Using Standards-Based Grading to Assess Labs
I don’t like grading lab reports, and I especially don’t like grading them when I get the sense that the lab report is not adding to a student’s content knowledge, lab skills, or communication skills. For years I did lab reports in a pretty standard way: I’d introduce some equipment ask the students to make some observations, ask what they could measure, refine the list of variables, discuss a procedure, and set the students loose to gather and analyze data. Afterwards, the students would share their results and through our discussion we’d develop the important ideas and definitions. Students are sent home with a rubric to eloquently summarize what happened in the lab and what they learned about physics. — and then I’d end up with a pile of grade reports to assess that generally didn’t accomplish what I was looking for. They regurgitated things without thinking, didn't dig into the analysis like I asked, skipped important points, and spent way too much time on making everything pretty. My grading was filled with many cringeworthy moments. I wrote futile comments and suggestions that would be long forgotten by the next lab report, if read at all, as students looked only at their grades before stuffing them in their backpacks.
I’ve been using standards-based grading in my course for five years, and I wanted to meaningfully extend it to laboratories. After several iterations, I think I’m headed in the right direction.
I chose to assess lab skills in labs, and to assess physics content through other assessments. There are many times, of course, when the two complement each other so well that I assess both in the same lab report.
I’ve been thinking about the lab skills I want students to acquire and demonstrate. I consulted existing standards, particularly the Next Generation Science Standards’ Science and Engineering Practices, which provides a pretty comprehensive starting point. For practical purposes, learning objectives for SBG need to be fairly broad so that the number of objectives doesn’t become overwhelming. As you develop objectives for your students, decide what’s really important for students to be able to do, and then design your instruction, labs, and assessments to scaffold the students in meeting those objectives.
I originally used just three lab objectives, and for more advanced students, this may be a good starting place.
Lab.1 I can conduct an experiment, record data, and distinguish between independent, dependent, and controlled variables.
Lab.2 I can analyze and represent data with graphs and equations.
Lab.3 I can construct explanations and models to interpret patterns observed in the lab.
For my 9th graders, I found that breaking down the objectives into smaller skills has helped enormously. I’ve reorganized my objectives several times and am currently using the following:
Lab.1 I can design and communicate data collection procedures with well-explained and labeled diagrams, distinguishing between independent, dependent, and controlled variables.
Lab.2 I can conduct an experiment and properly record qualitative and quantitative data.
Lab.3 I can represent data with graphs, linearizing as needed.
Lab.4 I can write the equation for the trend in lab data, using variables and units appropriately.
Lab.5 I can explain the relationship between variables and explain any physical significance of the slope and y-intercept.
Lab.6 I can make a scientific claim, support the claim with evidence, and provide the reasoning that connects the claim to the evidence.
Lab.7 I can quantify and explain the limits to my precision in my data collection and analysis.
I’ve listed them in lab-report order, but I don’t assess them in this order, as the first one is one of the hardest and takes the longest to learn. I model how we design experiments, but I’m very much guiding students to a particular procedure. Later in the year, there are a number of labs that lend themselves well to saying: here is what we don’t know, how are we going do do a lab to figure it out?
Lab.2,3,4 are the objectives students first learn in our kinematics and force labs. Mastery involves establishing consistent conventions and developing quantitative analysis skills. For older students, these probably don’t need to be broken into separate ideas, but I find it helpful to split up these tasks for 9th graders. After a few labs of discussion and guided practice, on a subsequent lab, we’ll do our typical pre-lab setup, and then I’ll dispatch students in their groups to demonstrate that they can conduct the investigation, record, graph, and mathematically represent the data — for credit. We then come together to interpret the meaning of the data, modeling a skill that will later be assessed. By the time we are developing quantitative expressions for calculating energy storage, I ask students to design an experiment to determine how to calculate kinetic energy. Electrostatics, electric circuits, and light, provide rich opportunities for developing models from largely qualitative observations and data through the claim-evidence-reasoning format. The practice for demonstrating this skill first comes through a qualitative introduction to energy lab and a conservation of momentum lab. Although we referred to uncertainty and precision, I never took the time to really ground the students in this concept, so I never assessed Lab.7. This objective fits so naturally into chemistry labs, I tend to kick that can down the road to next year’s science class. I also wrote an eighth objective that I didn't assess: Lab.8 I can describe further investigations that may clarify, refute, or expand my claim. This lends itself to sophisticated investigations, but can also be used in introductory courses. I try to introduce this idea so that it can be built upon in future courses.
This is an entire lab report! How does this make my life easier? [] The objectives signal to students the important ideas that will be seen throughout the year, so rather than ignoring a comment about how to set up data tables or graphs, students recognize that they will need that skill in upcoming labs, and that it’s worth figuring out how to execute the skill and earn credit for it.
[] Rather than nickel-and-diming students based on a rubric, I’ve placed the job on the student to convince me that they understand the objective.
[] The labs serve to develop key content for the course, but I assess content problem-solving understanding separately. This allows students to truly follow their data to logical conclusions, even if it isn’t “right”. We spend plenty of time in class debate after a lab to come to a consensus understanding, and the best ideas with the most support rise to the top.
[] I’m generally assessing only a few objectives in each lab. This helps the students to focus their preparation, it keeps the activities short enough to be completed in class, and it keeps the grading manageable. I ask myself, "did this student convince me that they can perform this lab skill?"
[] When students work together, the discussion between peers is dynamite. They work to convince each other that they are on the right track, and their arguments are rooted in the data they collect.
[] Students complete the labs for assessment in class - this helps to nip the issues with tutor-written lab reports. I do have the students work to complete the lab assessments in groups because it builds positive interdependence in the lab teams. As lab tasks get more difficult, the only way they can really accomplish it is to work as a team.
This feedback is so much more useful to the students than a letter or percentage grade. When they recognize they did not fully demonstrate their proficiency on an objective, they do go back to the notes I've made on the lab report and try to figure out how to accomplish that skill satisfactorily. They also know that these objectives will be assessed again in future labs, so it is to their advantage to ask questions and seek help in order to become proficient in that lab skill. By the end of the year, I feel that my students have not only developed these foundational lab skills, but they have developed a larger picture view about how the various steps in an investigation tie together and reinforce one another.
Originally posted on my previous website July 18, 2019.