redzils: (Default)
I think I have solved the outstanding nagging question for my dissertation defense prep: one of the people on my committee looked at my results section before Christmas, and wanted to know if changing my mediation-religion, essentially (Baron and Kenney, 1984 to McKinnon and colleagues, 2002) would have had me doing different analyses. The answer is, well, maybe: my high holy mathemeticians spell out the conditions under which it is appropriate to test mediation (which were not met, so I found another analysis), while hers' does not. There is no defensible version where I say, "oh your guy totally supports my assumptions" - he Just Never Says, Either Way.

So, I've been wondering for the last, oh, six weeks, on and off how I could prepare to get through that gracefully, and without doing Structural Equation Modeling (which is very time consuming and wouldn't have given me much). Tonight I looked at her guy's website again and realized there was just no way out. But, instead of giving in and spending the next four days doing SEM, I tackled the three paths we are quarreling about one at a time, via Sobel test. None of them were signficant, so my decision to not test the complete model was supported, regardless of the assumptions you are prepared to make.

Essentially: I win.

I still need to figure out distribution data for the measures I used when administered in other samples and compare it to what I found inmy sample, but this is not hard, just annoying. And then I will be as prepared as I can be, until I meet with my advisor tomorrow morning. I expect her to ask for a few changes and new thoughts, but am still really, really hoping to pretty much wrap my prep up by close of business tomorrow, so I can spend my weekend merely reviewing the paper and my notes, not frantically flailing.
redzils: (Default)
At some point I would like to post about the difference between significant relationships and meaningful ones, both in psychology and in my life.

And, it is worth capturing for posterity that the DBP director told us to "hum our way through" the overly detailed, insanely complex statistical methods employed in one of the papers we read for today.

Also, should I ask, the vacuum cleaner bags (Style 7) are on the floor in the box room. 
redzils: (Default)
Not a good sign: I just bought bought a pair of shoes online.  Why, yes, I am in the middle of a ridiculously expensive and complex statistical workshop.  My attention span is so blown.  Good thing it's almost lunchtime...
redzils: (Default)
Anybody remember the Gary Larsen cartoon that shows the kid in the crew cut with his hand up - "Teacher, may I be excused? My brain is full."  That is how I feel, and we still have 25 minutes to go.

The good news is that I think I am starting to get it - like the book says, hierarchical linear modeling is really just regression.  Essentially you are writing multiple regression equations, using the slopes and intercepts to parse out variance due to variability within and between individuals and the contexts in which they exist. I wish my understanding of regression was stronger since it seems like a vital foundation, but seem to be doing okay anyways.

The program itself is not particularly intuitive, so it is helpful to be running analyses and graphing equations under the direction of the people who conceptualized the framework.   Hopefully sometime in the next two days we will cover analysis of repeated measures data and why one would mean-center/grand-center/not center predictors.
redzils: (Default)
It is making me very nervous and cranky to realize that the SPSS license on my laptop will expire in 12 days, leaving me unable to work from Washington (state) or Alaska. I think I will probably buy my own copy - another expense no one tells you about in grad school. Does anyone have a recommendation for where or how to do that? seems to offer it semi-cheaply - does anyone have experience with them?
redzils: (Default)
Self -

You managed this morning.  You got up and attempted a run, and walked when you realized the run was trying to kill you.  You got your butt downtown and in to a third floor library chair.  Good work.

Now, DO SOMETHING PRODCUTIVE. Butt-in-chair time only counts if your brain is turned on, and it hasnt been for the last two and a half hours.  You have someplace to be in an hour.  Get something done between now and then. okay?


yours truly
redzils: (Default)
Dear self,

Did freaking out about the stats homework help you at all? 

Yeah, I didnt think so.

It didn't help you get it done any faster. It took a lot of energy to be that anxious.  And you missed several important hours of sleep.

It is good that you eventually calmed down, got a clue, and managed to boggle through most of it under your own power.  That way when you threw yourself on the instructor's mercy, he was able to take ten minutes and help you fix two things, getting it straightened out.

Remember, he is a nice guy.  Like every stats instructor you have had, he is a very nice guy.  (Also, where are all the women in statistics?). So cut the despair, the dramatic stomping and sighing, the middle of the night tossing and turning, and just GET SOME WORK DONE.

 the brat
redzils: (Default)
The pup just woke me up out of a long and repetative dream about building a regression model for something.

It has begun - statistics is hard for me, so when I am working on something statistical, I tend to find the project echoed in my dreams.

As a first year graduate student, I would study stats until my head hurt then go to bed and watch the equations and examples scroll by in my dreams.

Fun. Or something.
redzils: (Default)
I analyzed data!

Not much, etc. but it was meaninful.

I am, of course, not finding what I want, but at least I am doing something.

And I have a pounding headache and an evil meeting coming up.  Blech.
redzils: (Default)
When I get neurotic, I seem to cope in a variety of ways.

I post lots of manic LJ entries.

I run useless errands and do all the piled up dishes.

I read obsessively (to the point of reading while showering) to avoid spending time with my own brain.

I sleep so little that my mind gets all fuzzy.

And I spends lots and lots of money at Amazon**

I love Amazon and like to delude myself that Amazon loves me back.

Tonight I am throwing money in two directions: a waterproof case for my Ipod so I can listen to music and/or podcasts while wrestling fish and guides to SPSS.  Four guides to SPSS, to be precise.  Two basic ones - one known for its accesibility and the other known for its accuracy - and two that provide more detailed information on Syntax manipulation and practical applications of the program.  I am excited.  Now if I can just, you know, read them when they get here, all will be well.

I am justifying them by reminding myself that I am expected to teach this material in the fall, so between that and my thesis is it very necessary.

**(well, it feels like lots and lots of money to me, but tonight's $240 tab is the largest on record. And that isnt bad when you realize I am buying textbooks).
redzils: (Default)
I woke up prepared to accomplish great things, but couldn't dive into that right away since I have a more immediate crisis: meta-analysis homework.

I have spent all morning translating formulas into Excel-ese and getting back impossible numbers. I am now something like 90% of the way through, but the homework is winning. The amount of variance accounted for by corrections (as far as I can tell) has to be between zero and one. Which means 1.5 = wrong, every damn time.

I finally sent out a desperate email to another girl in my class, but haven't heard back from her. I would send it to John as is, but a) that is embarrassing, and b) it *cannot* be as hard as I am making it. Probably there is some error elsewhere in the project, but I don't even know where to look at this point. Damn meta-analysis.

I am giving up, temporarily. Resting and building my strength for another pass at it later on.


redzils: (Default)

November 2010

2122 2324252627


RSS Atom

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Sep. 26th, 2017 02:44 pm
Powered by Dreamwidth Studios