Nekokoneko
Somehow, I've got where I'd hoped I'd be by now. A giant Excel file, about 2mB, with over 3000 rows. I've added some variables the LIWC doesn't offer: date and time, neatly atomized into Year, Month, Day, Hour, and Minute. Why not Second? It's not beyond my power. Tomorrow, with the help of whichever of the faculty I can find, I'll get some preliminary reliability coefficients. One for each language dimension, plus Hour (that should be especially interesting). Then I can test for reliability differences between my age cohorts (no mean difference--not remotely).
Fun details. I think I should correlate my reliabilities (at least for the higher-order categories) with posts per year, month, time of day, and maybe average post length. If any of those covary with reliability, I'll get reliability residuals from a regression equation and see if age differences or mean reliabilities change at all.
But the really exciting business is going to start after Friday. I'm only a few steps from completely automating data collection--a huge expansion of my sample. There's no reason I can't narrow my age cohorts (currently 14-15 and 19-20) to single years and make more of them, and there's no reason I can't get 500 subjects in each bracket. Then, while I'm working on factory analysis (and reading about it--there's a book recommendation waiting, from Eric Turkheimer by way of Doug Davis), I send out my survey by email.
An interesting diversion: if I do get around to a survey, I'll be able to relate all kinds of linguistic variables to rates of response. I expect bloggers who post more often will be more likely to respond to my solicitation--that's pretty dull. But I wonder if (and how) frequency of words that describe "cognitive processes" relates to response likelihood. Then there's the "real" content of this survey.
A few maybe-glitches remain, but they're probably of little consequence. Little enough that I can ignore them until real programmers can help me get my code straight. Which it's not. It's ugly, but it seems to work--I took something from my surroundings and made of it something more.
I got edgy outside the lounge. I'll chalk it up to not eating today (an exaggeration: I had a muffin, a bowl of Cheerios, and about a third of a bittersweet chocolate bar).
I haven't thought of what to do about my parents, either. They deserve a divorce, I think. I can't imagine that either of them's happy, or anything like it. I think I'd like to bring L. out here to live with me. Which is the obvious solution to the problem of wanting to have a kid without getting another person involved. Which I do. Or I can wait for a niece or nephew to come by.
And about Unbreakable. It succeeds because, unlike some of the other Shyamalan movies, it has some of its most memorable scenes at the beginning. Bruce Willis riding the train, with the camera looking back at him from the seat ahead. Blood diffusing across a swatch of gauze, a little out of focus. Mr Glass's fall. It's quiet all through. I saw it with Owen last year--over a year ago--and didn't say anything afterward.
Today I miss M.Z. more than usual. I won't see him until winter, probably, unless some accident lands me in China this summer. That would be some accident. This summer will be for work and work and work and pies.
Another game (see if you can follow):
AtypePeacemike: Mr. Godwinson
police girl josh: Mr Skye
AtypePeacemike: Mr. MOrgan
police girl josh: Mr Raving
AtypePeacemike: Mr. Unethical
police girl josh: Mr Believing
AtypePeacemike: Mr. Wheelerite
police girl josh: Mr University of Dueling
AtypePeacemike: Mr. Liandri Corporation
police girl josh: Mr Xan
AtypePeacemike: Mr. Kriegor
police girl josh: Mr Learn how to aim!
AtypePeacemike: Mr. You die too easily
police girl josh: Mr M-m-m-monster kill
AtypePeacemike: Mr. Headshot
police girl josh: Mr Opposing faces
AtypePeacemike: Mr. AS-Overlord
police girl josh: Mr Mandalia Plains
AtypePeacemike: Mr. Pherae
police girl josh: jo .. what?
AtypePeacemike: Mandalia plains=field in FFT?
police girl josh received C:\Documents and Settings\Charles Lee\My Documents\download\policegirljosh\(Ani-Kraze)Chrno_Crusade_Ep22(XviD).avi.
police girl josh: yes
AtypePeacemike: Pherae=Eliwood's home area
police girl josh: mad connexion
Fun details. I think I should correlate my reliabilities (at least for the higher-order categories) with posts per year, month, time of day, and maybe average post length. If any of those covary with reliability, I'll get reliability residuals from a regression equation and see if age differences or mean reliabilities change at all.
But the really exciting business is going to start after Friday. I'm only a few steps from completely automating data collection--a huge expansion of my sample. There's no reason I can't narrow my age cohorts (currently 14-15 and 19-20) to single years and make more of them, and there's no reason I can't get 500 subjects in each bracket. Then, while I'm working on factory analysis (and reading about it--there's a book recommendation waiting, from Eric Turkheimer by way of Doug Davis), I send out my survey by email.
An interesting diversion: if I do get around to a survey, I'll be able to relate all kinds of linguistic variables to rates of response. I expect bloggers who post more often will be more likely to respond to my solicitation--that's pretty dull. But I wonder if (and how) frequency of words that describe "cognitive processes" relates to response likelihood. Then there's the "real" content of this survey.
A few maybe-glitches remain, but they're probably of little consequence. Little enough that I can ignore them until real programmers can help me get my code straight. Which it's not. It's ugly, but it seems to work--I took something from my surroundings and made of it something more.
I got edgy outside the lounge. I'll chalk it up to not eating today (an exaggeration: I had a muffin, a bowl of Cheerios, and about a third of a bittersweet chocolate bar).
I haven't thought of what to do about my parents, either. They deserve a divorce, I think. I can't imagine that either of them's happy, or anything like it. I think I'd like to bring L. out here to live with me. Which is the obvious solution to the problem of wanting to have a kid without getting another person involved. Which I do. Or I can wait for a niece or nephew to come by.
And about Unbreakable. It succeeds because, unlike some of the other Shyamalan movies, it has some of its most memorable scenes at the beginning. Bruce Willis riding the train, with the camera looking back at him from the seat ahead. Blood diffusing across a swatch of gauze, a little out of focus. Mr Glass's fall. It's quiet all through. I saw it with Owen last year--over a year ago--and didn't say anything afterward.
Today I miss M.Z. more than usual. I won't see him until winter, probably, unless some accident lands me in China this summer. That would be some accident. This summer will be for work and work and work and pies.
Another game (see if you can follow):
AtypePeacemike: Mr. Godwinson
police girl josh: Mr Skye
AtypePeacemike: Mr. MOrgan
police girl josh: Mr Raving
AtypePeacemike: Mr. Unethical
police girl josh: Mr Believing
AtypePeacemike: Mr. Wheelerite
police girl josh: Mr University of Dueling
AtypePeacemike: Mr. Liandri Corporation
police girl josh: Mr Xan
AtypePeacemike: Mr. Kriegor
police girl josh: Mr Learn how to aim!
AtypePeacemike: Mr. You die too easily
police girl josh: Mr M-m-m-monster kill
AtypePeacemike: Mr. Headshot
police girl josh: Mr Opposing faces
AtypePeacemike: Mr. AS-Overlord
police girl josh: Mr Mandalia Plains
AtypePeacemike: Mr. Pherae
police girl josh: jo .. what?
AtypePeacemike: Mandalia plains=field in FFT?
police girl josh received C:\Documents and Settings\Charles Lee\My Documents\download\policegirljosh\(Ani-Kraze)Chrno_Crusade_Ep22(XviD).avi.
police girl josh: yes
AtypePeacemike: Pherae=Eliwood's home area
police girl josh: mad connexion