A Plethora of Variables
Behavior (specifically, learning objectives) is a function of what? Robin Barrow1 begins to show how complex the answer can be: "[Learning is related to the teacher's mode of] instruction, [to his] sarcasm, precision, discussion-groups, free and easy classes, disciplined material, disciplined demands on behaviour, humour, kindness, appearance, age, compassion, fear,...[and to factors such as] Henry on his own, Henry in the company of Jane,...in the context of a school like this, a school like that, parents of one sort, headmasters of another." Let's not stop there. "The overbearing manner of a teacher may itself add a new dimension of content to whatever is being taught by making it seem objectionable. An authoritative type of instruction may add a further lesson and have consequences that are distinct from, say, a discussion group in the same topic" (1981, p. 190).
Variables. That's what Barrow is spinning, and these variables can make testing one method of instruction over another invalid. A researcher might think he is testing a method (a variable) when in fact he is unwittingly testing a plethora of variables that are muddying up his work. For example, "different ages and different personalities amongst students may respond in different ways to different styles and techniques of teaching." Also, "different teachers because of their personal idiosyncrasies, strengths and weaknesses may be better employed using different approaches" (Ibid., p. 191). I'm my own example: I disliked the formalistic teaching methods of my grade two teacher (Mrs. Moore) and my grade three teacher (Miss McKormick), but I worked much, much, much harder in Miss McKormick's class, largely because she, unlike her grumpy-looking colleague, often flashed a pretty smile my way. I would have climbed Mount Everest for her!
I distrust variable-drunk (poor) research. How can I know what study tests what actual variables? By studying the actual published research? By considering the sample size, the variables tested and untested, and all other relevant components? Yes. If I have the time. But often I'd rather follow Barrow's advice. "Time spent clarifying one's concept of education would be more useful for a practising teacher than time spent studying behavioural objectives or observation research into teaching methods" (Ibid., pp. 188-189). He adds, "Better to get a thorough understanding of what you want to achieve, both in the long and in the short term, to take critical stock of such information as may be gleaned from sociology, psychology and classroom studies, and to adapt as seems intuitively sensible to the situation you find yourself in, in the light of your understanding and knowledge" (Ibid., p. 192).
I, sharing Barrow's frown over educational research,2 have tended in my career to pick methods according to what my experience tells me will work and according to what I perceive as the educational and socio-emotional needs of my students. That turns teaching into an art, doesn't it?
Yes. For me. Yes. And in my art, I have chosen methods and ideas from G. D. Fenstermacher, J. F. Soltis, S. Wassermann, R. M. Gagne, L. J. Briggs, W. N. Wagner, M. Hunter, P. Wolfe, K. Egan, R. D. Kellough, N. G. Kellough, J. Dewey, J. Rousseau, C. Glickman, R. Barrow, W. Glasser, and anybody else of note according to "what [I repeat] my experience tells me will work and according to what I perceive as the educational and socio-emotional needs of my students". I enjoy doing that much more than trying to figure out what "behaviour is a function" of what.
1 Robin Barrow presently serves as the Dean of Education at Simon Fraser University, Burnaby, BC, Canada.
2 That doesn't mean I don't value good research that actually reduces its number of variables to a comprehensible level.