Unit 8 stats Chi Square booty

kms.

Follow-up Analysis

The largest component(s) of x² is ______ because the observed counts of ______ was higher/lower than expected

Procedure (goodness of fit)

X² GOF test for P

H_o

The claimed distribution of categorical variable is true

H_a

The claimed distribution of categorical variable is NOT true

Largecounts

The smallest expected count is ____, so all expected counts are > 5

OR

If there is a table we can say that all values in the table are > 5

OR

if there are just a few values we can show that each of them are > 5

Calculate

X²_stat = (o-e)² / e + … + (o-e)² / e = test statistic

Calc command: X² GOF test (use L₁ and L₂)

Procedure (homo)

X² test for Homogeneity ——→ bc its AT LEAST 2 samples (can be more!)

Find expected counts for 2 sample

(RowTotal)(ColumnTotal) / (TableTotal). —→ for each square of table

10% check for 2-sample

Assume ___ < 10% (population)
Assume ___ < 10% (population)

Largecounts 2-sample

See expected table above, smallest expected count is ____, so all expected counts are > 5

CALCULATE homo 2-sample

X²_stat = (o-e)² / e + … + (o-e)² / e = test statistic, note down p-value

Calc command: X² - test —→ edit [A]: 3×2, enter O-values

[A] —→ Observed
[B] —→ Expected

df = (# rows -1)(# columns -1)

Hypotheses for homo 2-sample

H_o: There is no difference in the distribution of _____ and _____

H_a: There is a difference in the distribution of _____ and _____