Dawnbridge

Casio fx-CG50 field manual for IB Math

AI only · 31 of 44

31

χ2 goodness-of-fit test

Maths AI

Run a χ2 goodness-of-fit test on observed and expected counts entered in lists.

Goal

A die is rolled 120 times, landing on each face 18,  22,  20,  25,  15,  20 times. Test at the 5% level whether the die is fair.

1

From the MAIN MENU, open Statistics. Enter the observed counts in List 1 and the expected counts (120/6=20 each) in List 2, pressing after each value.

RadNorm2
d/cReal
List 1List 2List 3List 4SUB118202222032020425205152062020
GRAPHCALCTESTINTRDIST
2

Press (TEST), then (CHI), then (GOF). Set Observed: List1, Expected: List2, and df: 5 (categories minus one); CNTRB is the list the calculator writes each category's contribution into.

RadNorm2
d/cReal
χ² GOF Test
Observed:List1
Expected:List2
df:5
CNTRB:List3
Save Res:None
GphColor:Blue
LIST
3

Press (CALC).

Result

χ2=2.9 and p0.715, so there is insufficient evidence that the die is unfair.

RadNorm2
d/cReal
χ² GOF Test
χ²=2.9
p=0.71539951
df=5
CNTRB:List3

Your turn

Work each one on your calculator, then check the answer.

  1. 1

    A spinner with 5 equal sectors is spun 100 times, landing 23,  18,  25,  15,  19 times on each sector. Test at the 5% level whether the spinner is fair.

  2. 2

    A model predicts expected counts 20,  40,  40,  20 over 120 trials. The observed counts are 26,  35,  44,  15. Test the model at the 5% level.

  3. 3

    A die-like device with 6 equally likely outcomes is used 90 times, giving observed counts 12,  19,  15,  21,  10,  13. Test uniformity at the 5% level.