well done!
i am BY FAR the smartest chick on the block! i thought i had calculated almost everything i needed for my first research questions. got about 800 pages of statistical results. printed. binded as book 1 and book 2. annotated. and all was going well in cornland until i started to write chapter 4, and realized one thing after another:
- (book 1) i did most of the calculations by school. impossible to use if i want to keep my dissertation under 500 pages long!
- (book 1) i sometimes calculated the overall frequencies (great!) but without the univariates that go with it (standard dev, means, standard error, etc.)! and vice-versa.
- (book 2) i looked at the variables at the END of the questionnaire (school, age, gender, grades, etc.) but not at the BEGINNING of the questionnaire (country of origin of teacher, physical appearance, etc.)! must redo the whole darn thing!
- in the end, all this REALLY doesn't matter because i have to redo absolutely EVERYTHING, because THE WHOLE POINT OF THIS STUDY IS TO LOOK AT THE DIFFERENCES BETWEEN NATIVE AND NONNATIVE TEACHERS! so i am not sure how i'll do it, if i need to separate the data into two sets first and then do the calculations for each group separately... but then how do i know if it's significantly different? simple t-test? can you perform at t-test on anovas??? for example: does the first language of the students influence their responses to question 26? that's a simple anova. but now, i want to know what the difference will be between students of native and nonnative teachers. ??? like maybe all students of nonnative teachers, regardless of their first language, respond very negatively to question 26. aaaahhhhhh!!!! i'm entering the fourth dimension! (and einstein was wrong, the fourth dimension is not time, it's insanity!!)
so basically, in a nutshell, and in so many words, I CAN'T DO ANYTHING DURING THE BREAK since i don't know how to deal with all this. must meet with the statistical consultant ASAP in january! the only thing i can do for now is to keep entering the data (got more stuff today from two schools), polish my third chapter, and then think about how i want to present the info (tables, graphs, etc.). that last part is complex since i'll have native vs. nonnative PLUS pre vs. post questionnaires...
this is so frustrating though...
on the brighter side of things, i got one letter of rejection and one invitation for an interview, today.
3 commentaires:
Statistics suck !!
LOL
:-)
l'pingouin: salut toi :) ca fait plaisir de te voir par ici, huhu!
ben, heu, j'me suis perdu ;-)
Enregistrer un commentaire
<< Ã la maison