Saturday, May 28, 2016
Standards and expectations
Thankfully, good sense prevails once again. Old Man Cambridge is still front and centre with his guidelines for grading GP essays. We just have be more specific about our expectations per question and grade accordingly. Ok, I can live with that.
Thursday, May 26, 2016
Life got no standard
My thoughts on the 'standardization' of grading GP. Old Man Cambridge has already given us a fine set of guidelines that are a broad-based, flexible instrument that easily scales up and down as the need requires. Any tightening of these guidelines second-guesses the original intentions of the source and ultimately turns our subject into an assembly-line grading system, as the sciences and math have been for a very long time already.
If what we are grading is purely information content, then sure, an assembly-line makes the grading process quick and efficient. A tick here, a cross there, a rubber-stamped numerical assessment made assigning a piece of work into one quality band or another. Answers that are either right or wrong, without wiggle-room for interpretation, with the use of the appropriate formula applied correctly lend themselves to be graded via an algorithm. The math and sciences reign supreme here, some papers being done on an optical sheet and graded by computer even. If it can be done by them, then why can't it be done by us, goes the common sneery criticism.
Very simply, in GP we don't just grade for information content. The bulk of our attention goes into each student's ability to communicate their ideas to us, and communication is by nature a subjective thing. If I, the reader, am able to understand what is being communicated, then great! But that does not mean that my colleague will understand the same piece of work in the same way, or at all. That's because our medium of communication, the English language, is a slippery, imprecise means of transacting ideas, with lots of different interpretations and connotations complicating matters. And Old Man Cambridge's original guidelines are made broad enough to accommodate a general sort-of categorizing. But when we have to decide how 12 marks is different from 13, the endless quibbling effectively renders the point difference moot.
One concern that's been buzzing around is that the kids compare their scores now and wail when they got a particular grade while another who wrote the same thing got another. We got this whole grading thing wrong. It's because we made every assignment a high-stakes life-or-death survival Hunger Games type deal, so every kid is desperately scrambling for any point he or she can scrape from the bottom of any available empty barrel. When we stop making every assignment such a mercenary concern, the comparisons will stop. Problem solved. But by standardizing GP grading, we are making each assignment even more high-stakes than ever. Promise equality and we will have to live up to it.
Standardizing GP grading goes by the assumption that we are going to get a lot of similar answers. That could very well happen when we emphasize practicing past exam questions ad nauseum; when we base content knowledge on the same set of 'content packages'; and when we prescribe formulaic approaches in response to identified 'question types'. But if I encounter two or more similar answers in GP, my first concern would not be whether they get the same marks or not. My first concern would be that we have a plagiarism issue. Replicating one another's work is not a trait to be encouraged in JC as the kids will get into serious trouble for it when they get to the Uni. So let's please not start a process that will eventually teach the kids the wrongest of values to bring with them to the Uni.
Besides, it's good to have at least one subject graded on a subjective basis. That's life.
If what we are grading is purely information content, then sure, an assembly-line makes the grading process quick and efficient. A tick here, a cross there, a rubber-stamped numerical assessment made assigning a piece of work into one quality band or another. Answers that are either right or wrong, without wiggle-room for interpretation, with the use of the appropriate formula applied correctly lend themselves to be graded via an algorithm. The math and sciences reign supreme here, some papers being done on an optical sheet and graded by computer even. If it can be done by them, then why can't it be done by us, goes the common sneery criticism.
Very simply, in GP we don't just grade for information content. The bulk of our attention goes into each student's ability to communicate their ideas to us, and communication is by nature a subjective thing. If I, the reader, am able to understand what is being communicated, then great! But that does not mean that my colleague will understand the same piece of work in the same way, or at all. That's because our medium of communication, the English language, is a slippery, imprecise means of transacting ideas, with lots of different interpretations and connotations complicating matters. And Old Man Cambridge's original guidelines are made broad enough to accommodate a general sort-of categorizing. But when we have to decide how 12 marks is different from 13, the endless quibbling effectively renders the point difference moot.
One concern that's been buzzing around is that the kids compare their scores now and wail when they got a particular grade while another who wrote the same thing got another. We got this whole grading thing wrong. It's because we made every assignment a high-stakes life-or-death survival Hunger Games type deal, so every kid is desperately scrambling for any point he or she can scrape from the bottom of any available empty barrel. When we stop making every assignment such a mercenary concern, the comparisons will stop. Problem solved. But by standardizing GP grading, we are making each assignment even more high-stakes than ever. Promise equality and we will have to live up to it.
Standardizing GP grading goes by the assumption that we are going to get a lot of similar answers. That could very well happen when we emphasize practicing past exam questions ad nauseum; when we base content knowledge on the same set of 'content packages'; and when we prescribe formulaic approaches in response to identified 'question types'. But if I encounter two or more similar answers in GP, my first concern would not be whether they get the same marks or not. My first concern would be that we have a plagiarism issue. Replicating one another's work is not a trait to be encouraged in JC as the kids will get into serious trouble for it when they get to the Uni. So let's please not start a process that will eventually teach the kids the wrongest of values to bring with them to the Uni.
Besides, it's good to have at least one subject graded on a subjective basis. That's life.
Subscribe to:
Posts (Atom)