I just imagined the whole class having a wrestling match. We could throw chairs and ladders and everything.
Maybe a tamer idea would be to just build a bunch of jigsaw puzzles together.
That being said, I'm not sure what puzzles me or what I'm wrestling with in regards to the articles.
I know what bugs me though.
I have a problem with studies that have such small sample sizes. Moreover, none of these social experiments seem to have control groups or repeatable elements.
I think the reason why we argue so much in class over the discrepancies in the "study" articles we've looked at is because these articles can't really come anywhere near giving proper findings. I appreciated it when Dr. Rickly said that many of these articles are just meant to be testimonials--Hey, this is what I did and it worked. Try it out.--But I wish that there were more feasible studies that really could help give solutions to some of the questions about writing process, group work, etc. Are there no studies out there with large sample sizes that have been conducted over several years?
Maybe I'm just tainted by my job where I look at IMRAD methods all day long and critique my authors' scientific method. But couldn't pedagogical theory be just as practical if we spent time on more extensive studies? Surely this is worth a dissertation topic or two. Maybe I haven't looked hard enough.
Okay...so I'm wrestling with this. And I guess I'd really like to punch these shallow studies in the face. Bam!
Subscribe to:
Post Comments (Atom)
6 comments:
I like to think that "getting your hands dirty" in the classroom leads to becoming a better (and eventually the best) teacher you can be. I guess I'm more of a "do-er" than a "theorizer".
Lorna raises a good question: "Are there no studies out there with large sample sizes that have been conducted over several years?" The question came, I assume, after reading Emig's study of one 12th grade writer ("Lynn: Profile of a Twelfth-Grade Writer") and Perl's study of five unskilled college writers ("The Composing Processes of Unskilled College Writers").
We do have other studies with greater sample sizes. They can be found in the journal __Research in the Teaching of English__. Some of the best known are from these Carnegie Mellon authors (Wallace, Hayes, Flower). But from my experience, even these sample sizes are fairly small--say a class of 30 to 40 students. In order to measure development in writing the study most cover a fairly long period of time (at least a semester). Thus, the researcher's attention is directed "longitudinally" over time rather than "latitudinally" over population. Ideally we would have both, but again, all researchers face constraints.
Also, because writing instruction occurs in English departments, and because English scholars like stories, even empirical research often will focus on one or two main characters, such as Lynn. Another memorable study of the inculturation of a new PhD student to the academic style of writing focuses only on one student--"Nate." Yet, I would argue that it provides a lot of insights into how advanced students learn to write for their discipline.
One advantage of the Texas Tech online composition program is the massive amount of data that is available for research. You are right, Lorna, this data would feed several dissertations for those willing to tackle it.
I am also going to notify Dr. Rickly of your blog post here. She has more experience with composition research and may have other insights into your question. Thanks for posting it.
There are a couple studies with larger sample sizes, but few are experimental/quasi-experimental. Unfortunately, when you work with humans/social settings, it's nearly impossible to control variables.
We'll be reading a study by Lunsford and Connors about the frequency of errors in college writing (and Lunsford and Lunsford--her neice--have replicated it recently, with a much larger sample size).
That doesn't mean we can't have good, rigorous research; but it's hard, and larger samples are even harder. But we can make sure that we control as much as possible; that we apply systematic methods; that we use triangulation (either/both in methods and analysis); and so forth.
Rich Haswell has a great article on how NCTE/CCCC seems to have declared "war" on what he calls "RAD" scholarship (replicable, aggreable, and data-driven. I have a pdf if you're interested:
http://www.faculty.english.ttu.edu/rickly/5363/Haswell.pdf
That said, it IS hard to study writing/writing classes, because every one of them is different: different people, different experiences, and so forth.
And Ken's right: most of us are trained in literary studies and criticism, not in empirical research. Our field tends toward qualitative rather than quantitative, descriptive rather than correlational analysis. Ken's right, too, about the immense data set we have in the TOPIC data base....and several PhD students are looking at that.
Still, I think you ask good questions. We DO need more large scale studies (though Lunsford and Lunsford said that the hardest part was getting IRB approval from all the cooperating schools). So keep calling for them!
Lorna, I had never thought the size and extent of the studies as frustrating to the class, but you might be right. I suppose it depends on your interests. I am wondering if the authors deliberately use smaller studies to catch our attention. Personally, I find the surveys with one student more interesting, just because I get a stronger sense of the abilities and efforts of that one student. I remember the findings because I have something invested in the individual. When the studies are broad, I am not as likely to feel that connection. I do agree that larger surveys might put some of the smaller studies into perspective, and it is odd that there has not been a great deal of research in this area compared to the smaller case studies.
First off, I enjoy the semi-violent nature of your post because I now realize I am not the only one who imagines our class in a fistfight. Of course, my fight involves the invisible enemy, Freshman Comp.
I never really thought about the fact that the sample sizes are so small. Now that you mention it I realize you are right. We have been dealing with very small samples. Maybe a larger one would reveal more about the true nature of teaching composition.
Everyone else gets right to the meat of the matter - I'm still working thru the wrestling visual you propose!
I might propose that the issue here isn't study sample size, or statistical consistency or methodology, but the fact that we teach something in composition that, unlike the temperature at 19th and University at 8:02:35 evry Wednesday morning, cannot be so nicely articulated in a rational, comprehensive study (much to my chagrin)? It's like literature for me - I keep trying to find its structure - even though I'm told there isn't a universal one. In certain ways, we seem to be postmodern composition instructors searching for a structuralist answer?
Post a Comment