Sunday, September 30, 2007

Grading of Confidence

"Grading of Confidence" is actually a subsection in Polanyi's chapter on Probability, in _Personal Knowledge_. I know that probability is rhetorically constructed - that is, there is a rhetoric of probability within disciplines -- in the social sciences this probability is measured with statistics. In composition studies, which values the personal narrative, if someone said it happened and is due to X causality, then it is pretty surely true unless someone else comes along and says -- well I experience that differently, and I think it was more due to Y than to X. What you are saying is all in your head.

The rhetoric of probability is really what Polanyi is discussing, although he does not say so or call it rhetoric. And so, if we can measure levels of confidence statistically, then we should also be able to measure levels of confidence by asking people how confident they are. Either way, it is a "grading of confidence" that is constructed rhetorically.

What is the probability that something is random? Like the sidewalk that leads up to my front door. What is the probability that the cement arranged itself that way, accidentally? Polanyi discusses this sort of thing to make a point.

Ong and Vygotsky and Gestalt psychology

Gestalt psychology was important to Vygotsky because he was critiquing it in his own work. Gestalt psychology was a response to the behaviorists and the experimental psychologists, who agreed that psychological processes could be studied by analyzing “basic constituents” (Cole & Scribner’s intro to Vygotsky, _Mind in Society_). Gestalt psychology rejected the notion that complex processes could be accounted for by examining simple processes.

Vygotsky believed that the internalization of culturally produced sign systems brings about behavioral transformations and forms the bridge between early and later forms of individual development” (Cole & Scribner, p. 7). Vygotsky has a chapter on the importance of examining writing when studying how cultural sign systems are internalized. This reminds me of Ong’s argument in his chapter on how “Writing Restructures Consciousness” (1988). Except that, Ong comes at this from a different angle, because he is talking about “restructuring.” This is different than “internalization” in some ways, and in others it is not different. The problem with Ong’s discussion is that it makes it sound like people who do not write, are generally on a lower intellectual level than people who do write. In some ways though, this also maps onto Vygotsky’s discussion of children who are yet unable to internalize certain cultural tools. And yet, the stances of these two authors differ because Vygotsky does not seem as arrogant and patronizing as does Ong. Vygotsky points out that because “mentally retarded” children have difficulty thinking abstractly, one approach was to remove all abstract-thinking curriculum, or curriculum that encourages abstract thinking, from their education. But, this had a detrimental effect on the mentally retarded children’s development. Instead, Vygotsky argues that because they have difficulty with abstract thinking, mentally retarded children should certainly receive education that encourages this type of thinking. My point is, Vygotsky was trying to help marginalized people. Ong instead, judges them --- at least that’s how he reads sometimes. Not to say Ong doesn’t have good points. (Vygotsky also argued for the intellectual benefits of play, since humans play above their level of actual intellectual development – so if you want people to become smarter, you have to let them play).

Vygotsky and Ong map onto each other because both of them are concerned with how technology influences the way we think. Vygotsky was concerned with learning, while Ong was concerned with intellectual capability. He focused more on writing and rhetoric.

An older colleague of Vygotsky, PP Blonksy, had already developed the idea that “behavior can only be understood as the history of behavior” (Vygotsky, 1978, p. 8). I believe the following is key to why Vygotsky has become so attractive as a theorist to those studying digital writing:

Blonsky was also an early advocate of the view that the technological activities of people were a key to understanding their psychological makeup, a view that Vygotsky exploited in great detail” (p. 8).

Later in 1988, Ong argues that writing is a technology, and that “more than any other single invention, writing has transformed human consciousness” (p. 77). I'm not sure about this idea of "transforming" and "structuring." This I am still thinking about because I think I might disagree that that is what writing does. So, I have to think about the differences between "mediation" and "structuring."

What I wanted to talk about was the differences between Polanyi and Vygotsky, but I'm not there yet. Obviously their focuses are very different, but what I mean is how are they different when it comes to their views of knowing.

Friday, September 28, 2007

Population - 50% random selection -- 75%

I am going to pull another population from the census this weekend. Whether I will use it or not I don't know. It is not going to change the low response rate necessarily, so I have to ponder this. If I decide to simply contact everyone, then it is not longer a random selection, and some of the probability theory I was going to rely on will disappears. However, because I have such a low response rate anyway, that probability theory disappears regardless. I mean, probability is still an issue, but the fact that I randomly selected a population isn't going to matter much when I have such a low response rate.

Response bias is a concern. Inferential statistics used in analyzing date from surveys assumes that 100% of the respondents returned the surveys. response rate is one guide to the overall representativeness of the survey participants. Response bias is less when a large number of individuals return the surveys. Response bias is more when few return the survey. Respondents are self-selecting anyway, and that result is pronounced when few return the survey. As Babbie (2004) points out, with a low response rate, there is usually something more going on than unwillingness to take the survey when you get a low response rate.

So far, I have contacted 50% of my entire population -- my census. And so, I have to navigate whether I am going to go to 75%, or whether if I go to 75% I might as well go to 100% and have a survey of the entire census rather than a random selection.

Interestingly, my biggest concern here is that I told the first two groups that they were randomly selected, and so if I go with 100%, that won't be true. So I worry about my researcher image and its credibility.

Response Rates on Surveys - Follow-up Mailings

Babbie (2004) discusses response rates and follow-up mailings on pages 260-261. He states that return rates of 50% are acceptable to analyze and publish. 60% is good and 70% is very good. Of course, in our field of rhetoric & composition where strict social science methods are not necessarily as highly valued (as in the fields that Babbie refers to), I've seen surveys analyzed and written about with response rates much less than 50%. Presently, my response rate for population one, Phase One, was just over 10%. I've sent out a reminder, but have not yet checked if I received any additional responses. while a few contacts have provided excuses why they cannot participate, such a low response rate overall should be a danger signal. As Babbie states "a low response rate is a danger signal, because the nonrespondents are likely to differ from the respondents in ways other than just their willingness to participate in your survey. Rich and Bolstein (1991), for example, found that those who did not respond to a preelection political poll were less likely to vote than those who did participate" (p. 261).

I would imagine since the purpose of the survey was clearly stated towards digital composing and copyright, it might be that those who did not respond did not think they were qualified, did not think it was important, didn't understand what I was talking about, etc. I intuit that even when what appeared to be valid reasons for not participating, such as business-- still, the reasons might be more complicated. If your program is represented in the survey, and gets all the copyright questions wrong, that won't look good for the program. However, I really have no sure way of knowing who from what program answers in what way, since at the end of the survey individuals are asked to name their school - they can also choose other or can't answer.

Babbie does concede that the literature varies widely on what is an acceptable response rate. I'd think that since explaining context is so important in our field, regardless of my return rate, I will need to qualify taking anything seriously from the survey results. It may occur that the only purpose of the survey turns out the provide sorting for students who might be interviewed.
Babbie has some interesting observations and hints regarding follow-up mailings. There should be two follow-ups, at two week intervals. For Phase One, I've done the first follow-up. I will do one more the week after next after excluding those who responded to my first reminder. Next week I will do Phase Four, which is the follow-up to those I have not heard from regarding Phase Two (recruitment for folks from the second population).

Amazingly, after sending initial recruitment emails to both populations of 50, I had received exactly 6 responses from each group, for a total of 12. I used two different strategies for recruitment. In the first instance, I had a long formal email of 800 words. It included the faculty recruitment email cut and pasted at the bottom. I did not include a link to the survey. For the second phase, I cut the email down to 200 words, included the faculty recruitment email as an attachment, and included a link to the survey. In both cases, amazingly, 6 people responded. The only difference was that in phase two when I included the link to the survey, I received the responses faster.

Thursday, September 27, 2007

Personal Knowledge

I am reading Michael Polanyi's _Personal Knowledge: Towards a Post-Critical Philosophy_ (1958). I am also administering an online survey to randomly selected technical/professional writing programs in the US. So far, response rates are not that great. So I say,

. . . what does the survey really matter anyway - all that matters is Personal Knowledge -- there are statements about probable events, and statements that might be probable {as in truth} (but these aren't the same thing) but, why would I place higher value on knowledge created via a theory on the relationship between two *statements* about probable events, versus knowledge I experience first-hand, and my own assertions of what's probable (based on a logical relationship; empirically validated sensory experience between what I know and what I see).

Polanyi defines a theory as "something other than myself" (p. 4). He draws a connection between maps and theories. "A theory may be regarded as a kind of map extended over space and time" (p. 4). The question he poses, is, should we consider as more objective the form of knowledge that relies "to a greater measure on theory rather than on more immediate sensory experience" (p. 4). He goes on to point out the strengths of theory.

1) A theory is "objective knowledge" as far as it is not I. However it can be proved wrong when I use it.

2) A theory cannot be led astray by my personal illusions.

3) Theories can be constructed without regard to "one's normal approach to experience" (p. 6). However, for this to make theories more "objective" than personal experience means that we have to value the fact that theory may disregard our "terrestrial location," equally applying to inhabitants of all the earth: "provided they share our intellectual values" (p. 6).

Regarding the issue of Personal Knowledge and Vygotsky, although Vygotsky is not cited in the index, Polanyi speaks favorably of gestalt psychology, but Vygotsky does not. However, if as Vygotsky says, we internalize tools that help us know, it would seem to me if you bought that, you'd have to buy the existence and utter superior importance of Personal Knowledge because Personal Knowledge simply reflects an individual's internalization of the tools and ways of knowing in her culture.