MATH13 – ADVANCED STATISTICS

Lecture Note: Case of Two Independent Samples

Lesson No. 1: Fisher Exact Probability Test

Learning Objectives:
At the end of the lesson, you will be able to:
1. discuss the assumption and conditions of using the Fisher exact probability test
2. perform Fisher exact probability test
3. explain the results of the Fisher exact probability test

The Fisher exact probability test, named after Ronald Fisher, is an extremely useful
nonparametric technique for analyzing discrete data (nominal or ordinal) when the two
independent samples are small in size. It is more accurate than the chi-square test especially
for a contingency table with one or more cells with expected values less than 5.
Because the Fisher Exact tests can be resource-intensive, limits have been placed on the
sum of all the cells in the supported contingency tables. These limits are currently set at 2,000
for a 2 × 3 table, 1,250 for a 2 × 4 table, 360 for a 2 × 5 table, 175 for a 2 × 6 table, 110 for a 2 ×
7 table, 75 for a 2 × 8 table, 40 for a 2 × 9 table, 320 for a 3 × 3 table, 95 for a 3 × 4 table and
30 for a 3 × 5 table. There are no limits for 2 × 2 tables.

Methods
1. State the hypotheses.
Ho: Variable A is independent of variable B. (The two classification/categories are not different.)
Ha: Variable A is dependent of variable B. (The two classification/categories are different.)

2. State the rejection rule.

Reject Ho if p ≤ α.

3. Find the value of the test statistic.

( A+ B ) ! ( C+ D ) ! ( A+C ) ! ( B+ D ) !
p=
N ! A! B!C ! D !

for the given 2 x 2 contingency table

Var1 / Var 2 Category Category 2 Total
1
Group I A B A+B
Group II C D C+D
Total A+C B+D N
or for an m (rows) x n (columns) contingency table

( R1 ! R2 ! ⋯ Rm ! ) ( C 1 !C 2 ! ⋯C n ! )
p=
N ! ∏ aij
i, j

with row totals R1 , R 2 , ⋯ , Rm , column totals C 1 , C2 , ⋯ , C n, grand total N, and observations in each
cell a ij. Note that this is a multivariate generalization of the hypergeometric probability function.

4. State the conclusion and interpretation.

Example: Determine if there is a significant association between the journal and type of article
appearing. Use 5% level of significance.

Article Type/Journal Math Journal Science Journal

Applied Math 6 0
Biology 1 8

Solution:
Note that there are cells with expected values less than 5; hence, Fischer’s exact probabilities
test will be used.

Observed
Article Type/Journal Math Journal Science Journal Total
Applied Math 6 0 6
Biology 1 8 9
Total 7 8 15

Expected (e)
Article Type/Journal Math Journal Science Journal Total
Applied Math 2.8 3.2 6
Biology 4.2 4.8 9
Total 7 8 15

row total × column total

Note: e=
N

Step 1: Hypotheses
Ho: The article type appearing is independent of journal. (There is no significant association
between journal and type of article appearing.)
Ha: The article type appearing is dependent of journal. (There is a significant association
between journal and type of article appearing.)

Step 2: Critical value.

Reject Ho if p ≤ α = 0.05.
Step 3: Test statistic
1. Compute the marginal totals.
2. Compute p.

Article Type/Journal Math Journal Science Journal Total

Applied Math 6 0 6
Biology 1 8 9
Total 7 8 15

6 ! 9! 7 ! 8!
p= =0.0014
15 !6 ! 1! 0! 8!

Note that this is only the result for a one-tailed test (opposite direction).
To compute for the two-tailed test, we need to determine all possible, extreme cases (that is,
their p-value is less than the computed p-value for the one-tailed test) without changing the
marginal totals.

The extreme case for the given problem is

Article Type/Journal Math Journal Science Journal Total

Applied Math 0 6 6
Biology 7 2 9
Total 7 8 15

6 ! 9! 7 ! 8!
p= =0.0060
15 !0 ! 7 ! 6 ! 2 !

The total p-value for the two tailed test is 0.0074.

Step 4: Decision and Conclusion/Interpretation

Since p = 0.007 4 < α = 0.05, reject Ho. There is a significant association between journal and
type of article appearing.

Example: Determine whether passing rate is independent of the review centers. Use 5% level
of significance.
Review Center A Review Center B
Passed 2 7
Failed 9 3

Solution:
Note that there are cells with expected values less than 5; hence, Fischer’s exact probabilities
test will be used.

Observed
Review Center Review Center B Total
A
Passed 2 7 9
Failed 9 3 12
Total 11 10 21

Expected
Review Center Review Center B Total
A
Passed 4.71 4.29 9
Failed 6.29 5.71 12
Total 11 10 21

Step 1: Hypotheses
Ho: Passing rate is independent of the review centers
Ha: Passing rate is dependent of the review centers

Step 2: Critical value.

Reject Ho if p ≤ α = 0.05.

Step 3: Test statistic

1. Compute the marginal totals.
2. Compute p.

Review Center Review Center B Total

A
Passed 2 7 9
Failed 9 3 12
Total 11 10 21

9 !12!11 ! 10 !
p= =0.0225
21 ! 2! 9! 7 !3 !

Note that this is only the result for a one-tailed test (opposite direction).
To compute for the two-tailed test, we need to determine all possible, extreme cases (that is,
their p-value is less than the computed p-value for the one-tailed test) without changing the
marginal totals.

The extreme cases for the given problem are

 Review Center Review Center B Total
A
Passed 8 1 9
Failed 3 9 12
Total 11 10 21

9 ! 12! 11 ! 10 !
p= =0.0056
21 !8 ! 3! 1! 9 !

Review Center Review Center B Total

A
Passed 9 0 9
Failed 2 10 12
Total 11 10 21

9 ! 12!11 !10 !
p= =0.0002
21 ! 9 !0 ! 2!10 !

The total p-value for the two tailed test is 0.0283.

Step 4: Decision and Conclusion/Interpretation

Since p = 0.0283 < α = 0.05, reject Ho. Passing rate is dependent of the review centers.
Lesson No. 2: Chi-Square Tests for Two Independent Samples

Learning Objectives:
At the end of the lesson, you will be able to:
1. discuss the assumption and conditions of using the chi-square test for two independent
samples
2. perform chi-square test for two independent samples
3. explain the results of the chi-square test for two independent samples

The chi-square test is applicable to data in a contingency table only if the expected
frequencies are sufficiently large. It is used to determine if there is a significant relationship
between two nominal variables.

Method
1. State the hypotheses.
Ho: There is no association between the two variables.
Ha: There is an association between the two variables.

2. State the rejection rule.

2 2
Reject Ho if χ c ≥ χ α ,df ; df =( r−1 ) ( c−1 )
Refer to Table F for the critical values.

3. Find the value of the test statistic.

r c
oij −e ij
χ c =∑ ∑
2

i=1 j =1 e ij
where o ij is the observed number of cases in the ith row and jth column and e ij is the expected
number of cases in the ith row and jth column.

4. State the conclusion and interpretation.

Example: The data below shows the student’s preference of learning styles and reader
classification. Are the students’ learning styles preference associated with their reader
classification? Use 5% level of significance.
 Reader
Classification / Auditory Visual Kinesthetics
Learning Style
Non-reader 25 25 42
Intermediate 28 64 34
Independent 45 15 15

Solution:
Step 1: Hypotheses
Ho: Students’ learning style preferences are not associated with their reader
classification.
Ha: Students’ learning style preferences are associated with their reader classification.

Step 2: Critical Value

The degree of freedom is df =( 3−1 ) ( 3−1 )=4 and level of significance is α = 0.05. Reject Ho if
2 2
χ c ≥ χ 0.05,4=9.488

Step 3: Test statistic

Reader Classification /
Auditory Visual Kinesthetics Total
Learning Style
Non-reader 2 30.77 25 32.6 42 28.57 92
5 6
Intermediate 2 42.14 64 44.7 34 39.13 126
8 2
Independent 4 25.09 15 26.6 15 23.29 75
5 2
Total 98 104 91 293

( 25−30.77 )2 ( 28−42.14 )2 ( 15−23.29 )2

χ 2c = + +⋯+ =46.74
30.77 42.14 23.29

The probability value is < 0.001.

2 2
Step 4: Since χ c =46.74 ≥ χ 0.05,4 =9.488 and p<0.001 (which is less than α = 0.05), reject Ho.
Students’ learning style preferences are associated with their reader classification.
Lesson No. 3: Median Test

Learning Objectives:
At the end of the lesson, you will be able to:
1. discuss the assumption and conditions of using the median test
2. perform median test
3. explain the results of the median test

The median test is a procedure for testing whether two independent groups (not necessarily
of the same size) differ in central tendencies (median). The test may be used whenever the
scores for the two groups are in at least an ordinal scale. This is considered a two-sample sign
test.

Method
1. State the hypotheses.
Ho: The two groups are from population with the same median.
Ha: The two groups are from population with different medians.
The median of one population is higher (or lower) than the other.

2. State the rejection rule.

Refer to the rule under Fisher test or χ 2 test.

3. Find the value of the test statistic.

 Use χ 2 test when n1 +n 2> 40
 Use χ 2 test when n1 +n 2 is between 20 and 40 with no cell of expected frequency less
than 5; else, Fisher test.
 Use Fisher test if n1 +n 2<20
4. State the conclusion and interpretation.

Example: Determine whether the control and experimental groups have the same median.
Control 11 1 9 4 34 17 1 14 1 13 2 31
5 8 2 6
Experimenta 34 3 35 2 28 12 1 30 1 22 1 29
l 1 9 8 4 0

Solution:
First, compute the median of the combination of two samples. Then, create a 2 x 2 contingency
table whose first row consist of the number of elements in each sample that are greater than the
median and whose second row consists of the number of elements in each sample that is less
than or equal to the median.

Control 11 1 9 4 34 17 1 14 1 13 2 31
5 8 2 6
Experimenta 34 3 35 2 28 12 1 30 1 22 1 29
l 1 9 8 4 0

The median for the combined samples is 18. So, the 2 x 2 contingency table will be:

Contro Experimental
l
> median 3 8
≤ median 9 4

Note that the expected values are greater than 5.

Step 1: Hypotheses
Ho: The two groups are from population with the same median.
Ha: The two groups are from population with different medians.

Step 2: Critical value

The degree of freedom is df =( 2−1 ) ( 2−1 )=1 and level of significance is α = 0.05. Reject Ho if
2 2
χ c ≥ χ 0.05,1=3.841

Step 3: Test statistic

Control Experimental Total
> median 3 5.5 8 5.5 11
≤ median 9 6.5 4 6.5 13
Total 12 12 24

2 ( 3−5.5 )2 ( 9−6.5 )2 ( 8−5.5 )2 ( 4−6.5 )2

χ c= + + + =4.1958
5.5 6.5 5.5 6.5
The probability value is 0.0405.

2 2
Step 4: Since χ c =4.1958 ≥ χ 0.05,4=3.841 and p=0.0405 (which is less than α = 0.05), reject Ho.
The two groups are from population with different medians.

Lesson No. 4: Mann-Whitney U Test

Learning Objectives:
At the end of the lesson, you will be able to:
1. discuss the assumption and conditions of using the Mann-Whitney U Test
2. perform Mann-Whitney U Test
3. explain the results of the Mann-Whitney U Test

The Mann-Whitney test is a nonparametric alternative to the t-test of two independent

samples which requires only the assumption of any continuous and at least the data is
measured on an ordinal scale.

Method
1. State the hypotheses.
Ho: The two groups are equal.
Ha: The two groups are not equal.
One group is higher (or lower) than the other.

2. State the rejection rule.

Reject Ho if |z c|≥|z crit| .

3. Find the value of the test statistic.

 Compute U.
 n 1 ( n1 +1 )
U =n1 n2+ −R 1
2
 Compute z.
U−μ U
z=
σU
where n1 , n2 are the sample sizes of the two groups, R 1 is the sum ranks of the

observations in the first sample, μU =

n1 n2
2 √
, σ U = 1 2 ( 1 2 ) for no ties in ranks or
n n n +n +1
12

√( n1 n 2
)( )
3 3
N −N t −t
σ U= −∑ T for tied ranks such that N=n1+ n1 and T = for t is the
N ( N −1 ) 12 12
number of observations tied for a given rank.

4. State the conclusion and interpretation.

Example: (no ties)

Determine if there is a significant difference on the running speed of student athletes in a 100-
meter sprint during in the last two years. Use 5% level of significance.
Year 1 8.7 9.0 9.15 9.10 9.12 9.03 9.04 8.83 8.81
Year 2 9.11 8.75 9.14 9.05 9.07 8.83 8.84 8.79 8.80

Solution:
Step 1. State the hypotheses.
Ho: The running speeds of the two groups of student athletes are equal.
Ha: The running speeds of the two groups of student athletes are not equal.

Step 2. State the rejection rule.

Reject Ho if |z c|≥|z crit|=1.96 . (two-tailed test)

Step 3. Find the value of the test statistic.

 Rank the values from smallest to highest. Then, compute R1.
 Compute U and z.

Year 1 8.70 9.00 9.15 9.10 9.12 9.03 9.04 8.83 8.81
Rank 1 9 18 14 16 10 11 6 5
Year 2 9.11 8.75 9.14 9.05 9.07 8.88 8.84 8.79 8.80
Rank 15 2 17 12 13 8 7 3 4

R1=1+ 9+18+14+16+ 10+11+6+5=90

9(9+ 1)
U =9(9)+ −90=36
2
 9 (9 )
μU = =40.5
2

σ U=
√
9 ( 9 )( 9+ 9+1 )

36−40.5
12
=11.324752

z= =−0.3974
11.324752

Step 4: Decision and Conclusion/Interpretation

Since |z c|=0.3974<|z crit|=1.96, fail to reject Ho. The running speeds of the two groups of
student athletes are equal.

Example: (with ties)

Grade 3 pupils from two different schools were given the acceleration test. Test at 5% level of
significance that the two samples came from the same population.

Section 1 45 36 38 44 48 39 37 30 34 33 45 56 39 31 33
Section 2 46 44 32 28 26 34 49 36 35 44 36 38 32

Solution:
Step 1: Hypotheses
Ho: Scores in the acceleration test of the two groups of Grade 3 pupils are equal.
Ha: Scores in the acceleration test of the two groups of Grade 3 pupils are not equal.

Step 2: State the rejection rule.

Reject Ho if |z c|≥|z crit|=1.96 . (two-tailed test)

Step 3
 Rank the values from smallest to highest. Then, compute R1.
 Compute U and z.

Section 1 45 3 38 44 48 39 3 30 34 33 45 56 39 31 33
6 7
Rank 23.5 1 16.5 21 26 18.5 1 3 9.5 7.5 23.5 28 18.5 4 7.5
3 5
Section 2 46 4 32 28 26 34 4 36 35 44 36 38 32
4 9
Rank 25 2 5.5 2 1 9.5 2 13 11 21 13 16.5 5.5
1 7
 R1=23.5+ 13+ 16.5+21+26+18.5+15+3+ 9.5+7.5+23.5+28+18.5+ 4+7.5=235
15(15+ 1)
U =15(13)+ −235=80
2
15 ( 13 )
μU = =97.5
2
3
−2 23−2 23−2 33−3 23−2 23−2 33−3 23−2
∑ T = 2 12 +
12
+
12
+
12
+
12
+
12
+
12
+
12
=7

√( )( )
3
15 × 13 28 −28
σ U= −7 =21.666667
28 ( 28−1 ) 12
80−97.5
z= =−0.80769
21.666667

Step 4: Decision and Conclusion/Interpretation

Since |z c|=0.80769<|z crit|=1.96 , fail to reject Ho. Scores in the acceleration test of the two
groups of Grade 3 pupils are equal.

Lesson No. 5: Kolmogorov-Smirnov Two-Sample Test

Learning Objectives:
At the end of the lesson, you will be able to:
1. discuss the assumption and conditions of using the Kolmogorov-Smirnov Two-sample
test
2. perform Kolmogorov-Smirnov Two-sample test
3. explain the results of the Kolmogorov-Smirnov Two-sample test

The two-sample Kolmogorov-Smirnov test is used to test whether two samples come from
the same distribution. The procedure is similar to the one-sample Kolmogorov-Smirnov test. The
two tailed test is sensitive to any kind of difference in the distributions from which the two
samples were drawn. The one-tailed test is used to decide whether the values of the population
from which one of the samples was drawn are stochastically larger than the values of the
population from which the other sample was drawn.
Method
1. State the hypotheses.
Ho: Both samples come from a population with the same distribution.
Ha: The samples come from population with different distribution.

2. State the rejection rule.

Reject Ho if D ≥ D α . (Refere to Table G1- Equal Size; Table G2- Unequal Size)
2 2
Reject Ho if χ c ≥ χ α ,df ; df =2 (Refer to Table F)

3. Find the value of the test statistic.

D=max∨S n ( X ) −S n ( X )∨¿
1 2

2 2 n1 n2
χ c =4 D
n 1 + n2

4. State the conclusion and interpretation.

Example: Determine whether the two samples come from the same distribution. Use 5% level
of significance.

Age Men Women

21-22 4 7
23-24 11 4
25-26 5 1
27-28 7 11
29-30 0 12
31-32 5 4
33-34 9 2
35-36 13 4
37-38 20 8
39-40 6 9

Solution:
Step 1: Hypotheses
Ho: Both samples come from a population with the same distribution.
Ha: The samples come from population with different distribution.

Step 2: Rejection rule

Reject Ho if D ≥ D α =0.230114 .
 2 2
Reject Ho if χ c ≥ χ 0.05 ,2=5.991

Step 3: Test statistic

Age Me Women Pn ( X )
1
Pn ( X )
1
Sn ( X )
1
Sn ( X )
2
Sn ( X )−S n ( X )
1 2

n
21-22 4 7 0.050
0.0500 0.1129 0 0.1129 0.0629
23-24 11 4 0.187
0.1375 0.0645 5 0.1774 0.0101
25-26 5 1 0.250
0.0625 0.0161 0 0.1935 0.0565
27-28 7 11 0.337
0.0875 0.1774 5 0.3710 0.0335
29-30 0 12 0.337
0.0000 0.1935 5 0.5645 0.2270
31-32 5 4 0.400
0.0625 0.0645 0 0.6290 0.2290
33-34 9 2 0.512
0.1125 0.0323 5 0.6613 0.1488
35-36 13 4 0.675
0.1625 0.0645 0 0.7258 0.0508
37-38 20 8 0.925
0.2500 0.1290 0 0.8548 0.0702
39-40 6 9 1.000
0.0750 0.1452 0 1.0000 0.0000
80 62

D=max∨S n ( X ) −S n ( X )∨¿ 0.2290

1 2

10(10)
χ 2c =4 (0.2290)2 =1.04882
10+ 10

Step 4: Decision and Conclusion/Interpretation

2 2
Since D=0.2290< D α =0.230114 and χ c =1.04882< χ 0.05 ,2=5.991, fail to reject Ho. Both
samples come from a population with the same distribution.

Lesson No. 6: Wald-Wolfowitz Runs Test

Learning Objectives:
At the end of the lesson, you will be able to:
1. discuss the assumption and conditions of using the Wald-Wolfowitz Runs test
2. perform Wald-Wolfowitz Runs test
3. explain the results of the Wald-Wolfowitz Runs test

The Wald-Wolfowitz Runs test (or two-sample runs test) is used to determine whether two
samples come from the same distribution. The test orders the values in the combined sample
creating a sequence and then using the one-tailed version of the one-sample runs test.

If there are ties, then the number of runs will differ depending on how the tied values are
ordered. In this case, multiple versions of the test will be performed changing the orders of the
entries with tied ranks.

Note, however, that when there is a significant difference between the distributions of the
two samples, it will not determine differences in the means, medians, variances, skewness,
kurtosis, etc.

Method
1. State the hypotheses.
Ho: Both samples come from a population with the same distribution. (The variable A is the
same for the two groups)
Ha: The samples come from population with different distribution. (The variable A is different for
the two groups)

2. State the rejection rule.

 Small-sample case ( n 1 , n2 ≤20 ). Use Table E. Reject Ho if Gdata ≤ G crit ,lower or
Gdata ≥ Gcrit ,upper
 Large-sample case ( n 1 , n2 >20 ) . Use the standard normal table. Reject Ho if |z data|≥|z crit|

3. Find the value of the test statistic.

 Small-sample case ( n 1 , n2 ≤20 ). G data ≤ G ( numberofruns )
 Large-sample case ( n 1 , n2 >20 ) . Find the number of runs G. Then, calculate the
following:
2 n1 n2
μG = +1
n1 + n2

σ G=
√ ( 2 n1 n 2) ( 2 n1 n2−n 1−n2 )
2
( n1+ n2 ) ( n1 +n2−1 )

G−μ G
z data=
σG
4. State the conclusion and interpretation.

Example: The following tables contains the unemployment rate for independent random
samples of cities in two regions. Test whether the samples come from a population with
the same distribution. Use 5% level of significance.

Region A Unemployment Rate Region B Unemployment Rate

City 1 6.6 City 7 2.8
City 2 6.4 City 8 4.6
City 3 7.9 City 9 4.4
City 4 5.4 City 10 5.3
City 5 7.6 City 11 4.2
City 6 7.5 City 12 7.3
City 13 3.4

Solution:
Step 1: Hypotheses
Ho: Both samples come from a population with the same distribution. (The unemployment rate
is the same for the two groups.)
Ha: The samples come from population with different distribution. (The unemployment rate is
different for the two groups.)

Step 2: Rejection rule

Reject Ho if if Gdata ≤ Gcrit ,lower =3 or Gdata ≥ G crit ,upper =12

Step 3: Test statistic

Score 2.8 3.4 4.2 4.4 4.6 5.3 5.4 6.4 6.6 7.3 7.5 7.6 7.9
Group B B B B B B A A A B A A A
Run 1 2 3 4

Gdata =4

Step 4: Decision and Conclusion/Interpretation

Since G data =4 >G crit ,lower =3, fail to reject Ho. Both samples come from a population with the
same distribution. The unemployment rate is the same for the two groups.

Example: The following are test scores of two sets of kindergarten pupils. Determine if the two
groups of pupils differ in their test scores. Use 1% level of significance.

Group A Group B
20 23
 55 8
29 24
24 15
75 8
56 6
31 15
45 15
21
23
16
15
24
15
21
15
18
14
22
15
14

Solution:
Step 1: Hypotheses.
Ho: There is no significant difference between Group A and Group B with test scores.
Ha: There is a significant difference between Group A and Group B with test scores.

Step 2: Rejection rule.

Reject Ho if |z data|≥|z crit|=2.33

Step 3: Test statistic

Score 8 8 8 14 14 15 15 15 15 15 15 15 16 18 20
Group B B B B B B B B B B B B B B A
Run 1 2

Score 21 21 22 23 23 24 24 24 29 31 45 55 56 75
Group B B B B B B A B A A A A A A
Run 3 4 5 6

Note: This is only one possible sequence of runs, since there are 2 other possibility for ties of
rank (score 24: ABB = 6 and BBA = 5).

Gdata =6
 2( 8)(21)
μG = +1=12.586207
8+21

σ G=
√ ( 2× 8× 21 )( 2 × 8× 21−8−21 )
2
( 8+21 ) ( 8+21−1 )
=2.092964

6−12.586207
z data= =−3.1468
2.092964

Step 4: Decision and Conclusion/Interpretation

Since |z data|=3.1468 ≥|z crit|=2.33 , reject Ho. There is a significant difference between Group A
and Group B with test scores.

WORKSHEET NO. 3

Directions: Solve the following problems completely.

1. Determine if men seem to diet less than women do based on the following data. Use 5% level
of significance.

Me Women
n
Dieting 1 8
Not dieting 9 2

2. Consider the random samples of elementary pupils, classified according to grades, who
watched TV for a certain week. Test the proportions if the number of pupils watching TV is not
the same among grade school pupils. Use 5% level of significance.

Grades Watching TV Not watching TV

Grade 15 8
3
Grade 20 7
4
Grade 25 6
5
Grade 15 4
6

3. Determine whether the groups of students have the same median. The following are test
scores in a 35-item test in Mathematics.
Section A 2 14 22 23 25 13 11 15 18 17 9 30
0
Section B 2 32 17 17 18 11 26 15 20 15 18 32
5

4. Compare the whether the scores of the students from two universities differ from each other.
Use 5% level of significance.

Group A 135 139 142 144 158 165 171 178 244 245 256 267 268 289
Group B 131 138 138 141 142 142 143 145 156 167 191 230

5. In a photograph test to identify the low and high authoritarian personality the following data
have been recorded. Determine if persons with high authoritarian personality stereotyped more
photographs than persons with low authoritarian personality. Use 1% level of significance.

Number of photos stereotyped Low High scorers

scorers
0-2 11 1
3-5 7 3
6-8 8 6
9-11 3 12
12-14 5 12
15-17 5 14
18-20 5 6
6. Determine whether the samples below come from the same distribution. Use 1% level of
significance.

Sample A Sample B
11 25
19 19
18 23
12 21
11 22
12 35
11 37
15 31
21 39
22 31
20 35
21 25
23 22
19
17
23
24
10
11
36
30
26
27
27
Table F. Chi-Square Distribution
   α
df 0.995 0.975 0.20 0.10 0.05 0.025 0.02 0.01 0.005 0.002 0.001
1 0.0000393 0.000982 1.642 2.706 3.841 5.024 5.412 6.635 7.879 9.550 10.828
2 0.0100 0.0506 3.219 4.605 5.991 7.378 7.824 9.210 10.597 12.429 13.816
3 0.0717 0.216 4.642 6.251 7.815 9.348 9.837 11.345 12.838 14.796 16.266
4 0.207 0.484 5.989 7.779 9.488 11.143 11.668 13.277 14.860 16.924 18.467
5 0.412 0.831 7.289 9.236 11.070 12.833 13.388 15.086 16.750 18.907 20.515
6 0.676 1.237 8.558 10.645 12.592 14.449 15.033 16.812 18.548 20.791 22.458
7 0.989 1.690 9.803 12.017 14.067 16.013 16.622 18.475 20.278 22.601 24.322
8 1.344 2.180 11.030 13.362 15.507 17.535 18.168 20.090 21.955 24.352 26.124
9 1.735 2.700 12.242 14.684 16.919 19.023 19.679 21.666 23.589 26.056 27.877
1 2.156 3.247 13.442 15.987 18.307 20.483 21.161 23.209 25.188 27.722 29.588
0
1 2.603 3.816 14.631 17.275 19.675 21.920 22.618 24.725 26.757 29.354 31.264
1
1 3.074 4.404 15.812 18.549 21.026 23.337 24.054 26.217 28.300 30.957 32.909
2
1 3.565 5.009 16.985 19.812 22.362 24.736 25.472 27.688 29.819 32.535 34.528
3
1 4.075 5.629 18.151 21.064 23.685 26.119 26.873 29.141 31.319 34.091 36.123
4
1 4.601 6.262 19.311 22.307 24.996 27.488 28.259 30.578 32.801 35.628 37.697
5
1 5.142 6.908 20.465 23.542 26.296 28.845 29.633 32.000 34.267 37.146 39.252
6
1 5.697 7.564 21.615 24.769 27.587 30.191 30.995 33.409 35.718 38.648 40.790
7
1 6.265 8.231 22.760 25.989 28.869 31.526 32.346 34.805 37.156 40.136 42.312
8
1 6.844 8.907 23.900 27.204 30.144 32.852 33.687 36.191 38.582 41.610 43.820
9
2 7.434 9.591 25.038 28.412 31.410 34.170 35.020 37.566 39.997 43.072 45.315
0
2 8.034 10.283 26.171 29.615 32.671 35.479 36.343 38.932 41.401 44.522 46.797
1
2 8.643 10.982 27.301 30.813 33.924 36.781 37.659 40.289 42.796 45.962 48.268
2
2 9.260 11.689 28.429 32.007 35.172 38.076 38.968 41.638 44.181 47.391 49.728
3
2 9.886 12.401 29.553 33.196 36.415 39.364 40.270 42.980 45.559 48.812 51.179
4
2 10.520 13.120 30.675 34.382 37.652 40.646 41.566 44.314 46.928 50.223 52.620
5
2 11.160 13.844 31.795 35.563 38.885 41.923 42.856 45.642 48.290 51.627 54.052
6
2 11.808 14.573 32.912 36.741 40.113 43.195 44.140 46.963 49.645 53.023 55.476
7
2 12.461 15.308 34.027 37.916 41.337 44.461 45.419 48.278 50.993 54.411 56.892
8
2 13.121 16.047 35.139 39.087 42.557 45.722 46.693 49.588 52.336 55.792 58.301
9
3 13.787 16.791 36.250 40.256 43.773 46.979 47.962 50.892 53.672 57.167 59.703
   α
df 0.995 0.975 0.20 0.10 0.05 0.025 0.02 0.01 0.005 0.002 0.001
0
3 14.458 17.539 37.359 41.422 44.985 48.232 49.226 52.191 55.003 58.536 61.098
1
3 15.134 18.291 38.466 42.585 46.194 49.480 50.487 53.486 56.328 59.899 62.487
2
3 15.815 19.047 39.572 43.745 47.400 50.725 51.743 54.776 57.648 61.256 63.870
3
3 16.501 19.806 40.676 44.903 48.602 51.966 52.995 56.061 58.964 62.608 65.247
4
3 17.192 20.569 41.778 46.059 49.802 53.203 54.244 57.342 60.275 63.955 66.619
5
3 17.887 21.336 42.879 47.212 50.998 54.437 55.489 58.619 61.581 65.296 67.985
6
3 18.586 22.106 43.978 48.363 52.192 55.668 56.730 59.893 62.883 66.633 69.346
7
3 19.289 22.878 45.076 49.513 53.384 56.896 57.969 61.162 64.181 67.966 70.703
8
3 19.996 23.654 46.173 50.660 54.572 58.120 59.204 62.428 65.476 69.294 72.055
9
4 20.707 24.433 47.269 51.805 55.758 59.342 60.436 63.691 66.766 70.618 73.402
0
4 21.421 25.215 48.363 52.949 56.942 60.561 61.665 64.950 68.053 71.938 74.745
1
4 22.138 25.999 49.456 54.090 58.124 61.777 62.892 66.206 69.336 73.254 76.084
2
4 22.859 26.785 50.548 55.230 59.304 62.990 64.116 67.459 70.616 74.566 77.419
3
4 23.584 27.575 51.639 56.369 60.481 64.201 65.337 68.710 71.893 75.874 78.750
4
4 24.311 28.366 52.729 57.505 61.656 65.410 66.555 69.957 73.166 77.179 80.077
5
4 25.041 29.160 53.818 58.641 62.830 66.617 67.771 71.201 74.437 78.481 81.400
6
4 25.775 29.956 54.906 59.774 64.001 67.821 68.985 72.443 75.704 79.780 82.720
7
4 26.511 30.755 55.993 60.907 65.171 69.023 70.197 73.683 76.969 81.075 84.037
8
4 27.249 31.555 57.079 62.038 66.339 70.222 71.406 74.919 78.231 82.367 85.351
9
5 27.991 32.357 58.164 63.167 67.505 71.420 72.613 76.154 79.490 83.657 86.661
0
5 28.735 33.162 59.248 64.295 68.669 72.616 73.818 77.386 80.747 84.943 87.968
1
5 29.481 33.968 60.332 65.422 69.832 73.810 75.021 78.616 82.001 86.227 89.272
2
5 30.230 34.776 61.414 66.548 70.993 75.002 76.223 79.843 83.253 87.507 90.573
3
5 30.981 35.586 62.496 67.673 72.153 76.192 77.422 81.069 84.502 88.786 91.872
4
5 31.735 36.398 63.577 68.796 73.311 77.380 78.619 82.292 85.749 90.061 93.168
5
5 32.490 37.212 64.658 69.919 74.468 78.567 79.815 83.513 86.994 91.335 94.461
6
5 33.248 38.027 65.737 71.040 75.624 79.752 81.009 84.733 88.236 92.605 95.751
7
   α
df 0.995 0.975 0.20 0.10 0.05 0.025 0.02 0.01 0.005 0.002 0.001
5 34.008 38.844 66.816 72.160 76.778 80.936 82.201 85.950 89.477 93.874 97.039
8
5 34.770 39.662 67.894 73.279 77.931 82.117 83.391 87.166 90.715 95.140 98.324
9
6 35.534 40.482 68.972 74.397 79.082 83.298 84.580 88.379 91.952 96.404 99.607
0
6 36.301 41.303 70.049 75.514 80.232 84.476 85.767 89.591 93.186 97.665 100.888
1
6 37.068 42.126 71.125 76.630 81.381 85.654 86.953 90.802 94.419 98.925 102.166
2
6 37.838 42.950 72.201 77.745 82.529 86.830 88.137 92.010 95.649 100.182 103.442
3
6 38.610 43.776 73.276 78.860 83.675 88.004 89.320 93.217 96.878 101.437 104.716
4
6 39.383 44.603 74.351 79.973 84.821 89.177 90.501 94.422 98.105 102.691 105.988
5
6 40.158 45.431 75.424 81.085 85.965 90.349 91.681 95.626 99.330 103.942 107.258
6
6 40.935 46.261 76.498 82.197 87.108 91.519 92.860 96.828 100.554 105.192 108.526
7
6 41.713 47.092 77.571 83.308 88.250 92.689 94.037 98.028 101.776 106.440 109.791
8
6 42.494 47.924 78.643 84.418 89.391 93.856 95.213 99.228 102.996 107.685 111.055
9
7 43.275 48.758 79.715 85.527 90.531 95.023 96.388 100.425 104.215 108.929 112.317
0
7 44.058 49.592 80.786 86.635 91.670 96.189 97.561 101.621 105.432 110.172 113.577
1
7 44.843 50.428 81.857 87.743 92.808 97.353 98.733 102.816 106.648 111.412 114.835
2
7 45.629 51.265 82.927 88.850 93.945 98.516 99.904 104.010 107.862 112.651 116.092
3
7 46.417 52.103 83.997 89.956 95.081 99.678 101.07 105.202 109.074 113.889 117.346
4 4
7 47.206 52.942 85.066 91.061 96.217 100.839 102.24 106.393 110.286 115.125 118.599
5 3
7 47.997 53.782 86.135 92.166 97.351 101.999 103.41 107.583 111.495 116.359 119.850
6 0
7 48.788 54.623 87.203 93.270 98.484 103.158 104.57 108.771 112.704 117.591 121.100
7 6
7 49.582 55.466 88.271 94.374 99.617 104.316 105.74 109.958 113.911 118.823 122.348
8 2
7 50.376 56.309 89.338 95.476 100.749 105.473 106.90 111.144 115.117 120.052 123.594
9 6
8 51.172 57.153 90.405 96.578 101.879 106.629 108.06 112.329 116.321 121.280 124.839
0 9
Complete table can be accessed via https://www.medcalc.org/manual/chi-square-table.php
Table G1. Two-sample Kolmogorov-Smirnov Test (Equal Size)
Table G2. Two-sample Kolmogorov-Smirnov Test (Unequal Size)