5 Formative Assessment
Formative assessments are tests or assignments you give along the way. You can make them high stakes, low stakes, or no stakes at all depending on what you are trying to do at a given moment in the progression of a lesson or a course. The ITIP action of “check for understanding” is a simple kind of formative assessment. Did the class understand what you just said? If yes, proceed to next step. If no, loop back and try it again. If continuous improvement in understanding is your goal, frequent formative assessments are a way to get there. Lots of formative assessment is education’s answer to “test-driven development”. Software terms like unit tests, integration tests, regression tests, and acceptance tests all have their analogs in instructional practice. A unit test is some assessment you do after a daily lesson. Homework assignments are one example. Integration testing is a more comprehensive exam, like a module post-quiz. Regression tests are bigger picture. Think of a mid-term exam. Does the class still remember content from week 1 now that we are halfway through the course? Acceptance testing is basically summative – do you as instructor accept that each student has mastered enough content to receive a passing grade in the course? Would other stakeholders – employers, the school itself, educational funding sources – agree that each student has accomplished enough to justify whatever the grade is? Like we said before, turning in grades on time at the end of the quarter is a core teaching responsibility. How you arrive at those grading decisions is in large part up to you.
There is no law, for example, that says the end-of-term grades needs to derive largely from a high-stakes final exam. For EDI reasons and for more general continuous improvement thinking, there is a lot to be said for calculating the end-of-term grades from numerous micro-assessments that happen along the way. Before the class even starts, you get to set the grading system in the syllabus. Below are two contrasting models.
| High Stakes Final | |
|---|---|
| Final Exam | 50% |
| Other Work | 50% |
| Total | 100% |
| Continuous Assessment | |
|---|---|
| Daily Work | 20% |
| Weekly Quizzes | 20% |
| Midterm Exam | 10% |
| Labs | 20% |
| Personal Project | 10% |
| Final Exam | 20% |
| Total | 100% |
In systems terminology, the High Stakes final model is “tightly coupled”. If anything goes wrong during finals week, it will cause major problems for you, the students, or both. The Continuous Assessment model by contrast is “loosely coupled”. No single exam, assignment, or assessment style is a deal breaker. Under Continuous Assessment, to earn a 4.0, and student needs to perform comprehensively well in all phases of the course. Fair enough. That makes 4.0 an actual accomplishment, not just a pat on the head. But under Continuous Assessment, there are numerous alternative pathways built into the course that will allow a respectable passing grade like 2.0. In case of circumstances beyond anyone’s control, you can even skip the final and just interpolate that part of the grade based on other performance metrics along the way. Continuous Assessment gives you a lot of flexibility in where you put the grading weight, how you address unusual learning needs for specific students, and how you can work around ice storms, pandemic outbreaks, and other challenges life throws at us. From an EDI standpoint, there is nothing to choose here. Continuous Assessment supports diversity at a system design level. A High Stakes Final is going to have binary impacts (up or down) on different types of students, and there will be no graceful way to spin that. If you hand out diversity accommodation bonus points or some other artificial compensation measure on a High Stakes final to make your instructional outcomes more equitable, that leads to a whole other dimension of awkwardness. So just to be clear about this – the royal road to diversity support across your instructional design is diversity of assessment model. Use lots and lots of different types of formative assessments along the way to derive the end-of-quarter summative grading assessment. Students are predictably unpredictable. Just when you think you’ve seen it all, you will surely be facing some brand new challenge! Loosely coupled, continuous assessment instructional design gives you the agility to respond to student learning needs you did not imagine could even exist.