inferential stats

null hypothesis =

no difference between sample and population mean

alternative hypothesis =

significant difference between sample and population mean

to reject the null means there ____

is significant difference

do not reject the null means there _____

is no difference

type 1 error =

null is true but you reject it

type 2 error =

null is false but you do not reject it

type of error that is more risky

1

type 1 error means you said ____ when actually ____

there is a significant difference, is not

type 2 error means you said ____ when actually ____

no significant difference, actually is

alpha signifies the ____

probability of making a type 1 error

alpha = 0.05 means

5% change of incorrectly rejecting the null

which test is non directional

2 tailed

2 tailed test

• when alpha=_____, then critical region is outside _____ (95% CI)

• if z falls within critical region, then _____

0.05, ±1.96, significant difference

test that is directional

1 tailed

1 tailed test

• more____

• when z falls in critical region, then ____

powerful, sig difference

inferential tests measure ____ or _____

differences, relationships

non-parametric tests

• _____ free

• ex: ____

assumption, log regression

parametric tests

• require _____

• ex: ____, t-test, ____ regression, pearsons _____

3 assumptions, anova, linear, correlation

3 assumptions with parametric tests

normality, homogenity, data type

parametric assumption - normality

• ____ sampling from a population that is ____

• ____ procedures

random, normally distributed, goodness of fit

examples of goodness of fit procedures =_____, for the assumption _____

shapiro-wilk, normality

parametric assumptions - homogenity

• some _____ will be present, but the degrees of ____ in groups should be roughly ___ and not _____

• ____ test

error variance, variance, equilivent, significant

parametric assumptions - data type

• data is ___ or ____

interval, ratio

4 levels of measurement

ratio, interval, ordinal, nominal

continuous levels of measurement

ratio, interval

categorical levels of measurement

ordinal, nominal

level of measurement - ratio =

numbers with equal intervals measures from true zero

example of ratio data

distance, age, time, weight

level of measurement that has a true zero

ratio

levels of measurement - interval =

numbers with equal intervals, no true zero

example of interval data

temperature

levels of measurement - ordinal =

numbers indicate rank

example of ordinal data

MMT, pain

levels of measurement - nominal =

numbers are category labels

examples of nominal data

gender, blood type, Dx