On stats

On stats
By Zeal
Because you know you weren't paying attention.*

TYPES OF DATA:
Quantitative - numbers and measurement
-Discrete: only whole numbers (1, 2, 3) or intervals (number of people born today)
-Continuous: In-betweens and whatnot (length, height, time)
-Ratio: an absolute zero with no negatives (distance, temperature in Kelvin)
-Interval: negatives are included (temperature change)

Qualitative - categories of stuff
-Nominal: can't be ordered (colors, your favorite variety of cheese)
-Ordinal: can be ordered (duh) (rankings of how much you like cheese, from 1 to 5)
NOTE: Stuff like rankings are qualitative, not quantitative, because they're measured

VARIABLE STUFFS
Dangit this table was hard to type

          Sample Population
Mean      x-bar  μ
St. dev.  s      σ
Number    n      N

FORMULAS GALORE

For the record, I use "mean" rather than x-bar in these formulas because I don't wanna get the hyphen in x-bar confused with a minus sign

Mean:
sum of stuff
___________
     n

Standard deviation:
  ______________
 | sum of x - mean (take each individual x and subtract the mean, then add the results)
 | _______________
√        n
(waaaaaaaay easier with a calculator)

Standard error (aka SEx) (maturity, guys)
 s
__
√n

T TESTS
Used for testing if two sample means are actually equal or if we just got the result by chance

Confidence level: chance that we got the result by just being really lucky
Degrees of freedom: n1 + n2 - 2
Use a t-distribution table to look up critical value

mean1 - mean2
________
  ________________
 | (s1)^2 + (s2)^2
 | _______________
√         n

If the t-test value is greater than the critical value, they are equal!

*I don't blame you.