ASSIGNING GRADES: THE PROCESS OF ASSESSMENT

With the advent of standards, which are statements constructed by authorities that outline the specific outcomes for a class, a new era of testing is emerging for teachers and learners. With standards, tests are based on acquisition of the essential content for each grade level. Standards based learning is different from traditional "norm referenced" learning. With norm referenced tests, the class average becomes the central system for rating the learner with others in the class but fails to compare the child on a state or national scale. Most states have some system of large scale testing that shows individual progress. Individual teacher made tests promote arguments and confusion.


We have already noted that some of the imprecision of the testing process may be eliminated through the development of good objectives. A good rule of thumb is to test for achievement of objectives. Poor teachers miss on this, causing ill feelings on the part of students and poor course evaluations.
Before noting some approaches to assigning grades and assessment, it may be useful to reflect on the grading, testing and assessment process. The following questions are suggested for your consideration:

  • What is worth knowing?
  • Should one test for what is known or unknown?
  • What kinds of tests are the most effective measures of achievement?
    • essay
    • multiple right
    • contracts
    • multiple choice
    • mastery learning
    • true/false
    • pass/fail
    • anecdotal records
    • completion
    • problem solving
    • normal curve
    • matching
    • self grading
    • mastery learning
  • What are the causes of grade inflation?
  • What is the difference between testing and grading?
  • What is the relationship between objectives - methods, testing & grades?
  • Is there any relationship between good grades and learning?
  • Quality vs. frequency. Which enhances learning more?
  • What is the purpose of testing?
  • What responsibility does learner/teacher have in evaluation?
  • What is the relationship of time on learning?
  • Can letter grades appropriately measure "learning"?
  • What follow-up to a test is appropriate? Can a Gausian curve (for bio material) be applied to a human designed test?
  • Should students be allowed to repeat a course until "perfect" works are earned?
  • Can rank in class really mean anything?
  • Is an A at Harvard comparable to an A at Boise State? What are the variables?
  • Comment: "If the process by which any product is produced be faulty, then the product will be faulty."
  • If it isn't to be tested then is it unimportant?
  • Should we test for writing, thinking, or creativity?
  • How many tests are right?
  • How does the test type determine study role?
  • What is a learning curve?
  • What is meant by test sophisticated? Is this true of ACT, CEEB-GRE, LSAT, MCAT?
  • What is the instructor's role in testing? "Instructor autonomy?"
  • What would be the results of national tests in different subject areas?
  • Is it common to select courses/instructors to enhance grades?
  • Should students be required to frame the important questions rather than answers?
  • How should one be different after having taken a course than before it was taken?
  • What kinds of questions are asked on GRE-type tests? - spatial aptitude and analogies.

Comment: "The major emphasis in American Universities is on knowing facts."
Many universities post the traditional percentage grading values in their public documents. The general scheme often looks like this:

90% - 100% Excellent - A
80% - 89% Superior - B
70% - 79% Average - C
60% - 69% Poor - D
Below 60% Failure - F

For any type of assessment a percentage grade may be assigned, even when quantification can't be easily used as in the case of essay or performance based examinations.


A variation of the percentage scheme often found in university settings is the normal curve grading system.


This is the system whereby the instructor determines the percentage of A's, B's, etc., according to a pre-established distribution based on the Gausian curve: A's = 7%; B's = 24%; C's = 38%; D's = 24% and F's = 7%.


Two approaches have emerged in recent yeas that are receiving attention through practice and discussion. These are portfolio assessment and mastery or performance assessment.


Mastery learning is based on a preset list of competencies that students should know or be able to do. Various schemes may be used to establish a final grade; usually the attainment of a particular number of competencies (I call this the merit badge approach). Performance or mastery is the chief criterion for success.

Portfolio assessment represents a collection of evidence that convinces the instructor that the objectives of the course have been met. The following notes outline the main ideas of the portfolio approach:

1. Portfolios are well developed models of evaluation in other disciplines (art, photography, architecture) and programs (Scouts - done with a mentor; awarded with ceremony).


2. A portfolio is a collection of evidence that constitutes a compelling argument that you have achieved or made progress toward a goal. (Evidence tells anyone who may see your portfolio that you have learned something - always articulate the relationship between evidence and purpose).


3. Ask yourself "What evidence will show that the purpose of my portfolio has been met?" Types of evidence may include:


4. The organization for the portfolio should include:


5. With portfolios you know what is expected as you begin -- the system's fair, reasonable, dynamic, it displays strengths, multi purposes and is integrated.


6. I will provide you with additional information as needed particularly the team aspects of verifying evidence via the value added process. (What will the assessor know about me from this piece of evidence that would not be known if I hadn't included it?)


Used by permission: Copyright © 1998, 2001, 2004 Dr. Terry Armstrong, Dr. Mitchell Klett, Dr. Scott Graves: Idaho Virtual Campus