Rubrics

We’re all being pressed to use rubrics. For those of you not in the ed biz, a rubric is described pretty well in Wikipedia:

A rubric is a scoring tool for subjective assessments. It is a set of criteria and standards linked to learning objectives that is used to assess a student’s performance on papers, projects, essays, and other assignments. Rubrics allow for standardised evaluation according to specified criteria, making grading simpler and more transparent….

Rubrics are generally thought to promote more consistent grading and to develop self-evaluation skills in students as they monitor their performance relative to the rubric.

Clearly rubrics are useful. Those two paragraphs are full of Good Things: criteria, standards, assessment, standardization, simplicity, transparency, consistency, self-evaluation skills. Everyone is in favor of these, so everyone is in favor of rubrics.

We generally use rubrics to determine partial credit on math tests. Instead of subjectively deciding that one student’s work is worth 10 points out of 16 and another student’s work is worth 12 points, a teacher just consults the rubric (usually developed in conjunction with other teachers on the team):

4 pts for stating appropriate law, 4 for finding one angle R, 4 for finding one PR, 4 for other;  –1 pt for less precision than requested, no penalty for extra precision.

There are still unstated but implied decisions to be made. For instance, if the student is to get 4 points for finding one angle but makes a mistake along the way, is it a 2-point mistake? Nevertheless, the rubric is extremely useful and helps to promote fairness. More importantly, it promotes the perception of fairness.

In math, however, we can’t usually share the rubric with students in advance (except perhaps on projects), since the rubric itself gives away the solution method. Having the rubric in advance could reduce the problem to mere rule-following rather than thinking. As a result, we can’t “develop self-evaluation skills in students as they monitor their performance relative to the rubric.”

Rubrics are also used effectively by the College Board in scoring answers to open-response questions on Advanced Placement exams. Experience shows a very high degree in reliability among the numbers given by different well-trained scorers to the same response.

But there are some downsides to rubrics. My upcoming post on Making the Grades will explore some of the them. An additional problem not covered in that book is that the current push for rubrics includes some preposterous goals. For example, here’s what the New England Association of Schools and Colleges lists as #1 among the 2011 accreditation standards for “Assessment of and for Student Learning” :

The professional staff continuously employs a formal process, based on school-wide rubrics, to assess whole-school and individual student progress in achieving the school’s 21st century learning expectations.

“School-wide rubrics”? There’s the rub. If a rubric is general enough to be used school-wide, from math to art, from history to science, then it’s going to be so general as to be meaningless. If a rubric is going to be useful to a math student in assessing his or her work, it has to be specific to math.



Categories: Teaching & Learning