The Kruskal-Wallis H Test

• The Kruskal-Wallis H Test is a

nonparametric procedure that can be used to
compare more than two populations in a
completely randomized design.
• All n = n1+n2+…+nk measurements are jointly
ranked (i.e.treat as one large sample).
• We use the sums of the ranks of the k samples
to compare the distributions.
 The Kruskal-Wallis H Test

Here
Here are
are thethe steps
steps in
in doing
doing the
theKruskal-
Kruskal- Wallis
Wallistest:
test:
a.a. State
State the
the null
null hypothesis.
hypothesis.
Ho: The kk population
Ho: The population distributions
distributions are
are
identical.
identical.
b.
b. State
State the
the alternative.
alternative.
Ha:
Ha: atat least
least two
two of the kk population
of the population
distributions
distributions differ.
differ.
 Example
Four groups of students were randomly
assigned to be taught with four different
techniques, and their achievement test scores
were recorded. Are the distributions of test
scores the same, or do they differ in location?
1 2 3 4
65 75 59 94
87 69 78 89
73 83 67 80
79 81 62 88
 Teaching Methods
1 2 3 4
65 (3) 75 (7) 59(1) 94 (16)
87 (13) 69 (5) 78 (8) 89 (15)
73 (6) 83 (12) 67 (4) 80 (10)
79 (9) 81 (11) 62 (2) 88 (14)
Ti 31 35 15 55

Rank
Rankthe
the16
16 HH00::the
thedistributions
distributionsof
ofscores
scoresare arethe
thesame
same
measurements
measurements HHa::the distributions differ in location
a the distributions differ in location
from
from11toto16,
16, 2
and 12 T
andcalculate
Test statistic: H =
calculate ∑ i − 3(n + 1)
the
thefour
fourrank
rank n(n + 1) ni
sums.
sums. 12  312 + 352 + 152 + 552 
=   − 3(17) = 8.96
16(17)  4 
 Teaching Methods
HH00::the
thedistributions
distributionsof
ofscores
scoresare
arethe
thesame
same
HHa::the
thedistributions
distributionsdiffer
differin
inlocation
location
a

12 Ti 2
Test statistic: H = ∑ − 3(n + 1)
n(n + 1) ni
12  312 + 352 + 152 + 552 
=   − 3(17) = 8.96
16(17)  4 
Reject
RejectHH00..There
Thereisissufficient
sufficient
Rejection
Rejectionregion:
region: For
Foraaright-
right-
evidence
evidenceto toindicate
indicatethat
thatthere
tailed
tailedchi-square
chi-squaretest withαα==..
testwith there
05 isisaadifference
differenceinintest
testscores
scoresfor
for
05and
anddf
df==4-1
4-1=3,
=3,reject
rejectHH00ififHH
the
thefour
fourteaching
teachingtechniques.
techniques.
≥≥7.81.
7.81.
 Activity
Four different teaching techniques in Physics were compared with one
another. Four classes were randomly assigned to undergo one of the four
teaching techniques. After being taught for one month, the students were
given in achievement test. The ff. are their achievement scores.
Teaching Technique
1 2 3 4
65 75 59 94
87 69 78 89
73 83 67 80
79 81 62 88
81 72 83 95
69 79 76
90

Test the null hypothesis that four teaching techniques in Physics do not
differ in effectiveness.