PARAMETRIC TEST NON-PARAMETRIC TEST

Paired Sample T-Test (Compares two Wilcoxon Signed-Rank Test

dependent samples
Two Independent samples T-Test Wilcoxon Rank Sum (Mann-Whitney
Test)

Null Hypothesis Alternative

H0 Hypothesis
Ha or H1
Left-Tailed The population The population
median (η) is greater median(η) is less
than or equal to than the
hypothesized hypothesized
median(η0) median(η0)
 η≥ η0
 η<η0
Right-Tailed The population The population
median (η) is less median(η) is greater
than or equal to than the
hypothesized hypothesized
median(η0) median(η0)
 η≤η0
 η>η0
Two-Tailed The population The population
median (η) is equal median(η) is not
to hypothesized equal to the
median(η0) hypothesized
 η=η0 median(η0)
 η≠η0

ONE SAMPLE WILCOXON

Is a rank based test and it compares the standard value (theoretical value) with
hypothesized median.

We reject H0 if the calculated value is less than Critical Value

ASSUMPTIONS:

 Differences between the data value and the hypothesized median are
continuous
 Data follows symmetric distribution
 Observations are mutually independent to each other
 Measurement scale is at least interval
Example 1.

In a law college, a random samples of 10 students marks are noted

below, is there evidence at the 5% confidence level to suggest that the
median mark is greater than 67?

Solution:

Hypotheses:
H0: The population median value ≤ 67 marks
H1: The population median value > 67 marks

Right-Tailed

a. Identify the difference between each individual value and the

median.

If the difference of individual value and median is zero, ignore it.

b. Ignore the signs of the difference values and assign lowest rank to
the smallest difference value. If the values have tied, then consider
the mean value.

c. Compute the Sum of Ranks:

 POSITIVE Difference Values

NEGATIVE Difference Values

d. Compute the sum of ranks of positive difference values, and

negative difference values (W+ and W-).
 Therefore the Test Statistic, W, is 15 (Because it is the lower/smaller
of the two Rank Sums (W+= 40 & W-= 15), see Table above)

e. , Right-Tailed & n= 10
 n = 10

From the Wilcoxon Table: Since it is One-Tailed, has to

Double the value (0.05*2=0.10)

Critical Value (CV(0.10,10) = 10

f. Compare the test statistic, W, with the critical value in the tables;
the null hypothesis can be rejected if W is less than or equal to the
critical value. Interpret the decision in the context of the original
claim.

Since,

The Test Statistic, W, (15) > the Critical Value (10), then we
Fail to Reject the Null Hypotheses. There is no significant
evidence that the median rank is greater than 67.

82=0.10)
Example 2.

Bank of America West Palm Beach, FL branch manager indicates that the
median number of savings account customers per day is 64. A clerk from
the same branch claims that it was more than 64. Clerk collected the
number of savings account customers per day data for 10 random days.
Can we reject the branch manager’s claim at 0.05 significance level?

Solutions:
Hypotheses:

 Null Hypothesis H0: Savings account customer median = 64;

 Alternative Hypothesis H1: Savings account customer median >64

One-Tailed Test
 Ignore the signs of the difference values and assign lowest rank to the smallest
difference value. If the values have tied, then consider the mean value.

From the Table Below,

@ Column D, there are TWO 4’s at Rows 2 & 6 with Ranks 3 & 4

To compute for the Adjusted Rank: 2 bec. There are 2 Numbers, 3 &
3 + 4= 7
4
Then get the mean value: 7/2 = 3.5

Get the POSITIVE Rank

Get the NEGATIVE Rank

Compute the Rank Sums, W+ & W-

Therefore, the Test Statistics: 13.5 (Because it is the lower/smaller

of the two Rank Sums (W+= 41.5 & W-= 13.5), see Table above)

To find the Critical Value,

G * 2 = 0.10 (Since it is One-Tailed)
Sample Size (n) = 10

From the W Table Below,

The Critical Value, CV(0.10,10) = 10

CONCLUSION:

Since 13.5 > 10, Therefore we Failed to Reject the Null Hypothesis.
There is no significant evidence that the savings account customers per
day are more than 64.
ACTIVITY 1

Can we say that the median value of the Population data is less than
50 from the sample data as follows:
10,20,90,60,58,13,19,73,85,63,29,30,27,84,83,80

WILCOXON SIGNED-RANK TEST FOR PAIRED SAMPLES

 Two samples are matched if each observation in one sample is matched

with an observation in another sample.
 The test does not assume normality and can be used as a nonparametric
alternative to the paired t-test for dependent samples.

Example 3.

Suppose an engineer want to know if a new fuel treatment leads to a change

in the average miles per gallon of a certain car. To test this, he measures the
mpg of 12 cars with and without the fuel treatment as shown below:

Solution: Using Excel

Hypotheses:

H0: The mpg is equal between the two groups

HA: The mpg is not equal between the two groups

Two-Tailed Test

a. Calculate the Difference Between the Groups.

If the difference of individual value and median is zero, ignore it.

EXCEL Formula. But if solve by Hand,

Subtract 24 from 20 (20-24 =-4; 23-25=-2;
…). The result is shown below (See
Column C).

b. Get or Calculate the Absolute value of the Difference.

 Absolute Value of -4 is 4
or (|-4| = 4)

0 Difference is IGNORED

Rank all observations in increasing order of magnitude, ignoring their sign.

Assign the lowest rank to the smallest difference value. If two observations
have the same magnitude, regardless of sign, then they are given an average
ranking
OR
ADJUSTED
RANK

There are 2, 1’s.

Therefore, the
rank will be 1 & 2.
So, 1+2=3; Then
get the average:
3/2=1.5

Note: Further explanations on how Ranks @ Column E are derived (See

Table above) if computation is done by hand:

@ Column D, @ Column D @ Column F @ Column E

For |D| Equal See, The Rank is Get the Average
to
2 Rows 4,13 3+4=7 7/2=3.5
Rows 4 & 13
3 Rows 6,12 5+6=11 11/2=5.5
Rows 6 & 12
 4 Rows 3,10,11, & 14 7+8+9+10=34 34/4 = 8.5
Rows 3,10,11, & 14

4, because there are 4

occurrences of |D|=4 (See col.
E, Rows 3,10,11 & 14

a. Calculate the POSITIVE Ranks

b. Calculate the NEGATIVE Ranks
 c. Calculate the Test Statistic which is simply the smaller of the sum
of the positive ranks or the sum of the negative ranks:

10.5 < 55.5

Therefore, the Test Statistic is 10.5 (10.5 since it is the smaller between
the two Ranks Sum: W+= 10.5 & W- = 55.5)
d. Calculate the sample size, which is the total number of ranks that
are not equal to zero.

Sample
Size = 11
 e. Find the Critical Value:

From the Wilcoxon Signed Rank Test Critical Values Table:

The critical Value = 10

f. Compare the test statistic with the critical value. The null
hypothesis can be rejected if test statistic is less than or equal to
the critical value. Interpret the decision in the context of the original
claim.

CONCLUSION

Since the Test Statistic of 10.5 > the Critical Value of 10, We therefore
Failed to Reject the null Hypothesis. We do not have sufficient evidence to
conclude that the mean mpg is equal between the two groups.
Example 4.

Suppose that a new post-surgical treatment is being compared with a

standard treatment by observing the recovery times of n treatment subjects
and m controls. Suppose the data are matched m=n=9 and that the observed
recovery times (in days) are:

Solution:

Critical Value is 5. (From the Wilcoxon Signed-Ranks Table below)

Since 4 < 5, we reject the Null hypothesis.

ACTIVITY 2.

A research team wants to test whether a new teaching method increases

the literacy of children. Therefore, the researchers take measure the
literacy of 20 children before and after the teaching method has been
applied. The literacy is measured on a scale from 0 to 10, with 10
indicating high literacy. The initial baseline shows an average literacy
score of 5.9 and after the method has been used the average increases to
7.6.
Source: https://www.real-statistics.com/statistics-tables/wilcoxon-signed-ranks-
table/

References:
https://sixsigmastudyguide.com/1-sample-sign-non-parametric-hypothesis-test/

https://www.statisticssolutions.com/free-resources/directory-of-statistical-
analyses/wilcoxon-sign-test/
https://sixsigmastudyguide.com/1-sample-wilcoxon-non-parametric-hypothesis-test/

http://www.analystsoft.com/en/products/statplus/content/help/analysis_nonparam
etric_statistics_comparing_two_dependent_samples_wilcoxon_matched_pairs.html
https://www.statology.org/wilcoxon-signed-rank-test-excel/