Friday, October 9, 2015

My teachers will be so proud. I said "Analytic Statistics" on YouTube.




Analytic Statistics.


One on my new growing passions is learning about how researchers:

take DATA and turn it into KNOWLEDGE


(I actually really like the way my notes look when I include, charts, diagrams and equations. I feel nerdy. And they look awesome. My note book has been pre-blessed by my kids, so I am often taking notes around their cute little drawings. I love that.)

Data can be words people spoke in an interview, measurements you took in an experiment, or the amount of steps it took for you to make a peanut butter sandwich. Data has existed, and an attempt has been made to measure it.



Knowledge? Knowledge is spiritual. It is the stuff of human evolution and it is the purpose of existence. We know, therefore we exist. We scrape and search for data because the screaming, pulsing, driving impulse to KNOW is hard wired into the very cells and spirits of our beings.

And one funny, silly little way we have to satiate this yearning is to MEASURE, and to DOCUMENT. And so we test, and experiment, and write down numbers, and then hope that this will let us..oh please will it...let us KNOW a little bit more about something.
Statistical Analysis is sobering. 
(It lets us know just how little we actually do KNOW about things through DATA.)

But then, a little sobriety doesn't hurt.

  • Statistical Inference. Is it likely that two data sets are the same? How likely are you to get similar results in future tests? How much variance is there? Error deviation? What is the normal Distribution?
  • The standard error of the mean is dependent on the sample size. How can you figure out where variance comes from - is it due to odd-ball participants or due to the condition that you are testing?
  • How can you analyze variance? ANOVA test? What if you are testing tow things that interact and you want to test the "interaction effect?" What if your data is non-parametric? 
  • What is the Mann-Whitney-Wilcoxon - Utest?

No comments:

Post a Comment