Classroom dynamics are shaped by many things, including attributes, dispositions, knowledge, and skills of the teacher, students’ personalities and backgrounds, and, among other things, students’ academic abilities. Margarita Pivovarova (2014) seeks to isolate and measure the effect of students’ academic abilities, as measured by a low-stakes standardized test, on the performance of other students on subsequent exams, in effect, attempting to address the title question of whether students should be homogenously tracked or heterogeneously mixed according to academic ability. As I read this article, there were several things that I took issue with, each of which I want to address in turn.
Relatively early in the article, Pivovarova (2014) begins to use certain words to describe students’ academic performance that I found to be rather unsettling and dehumanizing in nature; by simply reducing them students to statistics on a page and using descriptors such as, good, bad, average, and marginal, she fails to positively recognize that the data she are interpreting are individual humans who are much more complex than simple adjectives. In an attempt to distance herself from the inescapable connotations associated with such words, she includes the following in her notes section, nearly three fourths of the way through the article:
4In order to make the interpretation easier, instead of labeling students by the level of achievement as 1 to 4, I will call students at the lowest level of achievement “bad” students without attaching the actual meaning of the word “bad”; students at the highest level – “good” students, and students in the middle of the achievement distribution – “average”. Among “average” students, I will distinguish between “marginal” (those whose achievement is below provincial standards, or level 2) and just “average” (level 3). (Pivovarova, 2014, p. 29)
While I recognize that it can often be easier to simplify data for ease of writing and communication with the reader, I found the inclusion and repeated use of these words to be over-simplified and indelible choice. The visceral reaction I had when reading these words, to me, underscores the importance and value of being very conscious and intentional in my word choices when making qualitative judgments and assessments about data points, and to always remember what the data I describe actually represent, which, in this case are students.
When I began to reflect on the situative context behind Pivovarova’s (2014) work to better her dispositions and analyses, it became clear that her approach was quite distanced from any actual interaction with the students themselves. With such a large sample size (n=228,947 students), it seems likely that this information was reported to her by, or obtained through an institution involved with Ontario’s standardized testing, as opposed to being collected by she herself. While there were likely human-to-human interactions in the collection and analysis of these data, it seems as though one could reproduce such a study without ever seeing an actual person represented by the data. I see this as a major weakness of her methods; by never interacting with the ‘subjects’ of a study, it seems as though it would be quite easy to make such qualitative assessments that fail to acknowledge the humanity of the data points. As I read more about the author and her background, I found that her focus is in field of economics, which can have a cold, sterile distance to it, which was the feeling conveyed through this article.
Despite my above sentiments, when I began to reflect on the motivation behind Pivovarova’s work, I see the most noble of intentions behind it. The article intends to dispel current thinking on the linear relationship between the effects peers have on an individual achiever’s learning. This is done in the name of improving school effectiveness and efficiency, with the ultimate goal of improving student achievement. If current practice dictates that students be grouped into classrooms in a certain way, homogeneously by achievement, for example, and new information suggests that more students would benefit to a greater extent if they were grouped using a different method, then the paper serves to fill an invaluable need that will improve outcomes for a significant number of students, something that has limitless value. An example of this, representing another particular strength of Pivovarova’s (2014) article is the manner through which she debunked an oft-used excuse of educators: the idea that a “bad apple” student can ruin the learning environment for all other students. Her data suggest that an increased number of low-performing students do not, on the whole, negatively impact the outcomes of high achieving students (Pivovarova, 2014). To use data to soundly reject such a dehumanizing (of students) notion is one of my most valuable take-aways from this research.
Through reading this article, I have gained a valuable insight about how I will implement my innovation into my own practice. I hope that collecting data through participative action research and utilizing methods that always actively seek to humanize my participants, my innovations will not reduce the lives and abilities of those involved to mathematical formulas, algorithms, and simple numbers on paper. But rather, that my methods will always refer to the data in ways that respectfully acknowledge the various backgrounds and stories of those involved in the study.
Pivovarova, M. (2014). Should We Track or Should We Mix Them? Mary Lou Fulton
Teachers College. Tempe: Arizona State University.