« The Iraqi Study Group Report | HomePage | IE7, Google Desktop, and MSHTML »

Thursday, December 07, 20061165537500

Classrooms Evolved, Part I: Traditional Methods

The most common style of teaching ignores my definition of a class Student society, and student concern for their grades, are sidelined throughout the class hour. A professor will stand up with a lesson plan, often written years ago, and talk. For an hour or two or three, depending on the class, students just sit and listen. Students are essentially uninvolved with the lecture, except for the mental work required in paying attention. Then after class, students (supposedly) study, typically individually, and for most classes ineffectively. Standard ways of grading in these classes, such as exams and essays, individual and group work, do not help much.



Multiple-choice exams are not effective. Doing well on such a test requires the student to memorize terms for a tremendous amount of material. “A” students learn to sacrifice understanding to breadth. Students do not leave the class any smarter, or even more able to use concepts in conversation, let alone in life. This sort of studying is exhausting. Worse, it encourages lazy teaching. "It's in the book" is a reflexive answer to a student complaint. It is hard to see multiple choice exams, as they are regularly given, as anything other than a rotten instrument.


Yet essay tests are counterproductive too. They are more useful because they require linguistic reconstruction of information, and this writing makes writing both more cognitively valid than multiple-choice exams. But essays encourages lazy teaching. By essentially punishing the teacher for giving this sort of work (it takes h o u r s to grade these things), essays act as constant reinforcement against instructors who assign them. Standard essay tests punish the professors who use them. Additionally, the temptation to use a rubric (and sacrifice measured comprehension for something mindlessly easy to grade) becomes too great – and at that point, why not just use a multiple-choice exam?

Assigning individual work is full of dangers. Private assignment deprives scholarship of a social component. The human animosity towards isolation can keep student from putting in the necessary time on individual projects. Likewise, individual work channels students in their own ignorance, too often making the assignment either a desperate example of fumbling in the darkness. Comprehension is ignored in favor of meeting the teacher's requirements.

Not that typical group work is much better. Until graduate school I despised group work, because such assignments really meant that high-performing students would just carry the load for low-performing students. Group work without accountability was some sort of prisoner's dilemma game, with one player being both unwilling and unable to meaningfully cooperate. Indeed, I think many teachers' overemphasis on individual work comes from a learned aversion to poorly contributing group-members.

So what is needed is a grading system which avoids the memorization of practice exams, the grueling mind-numbingness of essays, the isolation of individual work and the free-riding of group work. Thus must be done in the context of a teaching style that embraces society. It requires a new style of classroom – a style I have been developing for years.




Classrooms Evolved, a tdaxp series
1. Traditional Methods
2. Social Grading
3. Deliberative Learning
4. Overcoming Doubt
5. Conclusions

Comments

Excellent post Dan. Some comments.

"“A” students learn to sacrifice understanding to breadth"

All too often and it is a terrible waste of talent, keeping the brightest operating at the lowest cognitive common denominator, except with greater accuracy over larger amounts of data.

" A professor will stand up with a lesson plan, often written years ago, and talk"

Ah the yellowed notes of the soon to be emiritus...LOL! True. Lectures are useful when used sparingly and with a cognitive objective in mind beyond conveying information, which usually means some student-teacher engagement beyond talk-listen-record.

"Multiple-choice exams are not effective"

No but they are cheap substitutes for a mass educational system that does not want to pay for real teaching or education vs. inculcating socialization norms.

"By essentially punishing the teacher for giving this sort of work "

The punishment is actually in the high student to teacher ratio that mitigates against all kinds of useful and productive interactions, except on an arbitrary basis. It's hard to get to know new 200 people well every year much less every semester

"Assigning individual work is full of dangers"

Depends on who is ready for it but feedback needs to follow

"Group work without accountability was some sort of prisoner's dilemma game, with one player being both unwilling and unable to meaningfully cooperate"

And the number of professional educators who will viciously attack you for making that truthful observation the moment you begin to identify demographics, are legion.

Posted by: mark safranski | Friday, December 08, 2006

Much of what you say is true, but you're ignoring reality -- that is, there aren't workable alternatives when you have 240 students in a class. As for multiple choice tests, you're overgeneralizing. The quality of a multiple choice test as an assessment tool depends entirely on the quality of the questions, and has nothing to do with the test format.

Posted by: rightwingprof | Friday, December 08, 2006

Mark & RightWingProf,

Thank you both for your comments.

"And the number of professional educators who will viciously attack you for making that truthful observation the moment you begin to identify demographics, are legion."

I wasn't thinking of any demographics in particular while I wrote those, but I suppose you are right.

Stay tuned for "Wary Guerrillas" (the project formerly known as "Systems Administration for Phenotypes"), which will begin immediately after part V of this series. It looks at the demographics & implications of cooperation.. and comes to some interesting conclusions.

"The quality of a multiple choice test as an assessment tool depends entirely on the quality of the questions, and has nothing to do with the test format."

Essay and multiple-choice exams are different creatures that require different conditions. Broad and vague questions can suffice for essays, because you are measuring a student's writing and thinking ability. However, questions have to be more carefully designed in multiple choice exams.

One way of thinking about the difference is if you want to spend more time in prep or more time in evaluation for a test. Essays save you prep time, while multiple-choice exams save you evaluation time.

Posted by: Dan tdaxp | Friday, December 08, 2006

"One way of thinking about the difference is if you want to spend more time in prep or more time in evaluation for a test. Essays save you prep time, while multiple-choice exams save you evaluation time."

Time is an important consideration, but the major problem with essay and short-answer tests is objective assessment. If you have only one class of 30, this isn't too much of a problem. If you have four classes of 30, it becomes a serious problem, because even the same instructor drifts while assessing exams.

You could, of course, assess these exams much like ETS assesses writing: with multiple readers calibrated to a single rubric and set of texts, where second readers do not know what score the first reader has given, and any paper given two scores more than one apart must be evaluated by a third reader. However, this is very time and work intensive, and few faculty would be willing to do this.

And even if you are objective (for the sake of discussion), giving essay and short answer tests opens you to the charge of subjectivity, a charge that is difficult to fight on today's campuses, more difficult if you have a spineless chairman or worse, dean who is more interested in maintaining political ties than standing behind his faculty.

I have five classes, each with 240+ students. Do I give essay or short-answer tests? God no. I don't have the time to grade them; 48 hours after the final has been given, I have to turn in final grades. It's not feasible. Even if I wanted to, I couldn't, because it's a departmental final.

I'm merely pointing out that essay and short-answer exams have their own share of problems.

Posted by: rightwingprof | Saturday, December 09, 2006

Right Wing Prof:

I agree with you:

"Essays encourages lazy teaching. By essentially punishing the teacher for giving this sort of work (it takes hours to grade these things), essays act as constant reinforcement against instructors who assign them. Standard essay tests punish the professors who use them. Additionally, the temptation to use a rubric (and sacrifice measured comprehension for something mindlessly easy to grade) becomes too great – and at that point, why not just use a multiple-choice exam?"

Posted by: Dan tdaxp | Monday, December 11, 2006

Post a comment