actually feel like scientists. However, they usually supplement unit tests so they can demonstrate their knowledge and understanding of key concepts within a unit.

I'm tinkering with my summative labs for next year along with the assesments of them, but feel I'm on the right track. I'd love any input, criticism or advice if any of you have used similar assessments in your science curriculum. We're all learning!

Popular Posts