T-what?

I can't say I enjoyed this week's reading, but I did understand some of it.

L&A
I was fascinated by the study by Paul Diedrich using scorers from across a broad spectrum to grade essays to look for some method of consistency. This would make a fascinating study among high school teachers within the same school and grade level. I've been a scorer at the state level and almost a whole day was spent making sure we were trained to be consistent in our scoring.

The section on subject selection made clear some of last week's reading on representative samples. I liked the quick mathematical rule for quantified research, "Generally a safe number is at least ten times as many subjects as variables" (84). This makes sense. You have to make sure things don't happen by chance alone and this number tries to ensure that. Then you have to make sure, with a random sampling, to overcome the problem of people dropping out of the study or not getting enough people from the sample to respond. The term "representative" becomes very important.

There was a lot in this I did not understand. I feel like I missed Algebra I and II and now I'm being promoted to Calculus class without the fundamentals. What is "passive construction per T-unit" (90) and I'm completely lost by the first paragraph under Correlation on page 93. I'm glad I don't have to present this material!

In K&S I got caught up in the questions procedure of the 7th grade teacher who wanted to examine the value of teacher-student dialogue journals. She asked series of questions such as, "To what degree did the students elaborate on their entries?" then later, "She used these questions to devise some interview questions, attitude scales, and categories for analyzing the journal exchanges" (222). Here it's clear that the questions are just as important as the answers. Her initial questions were used to develop further quantitative research strategies. I think this connects to what Sandy asks in her blog about the relevance of "the extent to which non-academic writers were sensitive to rhetorical issues..." (L&A 83). It would help to know what the questions were that incited this study, because they certainly aren't obvious from the material we are given.

There was a great deal here I did not understand either. What's a regular ANOVA? What is a 2x2 design?

I would like to read Martin Nystrand's study "Making it Hard" because of the value it seems to place on longer responses to literature (231) and how about the use of "you" in improving literary interpretation responses! (232). Who would have thought?

I know I'm getting caught up in the content of these studies rather than the how to's and mathematical formulas. These leads me to ask more questions. Since most English and composition teachers are right-brained (yes, I know that's a generalization for which I have no quantifiable data), how many studies go undone because we, as a group, can't or won't get caught up in T-cells, 2x2 designs, or variables? I keep thinking of going into partnership with my friend Kay, a math teacher, who lives for probability and statistics.