We can differentiate assessments into four different types. This is a macro view of assessment, a first step in crafting an authentic, research-guided approach to assessing your student’s learning. Assessments all fall into two categories. Within these categories, there are further subcategories (low-stakes, high-stakes, academic, authentic, etc.), but all assessments are either summative or formative.
Formative assessments have a primary goal of improving the depth and quality of student learning, not to provide final evidence of learning or yielding grades. You will use these assessments to inform your teaching and to provide feedback to students. Students will use these assessments to inform their own view of their mastery and understanding of the subject matter.
Some examples of formative assessments for the virtual classroom:
- Pre-tests of foundational knowledge
- Muddiest Point discussion forums
- Draft submissions of project sections
- Peer Review
- Self-check quizzes
Dylan Williams, a leading researcher in the field of assessment describes formative assessment like this; “Formative assessment describes all those processes by which teachers and learners use information about student achievement to make adjustments to students learning to improve their achievement.”
A summative assessment has a primary goal of providing proof of learning and assigning a grade or defining the level of knowledge a student has attained. You will use these types of assessments to decide if the student has achieved mastery of the course material. Students will use these assessments to judge whether they are capable of continuing in the field of study or similar high-‐level decisions. These assessments do not generally offer opportunities for revision or improvement.
Some examples of summative assessments for online learning;
- Research Papers
- Problem Sets
To get further perspective on the difference between summative and formative assessments, please watch the video interviews with Dylan Williams and Rick Wormeli.
Another way to differentiate assessment types is by relative value vis a vis their effect on a student’s ultimate assessment, their grade. These two categories are:
- High-stakes assessments
- Low-stakes assessments
What is a high-stakes assessment? High-stakes assessments are defined by value more than type. They have four major characteristics.
- Given infrequently
- Is a single, defined assessment
- Has obvious, significant consequences for both success & failure
- High value relative to student’s final grade
- Summative in nature
- Feedback, if given is not timely
For example, if you assess your students with a midterm and a final exam that together yield more than fifty percent of a student’s final grade, even if students complete other forms of assessment, you are using high-‐stakes assessments because failure on these assessments will certainly result in failure of the course.
High-stakes assessments are in the news a lot. Standardized tests like the SAT, Virginia’s Standards of Learning (SOL), and the New York Regent’s Exam are high-stakes assessments that can have profound negative effects on students' lives. These exams have spawned a nation-wide, $4 billion test preparation industry (Wired, 2013) founded on students’ (and their parents’) fear of failure on these exams.
This is important because fear is a primary motivator for students required to take high-stakes assessments. They will do whatever they need to do to succeed because there is so much at stake. This includes spending $200/hour on test prep classes, cheating, plagiarism, and cramming, which get them through the exam successfully but may not yield a student who actually knows what the test claims to prove they know.
Engineering students are the second most likely group of students (business students, not a shock, are #1) to engage in cheating (McCabe et al., 2001) primarily because of the discipline’s reliance on traditional high-stakes assessments in combination with the difficulty of the material. If you have doubts about the prevalence of cheating in higher education, read The Shadow Scholar, an article published in the Chronicle of Higher Education.
High-stakes assessments commonly used in higher education like finals, midterms, term papers, theses, and dissertations are not surprisingly the source of many academic integrity problems.
High-stakes assessments are also more likely to result in the practice of “curving” grades, either on the exams themselves or on the students’ final grades. In an applied, competency-based program, like EP, this is seriously negative practice because a student who has earned a ‘D’ has not achieved mastery, even if that ‘D’ is changed to a ‘B’ because the student did better than the average student. The result is a student who does not actually know the material passed forward as ‘competent’ because they did not fail quite as badly as his or her peers. Grading curves also foster a competitive learning environment as opposed to a collaborative one.
Like the high-stakes assessments mentioned above, low-stakes assessments are defined by their relative value in relation to a student’s final grade. They also have certain recognizable characteristics:
- Given frequently
- They individually have a low impact on the final grade
- Formative in nature
- Feedback is timely
As championed by Dr. Scott Warnock of Drexel University, Frequent Low States (FLS) assessments are about feedback. By giving more frequent assessments you dilute the impact of any one grade (good or bad) and reward consistent quality efforts. While high-stakes testing; “discouraged teachers from using strategies which promoted inquiry and active student learning…” FLS methods do the opposite. FLS approaches foster transparency in grading and a collaborative academic atmosphere.
McCabe, D. L., Treviño, L. K., & Butterfield, K. D. (2001). Cheating in academic institutions: A decade of research. Ethics & Behavior, 11(3), 219. https://doi.org/10.1207/S15327019EB1103_2