After looking at the questions each group settled on in Crayfish Structures for Sensing, I could tell my class would need more support to plan effective experiments. I gave my class a related reading informational text assignment, then called each group up one by one to develop questions and experiments. Most conferences took about 15 minutes. In this excerpt from the "taste" group, you can hear (but not see, sorry) a good discussion going on about controlling variables.
This group ultimately decided to place small amounts of several different food around the edge of the tub, and see which food they crayfish had a preference for.
Once each group had the big idea settled, I had them create a shared Google Doc (we are a Google Apps for Education district) that included the question, materials, procedure, and hypothesis. I had them use a Doc so they could work on it simultaneously, but a recorder could have been assigned to write down ideas in science notebooks.
Students actually testing their hypothesis happened the next day to provide time for me to gather questions, and a few teams needed time to revise and clarify their procedures before testing. The most difficult part of their experiments was that they required patience. Most students seemed to expect the crayfish to dart right over to the food, and if it didn't within a few seconds, they'd give it a little bump. I tried to explain that their actions were interfering with the experiment by introducing another variable.
If I were to do this again, I would certainly spend some time preparing them to wait for a result by explaining how scientists in the field might hide out for weeks or months so that they could observe animal behavior in the wild without influencing it.
Once each group concluded their trials and recorded their data, I worked with them to form a conclusion that summarizes their findings, as well as supporting or disproving their hypothesis.
Because we were approaching our annual Family STEM Night (formerly Family Science Night, formerly Science Fair), I had each group display their findings on project boards to show at the event.