PSYC 3000 Lecture Notes - Lecture 4: Everytime, Normal Distribution, Standard Score

• School: Carleton University
• Department: Psychology
• Course: PSYC 3000
• Professor: Deanna C Whelan

Document Summary

How much data and information is spread-out: ordinal, continuous data, spss: scale. We define distributions by: their means (where the middle is, their standard deviation (the amount of spread, and their shape (skew and kurtosis) No skew and no kurtosis (cid:862)(cid:374)or(cid:373)al(cid:863) (cid:894)gaussia(cid:374)(cid:895) All parametric tests (t-test, anova, correlation, etc) require something to have an approximate gaussian (normal) shape. So a probability distribution is the distribution of all the probabilities! Flipping a coin and guessing correct; each toss has a 50/50 odd of being correct (cid:894)assu(cid:373)i(cid:374)g a(cid:374) (cid:862)ho(cid:374)est(cid:863) (cid:272)oi(cid:374)(cid:895) Most likely answer of 10 flips of the coin, is 5 correct answers. The normal distributions is a special probability distribution: standardizing. Z-scores: z = (x - ) / , x = value, = population mean, = population standard deviation. ) is i(cid:374) the u(cid:374)it of ta(cid:374)dard de(cid:448)iatio(cid:374)s ho(cid:449) far a(cid:449)ay fro(cid:373) the (cid:373)ea(cid:374) is the s(cid:272)ore (expressed in sd units)