Hey all~~
Nicely done analysis of Pianko. There are several problems with the study, and collectively, you've identified about all of them.
Questions:
Let's return to Scott's question about genre. What difference does the publication venue make? Do the two articles on the same topic read the same way in the two journals? How does the genre of the journal shape what we know, what we write, how we read?
And let's think in terms of Rory's critique of the dearth of randomization. Suppose you want to do a case study, a set of case studies, or a study with a small N. How do you choose your students, and what difference does that make?
I'll look forward to talking with you all tomorrow about these and other items:
1. The Emig study. It's fairly classic social science epistemology. What do I mean when I say that? It's also not without its limitations: what are those, and how big an impact do they have on the study?
2. The Beaufort study. Emig's study was her dissertation; this is a report from Beaufort's dissertation. Beaufort is also social science in approach. How does it compare to Emig, some 30 years later? What are its limitations?
3. If you look on your syllabus, you'll see that you have a design piece due the week before break. What I'm looking for here is a very small study that you'll complete before the end of the term. We'll use the study I'm conducting with the undergraduate student as an example of a small study so you can see what it looks like.
Looking forward to seeing you tomorrow,
ky
Subscribe to:
Post Comments (Atom)
1 comment:
Let's return to Scott's question about genre. What difference does the publication venue make? Do the two articles on the same topic read the same way in the two journals? How does the genre of the journal shape what we know, what we write, how we read?
**Historically, CCC has been a more humanities-based journal, meaning that it's much less oriented to quantification of results. The articles in CCCs tend to provide accounts and interpretations, but there can be a lot of slippage there. By way of contrast, RTE is social science in approach, so more interested in quantification--so that any study can be *replicated* or repeated. The idea here is that if you can repeat something and the results come out the same, it's more likely to be
an accurate account of a phenomenon or activity. So in CCC, we might say that students paused and then reflected; in RTE, we'd say that one student paused for 10 seconds, and then reflected for 5. Is this what you all read?
And let's think in terms of Rory's critique of the dearth of randomization. Suppose you want to do a case study, a set of case studies, or a study with a small N. How do you choose your students, and what difference does that make?
**it makes a big difference, but a couple of things here are worth noting. For one thing, you can't generalize from a case study anyway, and the point of randomization is to permit generalization. For another thing, you're probably going to rely on what McNealy calls a convenience sample. In other words, you'd be studying the students who agreed to work with you. A third way to think about this is that it's a design limitation: in other words, all studies have design limitations, and yours will as well, so it's important to note what those limitations are.
1. The Emig study. It's fairly classic social science epistemology. What do I mean when I say that?
*I mean that she's very clear about her methodology, sufficiently so that you could replicate the study if you wanted to.
It's also not without its limitations: what are those, and how big an impact do they have on the study?
*as Shreiner suggests, without considering the context, Emig *seems* to be working off of a fairly romantic view of the writer, and that colors the conclusions she draws. It's also worth noting that her students are upper-middle class, so she's getting a particular demographic here--one that is hardly unversal. In another critique of Emig, Frank Voss notes that asking students to compose aloud while also recording them so distorts the process that it in effect invalidates the results. Do you agree?
2. The Beaufort study. Emig's study was her dissertation; this is a report from Beaufort's dissertation. Beaufort is also social science in approach. How does it compare to Emig, some 30 years later? What are its limitations?
Emig is without context (nearly); Beaufort is almost totally context. What else?
3. If you look on your syllabus, you'll see that you have a design piece due the week before break. What I'm looking for here is a very small study that you'll complete before the end of the term. We'll use the study I'm conducting with the undergraduate student as an example of a small study so you can see what it looks like.
*I believe you've seen this. What else would you like to know about it?
Post a Comment