1

CTED Statistical Tools

MODULES AND PRACTICUM PAPERS

Part I. Measures of Central Tendency

I will make this lesson very simple because it is very easy and had been learned by you in
your high school. However, it should not be ignored because these tools are foundations
in some descriptive and inferential statistics. In other words, this is just a review.

For every group of scores, there are those ones which are noticed as “most”. It is
these scores which commonly appear, obtained or grouped. Thus, this is called central
tendency which simply serves as an index in the clustering of X-value towards the central
value. There are three measures of central tendency. These are mean, median and mode.

I. Objectives:
At the end of the lesson, the student is expected to:
A. Differentiate the characteristics of mean, median and mode.
B. Determine the mean, median and mode in a given set of data correctly.

II. Topics:
A. Mean
B. Median
C. Mode

III. Presentation

A. Mean or average. This is probably the most common way of

describing central tendency. It is done by adding up all the values and then divided by
the number of values. There are two ways in the determining the mean. When n is
lesser than 30, the ungrouped data is used. When n is 30 or more, the grouped data is
used. (However, if you use excel, the grouped data can be done only after manual
grouping of values, but it is a waste of time. If n is more than 30, just encode them
individually until all are done).

Example: Ungrouped Data

84 79 82 93 76 = 414 Total
 2

414 5 = 82.8 Mean

B. Median. This is the score found exactly at the middle of set of
values. A simple way of looking for the median is to list down all the scores either from
highest to lowest or from lowest to highest. The score located at the center of the samples
is the median. If there are two values at the center, add them then divide it by 2. The
resulting value is the median.
Even number

Illustration: 16

Odd number 18

2 19

3 20 20 + 22 = 42

4 Median 22 42 ÷ 2 = 21 Median

7 23

9 25
number repeated many times
26

C. Mode. This is the score found as the most frequently occurring score
in a set of scores. An easy way to determine the mode is to arrange the scores in either
ascending or descending order.

Example: 44 37 37
32 45 30

37 42 45

33 36 39

30 32 33 36 37 37 37 39 42 44 45 45

What is the most frequently occurring number? 37 Mode

(If there is 1 mode, that is called unimodal; 2=bimodal; 3 =multimodal)

D. Weighted Mean. Weighted values that results to weighted mean uses the Likert
point-scale. The points are expressed in weights with arbitrary descriptions used
depending upon the situation or condition where it is applied. It could be
progressive or regressive when it is one directional; or integeric when descriptions

based on or subject to individual judgment or preference; it

is determined by chance, not by necessity.
 3

contain both positive and negative descriptions. It is also known as two-

directional.

I. Objectives
In this lesson, the student is expected to
A. Calculate values using the Likert-point scale.
B. Make simple interpretations about values obtained in calculations.

II. Topics
A. The Likert-point Scale
B. Calculation of Weighted Mean

III. Presentation

A. The Likert-Point Scale

This is still a part of the lesson on measures of central tendency. The description
of a Likert-point scale is already given in the introductory paragraph. In a sense, it is an
arbitrary way of giving a value or a number to a certain description so that the value or
values can be calculated.

Illustrations:

Progressive/Regressive Integeric
Weight Description Weight Description
5 Excellent 5 Strongly Agree
4 Very Good 4 Agree
3 Good 3 Uncertain
2 Fair 2 Disagree
1 Poor 1 Strongly Disagree

The weights in a Likert point-scale are not always 5. It can be 3 and any number
up to 10 (or more). To facilitate interpretation, each weight (point) or a certain distance
should have equivalent description. Here is an example on how to use a Likert point scale
in determining the weighted mean:

School “D” released the following report in 2011 regarding programs they have
conducted among their students

Programs Always Often Sometimes Rarely Never

5 4 3 2 1 ∑fx Wm Rank Interpretation
Deworming
27 51 19 13 10 432 3.60 2 Often
 4

Supplementary 15 28 42 30 5 378 3.15 1 Sometimes

Feeding
Conducting Phil-
IRI test 29 47 30 7 7 444 3.70 3 Often
Gulayan sa
Paaralan 77 36 4 2 1 546 4.55 4 Always
Overall Mean 3.75 Often
N=120

To calculate using a scientific calculator=

(27x5)+(51x4)+(19x3)+(13x2)+(10x1)=∑fx÷N=Wm

Calculation of Weighted Mean (Wm)

Procedure of Calculation:

1. Determine the number of cases (N) per row or horizontally.

2. Multiply the frequency per cell by the weight of that cell. This will give
you the partial product for the item.
3. Add the partial products to get the total products. For the item.
4. Divide the total product by the number of cases (N) to get the weighted
mean value.
5. Give the appropriate description about the weighted mean value obtained.

The formula or rule is simple:

∑ fx
Wm = where: Wm is the weighted mean value
N
∑ is the summation sign
f is the frequency
x is the weight
N is the number of cases
∑fx is the sum of the partial products

Some simple analyses that could be derived from the values are:

1. Gulayan sa Paaralan appears to be “Always” while supplementary feeding is

“Sometimes” a program conducted by School D in 2011.

2. Generally, School D “Often” conducted the four programs in 2011.

Note:

To match the Wm- values, you have to make a scale. In this example, subtract the highest
value (5) by the lowest value (1). So 5-1=4. Divide 4 by the highest value (5) and the
resulting value is 0.8. this will be used as the interval. Thus:
 5

Weigth/Interval Interpretation
4.21-4.00 Always
3.41-4.20 Often
2.61-3.40 Sometimes
1.81-2.60 Rarely
1.00-1.80 Never

Electronic calculation will be introduced to you when we meet.

Score

Practicum Paper
7
Measures of Central Tendency

Name: Via Bianca R. Beguia

A. What is the mean in the following set of scores?

1. 32 35 31 28 49 43 42
Total: 260
Mean:37.14
 6

2. Jun got the following scores in quizes:

Earth Science= 92; Literature= 81; General Physics= 79; General Chemistry= 84
What is the mean of his marks?
Mean:84

B. What is the median in each set of scores?

1. 79 82 73 81 85 80 77
73 77 79 80 81 82 85
Median- 80

2. 101 93 95 112 110 96

93 95 96 101 110 112
96 and 101
Median- 98.5/99

C. What is the mode in the following set of scores?

63 61 48 57
59 57 45 68
62 75 53 77
72 76 79 52
57 57 49 80

45 48 49 52 53 57 57 57 57 59 61 62 63 68 72 75 76 77 79
80
Mode: 57
Multimodal

D. The ages of teachers with osteoporosis are 43; 32; 36; 41; 30; 45; 47; 29

1. What is their average age? 38

Calculation: 43,32,36,41,30,45,47,29 total= 303 ÷ 8= 38

2. What is their median age? 39

29 30 32 36 41 43 45 47
36 and 41 =36+41=77÷2=38.5
 7

Score

18
Practicum Paper

Weighted Mean and Weighted Values

Name: Via Bianca R. Beguia

Determine the weighted mean values of the following data:

Scholastic Very High Moderate Low Very

Problems in High Low ∑fx Wm Rank Interpretation
District C 5 4 3 2 1

Developing
Readers 39 62 111 50 38 914 3.05 2 Moderate
Severly
Wasted 6 13 41 180 60 625 2.08 1 Low
Nutritional
Status
Low
Performance 46 95 89 42 28 989 3.30 3 Moderate
Level
Poor
Attendance 122 71 68 29 10 1,166 3.89 4 High
Low Student-
Book Ratio 86 143 51 12 8 1,187 3.96 5 High
3.23 Moderate
Overall Mean

Fill up the rows and columns with the correct data or information.

1. What is the value of N? 300

2. What problem was found as the highest? Low- Student- Book Ratio
3. What problem was found as the lowest? Severely Wasted Nutritional Status
4. What is the most alarming problem? Severely Wasted Nutritional
 8

Why? Because this is the lowest scholastic problem in District C. Books are one
of the most important to the students. It is the materials that can help in
learning.
5. Write a general statement about the data obtained.
Generally, in District C "Moderate" conducted the five scholastic problems.

Part II. Measures of Dispersion or Variation

The spread of value around the central tendency is called dispersion or “variability”.
Measures of dispersion serve as index of spread of x-values away from the central value.
There are four common measures of dispersion and variation. These are range, variance,
standard deviation and coefficient of variation.

I. Objectives:
At the end of the lesson, the student is expected to:
A. Define the different measures of dispersion.
B. Use the rule in calculating measures of dispersion correctly.

II. Topics:
A. Range
B. Variance
C. Standard deviation
D. Coefficient of variation

III. Presentation:

A. Range. This is the difference from the highest and lowest value.

Example: 62 – highest value or score

-35 – lowest value or score

27 Range

B. Variance. This is the value obtained by squaring each discrepancy, sum

up the values then divide it by the number of cases (N-1).

∑(dx ) ² 120 120

Example: V = = = = 10.91
N−1 12−1 11
 9

C. Standard deviation. It is a value that describes the homogeneity

(closeness) or heterogeneity of a set of data being analyzed. It computed
by taking the square root of the variance.

Example : sd =
√ ∑ (dx )²
N−1 √
=
120
12−1 √
=
120
11
=√ 10.91 = 3.30

Interpretation of sd values: 3-below = homogenous

more than 3 = heterogeneous

D. Coefficient of Variation. This is obtained by dividing the value of

standard deviation (sd) by the mean and multiplied by 100. The value is
expressed in %.

sd 3.30
Example: CV = x 100 = = 0.4125 x 100 = 41.25%
M 8

Score

Practicum Paper
Measures of Dispersion and Variation 10

Name: Via Bianca R. Beguia

Examine very well the examples at the left column then do the activities in the right
column.

Left Column Right Column

Student Xi dx=(Xi-M) (dx)²=(Xi-M)² Student xi dx=(Xi-M) (dx)²=(Xi-M)²

A 8 -4 16 Jun 12 -5 25
B 9 -3 9 Mark 12 -5 25
C 9 -3 9 Ken 13 -4 16
D 10 -2 4 Adelaida 14 -3 9
E 10 -2 4 Yasmin 15 -2 4
 10

F 11 -1 1 Tato 16 -1 1
G 13 1 1 Ismael 16 -1 1
H 13 1 1 Hubert 16 -1 1
I 13 1 1 Aisha 19 2 4
J 15 3 9 Ashley 21 4 16
K 16 4 16 Herbert 24 7 49
L 17 5 25 Gil 26 9 81
N = 12 ∑Xi=144 ∑dx= 0 ∑(dx)²= 96 ∑=
N=12 204 ∑(dx)²= 232
M= 12 M=1
7

A. Measure of dispersion using the number A. Find the Range using

line system. Subtraction:
13
9 10 13 26
8 9 10 11 13 15 16 17 Highest score
_____.____.___.___.___.__.___.___.___.____.__
__ ____ 12
8 9 10 11 12 13 14 15 16 17 Lowest score
____________
. 4 X 5
Mean RANGE 14
. 9 .
Range

B. Show the Range using Number Line system. Look at the model above.

12 12 13 14 15 16 16 16 19 21 24 26

14

Range

B.Measures of dispersion using the B. Measures of dispersion using the

Variance Variance

V = ∑(dx)²/ N-1
 11

❑
96
= 12−1 V=
∑
❑
(dx)²
=
¿ N −1
= 8.73 =232
12-1
=21.09

C. Measure of dispersion using C.Measure of dispersion using Standard

Standard Deviation Deviation

s.d. ¿
√∑ (dx )²
N−1

¿
√ 96
12−1
=√21.09
=4.59
= 2.95
D. Measure of dispersion using Measure of dispersion using
Coefficient of Variation Coefficient of Variation

s.d . C.V.= s.d. x 100

C.V. = x 100
M M
=4.59 x100
2.95 12
= x 100
12
=459
295
= 12
12
= 38.25%
= 24.58%

Part III. Inferential Statistics

A. Measures of Correlation

This lesson and the succeeding ones tackle on inferential statistics. It is in this
type of statistics where inferences about whether relationships/association/correlation
exists in the samples used. Testing the null hypothesis is involved here. The last set of
lessons reflected in the outline of topics in the Course Outline given to you may already
be integrated here. This is not only to save time but to make the learning of the unit more
interesting and meaningful. (Electronic calculation will be demonstrated to you when we
meet).

I. Topics:
 12

Measures of Correlation
1. Spearman’s Coefficient of Correlation
2. Kendall’s tau
3. Pearson’s Coefficient of Correlation
4. Chi-Square
5. Kendall Coefficient of Concordance W
(There are other tools used to determine correlation/association between 2 or
more variables, but these are the most common ones).
II. Objectives:

In this lesson, the student is expected to:

1. Use measures of correlation efficiently and effectively.

2. State some simple analyses based from values of data obtained.
3. Explain the result of the test of the null hypothesis

III. Presentation:

A. Spearman’s Coefficient of Correlation

This measure is also known as Spearman’s rho. Rho is the Latin for “r”
which is the symbol used to show relationships. This is used when N is 10 or
more. There are two distinguishing characteristics of the Spearman’s Coefficient
of Correlation. These are: 1) it is dichotomous, meaning two sets or groups are
correlated; and 2) it uses the ranks of values previously obtained like in
frequencies, percent or weighted mean.

2
6(∑ D )
Formula: r = 1- where:
N ( N −1 )
2

r = is the value of correlation

N= is the number of items or cases correlated
∑D² = is the sum of the differences squared

1 and 6 are constants. 1- means that the value of r should not be more than 1.
Thus, the table below is a guide to interpret the coefficient of correlation.

Computed r – value Interpretation

Less than ± 0.20 Negligible correlation
± 0.20 to ± 0.39 Low correlation (definite but small relationships)
± 0.40 to ± 0.69 Moderate correlation (or substantial relationship)
± 0.70 to ± 0. 89 High Correlation (or marked correlation)
± 0.90 to ± 1.00 Very high (perfect if r=1.0) correlation(or very dependable
relationship)
 13

The coefficient shows the strength (or weakness) of the association between two
variables. Types of correlation could be no correlation, perfect positive correlation and
perfect negative correlation.

Example:
Diseases commonly diagnosed in the Nutrition Department of “E” Hospital

Males Females
Diseases Frequency Rank Frequency Rank
Hypertension 9 2 12 3
Renal failure 78 10 54 8
Tuberculosis 32 6 42 7
Anorexia 11 3 17 4
Colon cancer 53 8 68 9
Anemia 25 5 97 10
Cardio-vascular (heart) 64 9 34 5
Diabetes 5 1 40 6
Obesity 41 7 11 2
Marasmus 22 4 8 1

Procedure in calculating the value of Spearman’s r:

1. Take the ranks of both groups. However, the ranks of the first group should be
arranged from lowest to highest.
2. The ranks of the second group will be paired with the ranks of the first group.
3. Determine the differences of the ranks horizontally and write them in column
called “D”.
4. Square the differences and write them in the column called “D²”.
5. Get the total value of the D². (∑D²)
6. Calculate the value of r using the rule.
7. Look at the chart for the meaning of the value.

Illustration:

Males Females D D²
1 6 -5 25
2 3 -1 1
 14

3 4 -1 1
4 1 3 9
5 10 -5 25
6 7 -1 1
7 2 5 25
8 9 -1 1
9 5 14 16
10 8 2 4
∑D² = 108

After determining the value of ∑D², proceed to the calculation of r:

6 (∑ D 2 ) 6(108) 648
r=1– 2 =1– 2 = 1- 990 =1- 0.65 = 0.35
n(n −1) 10(10 −1)

low correlation
(look at the table)

To determine the significance of correlation, the t – value is used. The formula is: t –

√
value = r
n−2
1−r ²
where: r = is the computed r- value

n-2= is the degree of freedom

In this case the null hypothesis (Ho) is also tested:

Null Hypothesis (Ho): There is no significant correlation between the diseases

commonly diagnosed among the males and females in the Nutrition Department
of “E” Hospital.

Alternative Hypothesis (Ha): There is a significant correlation between the

diseases commonly diagnosed among the males and females in the Nutrition
Department of “E” Hospital.

Assuming that the respondents represent a sample, then, there is a need to

determine the significance of the r-value by using the equation shown below:

Substituting the formula, it would look like this:

t-value = 0.35
√ 10−2
1−0.35²
= 1.06 if you are using a scientific calculator
 15

= 0.35
√ 8
0.88
if you are using ordinary calculator

= 0.35 √ 9.09 or mobile phone.

= 0.35 (3.01)

= 1.06 Computed value or Calculated value

The critical or tabular t-value at 8 degree of freedom on a 0.05 (5%) level of

significance is 2.896. (See the table of critical t at the last part of this paper).

To make a decision about the null hypothesis (Ho), compare the computed
value against the tabular value (or critical value) like 0.05 or 0.01 level of
significance.

 Reject the Ho if the computed value is greater than the critical or

tabular value.
 Accept the Ho if the computed value is lesser than the critical or
tabular value.

In this case (the example), the computed t-value is lower than the tabular value.
Therefore, the decision is to Accept the Ho. The Ho is to be taken as conclusion.
Therefore, there is no significant correlation between the diseases commonly
diagnosed among the males and females in the Nutrition Department of “E”
Hospital.

B. Kendall’s tau (T)

The characteristic of this tool is just like the Spearman’s r. However, the value of N
makes them different. The Kendall’s tau (taken from the author Maurice Kendall) is used
when N is lesser than 10. The formula is:

C−D
T= where: C= concordant pairs (or higher numbers)
C+ D
D = discordant pairs (or lower numbers)
 16

Example:

Non-teaching
Assignments of Male Female
Classroom Teachers Frequency Rank Frequency Rank
Librarian 34 3 61 4
Scout Leader/Master 67 4 27 3
Nutrition Coordinator 3 1 5 1
Guidance Teacher 8 2 9 2

Procedure:
1. Arrange the ranks of the first variable from lowest to highest.
2. Pair the ranks of the second variable side by side with the first variable.
3. Count the number of concordant and discordant pairs.
4. Collect the values in the C and D columns separately.
5. Determine the value of T
6. Calculate the significance of correlation using t-value.

Illustration:

Male Female C D
1 1 3 0
2 2 2 0
3 4 0 1
4 3
Total 5 1

C−D 5−1 4
T=
C+ D
= 5+1
= 6
= 0.67 (moderate correlation)

To get the values in the C and D columns:

Since you have already arranged and paired the ranks of the 2 groups, this time
forget or do not focus on the first colum anymore (in this example is the Male
column). Just focus of the second column (Female).
a. Look at the first entry in the Female column (1). Ask yourself silently: How
many numbers below are higher than 1? The answer is 3 (because 2, 4, and 3
are lower than 1). Hence, 3 is written in the C column. Next, ask the reverse
question: How many numbers are lower than 1? The answer is 0. Hence, it is
written in the D column.
 17

b. Do the same with the rest.

c. Add the values in the C and D columns.
d. Calculate the value of relationship.

To test the null hypothesis, use the t-value as illutrated in Spearman’s rho.

C. Pearson’s Coefficient of Correlation (r xy ¿

This is also known as Pearson’s r. It is a kind of correlation where two (2) values
from one respondent are correlated. It takes into account each and every score and
produces a coefficient between 0.00 and + 1.00. There are two methods of determining
the Pearson’s r: the deviation method and the difference method. (You have learned this
already when we discussed reliability and validity of instruments).

For facility, let us take the deviation method. The formula used is:

r ∑ dxdy
xy=
( N−1) ( sdx) ( sdy )

where: r xy is the value of the Pearson’s coefficient

∑dxdy is the product of the deviation between x and y variables
sdx is the standard deviation of variable x
sdy is the standard deviation of variable y
N is the total number of subjects/cases

Example: 10 students whose major is English had the following scores in Prelim and Midterm
in Literature. Find the degree of correlation between their scores in prelim (x) with their scores in
midterm (y).

Students deviation Deviation

of x of y
N x y (dx)²=(Xi- (dy)²=(Yi-M)² Dxdy
dx=(Xi-M) dy=(Yi-M) M)²
 18

A 12 8 6 1 36 1 6
B 10 12 4 5 16 25 20
C 8 6 2 -1 4 1 -2
D 7 4 1 -3 1 9 -3
E 6 9 0 2 0 4 0
F 5 10 -1 3 1 9 -3
G 5 7 -1 0 1 0 0
H 4 6 -2 -1 4 1 2
I 2 5 -4 -2 16 4 8
J 1 3 -5 -4 25 16 20

∑dx)² ∑(dy)² ∑dxdy

10 60 70 104 70 48
M=Mean

Mx=6 My=7

1. sdx =
√ ∑ (dx )²
N−1
=
√ 104
10−1
= 3.40

2. sdy =
√ ∑ (dy )²
N−1
=
√ 70 = 2.79
10−1

r
3. xy=
∑ dxdy
( N−1) ( sdx) ( sdy )

r
4. xy=
48
( 9 )( 3.40 ) ( 2.79 ) = 0.56 (moderate correlation)

To test the Ho, use the formula for t-value.

Ho. The correlation between the scores of the Clinical Nutrition students in
Quiz 1 and Quiz 2 in Biostatistics is not significant.

Ha. The correlation between the scores of the Clinical Nutrition students in
Quiz 1 and Quiz 2 in Biostatistics is significant.

t-value = r
√ n−2
1−r ²
= 0.56
√ 10−2
1−0.56²
= 0.56
√ 8
0.69
=0.56√ 11.59
 19

= 0.56(3.40) = 1.90 computed value or calculated value

computed t-value = 1.90

tabular t-value = at 8 df based on 0.05 = 2.896

Analysis Decision for Ho

Since the computed t-value is lesser than the tabular t-value on a 0.05 level of
significance, therefore, the Ho is accepted. With this, the correlation between the scores
of the Clinical Nutrition students in Quiz 1 and Quiz 2 in Biostatistics is not significant.
Conclusion

The evidence does not warrant that the relationship between variables does not exist. The
result showed that 95 in 100 trials, the variables are not related or are independent and
only 5 in 100 trials, the relationship exists among variable (Subong).

Score

Practicum Paper Pearson’s Coefficient of Correlation

16
Name: Via Bianca R. Beguia

Instruction: Calculate the degree of correlation using the deviation method between the
marks of the students in English and Mathematics Test also the null hypothesis (Ho)
shown below the data.

Given:

Student English Math

N x y dx dy dx² dy² dxdy
A 15 13 6 4 36 16 20
B 13 13 4 4 16 16 16
C 12 12 3 3 9 9 9
D 12 11 3 2 9 4 6
E 10 10 1 1 1 1 1
F 9 10 0 1 0 1 0
G 8 9 -1 0 1 0 0
H 7 8 -2 -1 4 1 2
I 7 7 -2 -2 4 4 4
J 6 6 -3 -3 9 9 9
 20

K 5 5 -4 -4 16 16 16
L 4 4 -5 -5 25 25 25
N= 12 ∑=108 ∑=108 ∑(dx)² ∑dy)² ∑dxdy
M=9 M=9

sdx =

sdy =

r xy =

Meaning:_____________________________________

3Calculate the t-value

t-value =

Null Hypothesis (Ho): The correlation coefficient between the marks of the students in
the English and Math is not statistically significant.

Degree of freedom (df) = 10 0.05= 2.228

Analysis:
____________________________________________________________________.
________________________________________________________________
_____.
Decision for Ho:
_____________________________________________________________.
Conclusion:
_________________________________________________________________.
_____________________________________________________________
____.
_____________________________________________________________
____.
_____________________________________________________________
____.
_____________________________________________________________
____.
 21

D. Chi-Square Test (X²)

This test is used to measure the association between two nominal variables. It is
only applicable for relationships of variables with nominal data. There are two types of
Chi-Square test. These are:

1. Chi-Square test in a one-sample case; and

2. Chi-Square in two-sample cases

( fo−fe)²
The formula is: X²= ∑
fe

where:X² is the Chi-Square value

fo is the actual observed frequencies
fe is the expected frequencies
Example A: one sample –case:

A class with 54 students in TVL focused on Bread and Pastry was surveyed as to
their willingness to observe different laboratory procedures done in Everlasting Bakery in
the city. The result showed that:

Willing =32 Undecided = 9 Not Willing = 13

The calculation would look this way:

Measure Willing Undecided Not Total

Willing
Fo 32 9 13 54
fe 18 18 18 54
fo-fe 14 -9 -5
(fo-fe)² 196 81 25
(fo-fe)²/fe 10.89 4.5 1.39

( fo−fe) ²
∑ 16.78
fe Computed X²=
Degree of freedom (df) = k-1; 3-1 =2

df = 2 at 0.05 level of significance = 5.99 (Look at the Chi-Square Table)

Ho. The responses of the students on their willingness to observe different laboratory
procedures in a bakery do not vary.
 22

Ha. The responses of the students on their willingness to observe laboratory procedures
in a bakery vary.

Since the computed value of the Chi-Square of 16.78 is greater than the tabular
value of 5.99 on a 0.05 level of significance, then the Ho that the students’ responses on
their willingness to observe different laboratory procedures in Everlasting Bakery was
rejected. The alternative hypothesis (Ha) that the students’ responses differ was accepted.
This shows that the students’ willingness to observe the laboratory procedures in the
bakery varies. In other words, the students’ responses were not common, so it is not a
wise decision to go on with the activity.

Example B. Two-Sample Cases

There were 50 Student Teachers who were on their off-campus or internship year.
Determine whether or not their attitudes in their training are associated with gender.

Data:

Gender
Attitudes Male Female Total
Very Good 9 5 14
Good 12 7 19
Fair 8 9 17
Total 29 21 50

Cell fo fe fo-fe ( fo-fe )² (fo-fe)²/fe

A 9 8.12 0.88 0.7744 0.0954
B 5 5.88 -0.88 0.7744 0.1317
C 12 11.02 0.98 0.9604 0.0872
D 7 7.98 -0.98 0.9604 0.1203
E 8 9.86 -1.86 3.4596 0.3509
F 9 7.14 1.86 3.4596 0.4845
X² =1.2700
computed value
To calculate fe: fe= VT x HT ÷ GT

df = (k-1)(r-1) =(2-1)(3-1) = (1)(2) = 2

tabular X² value: 0.05=5.991 Computed X² = 1.2700

The interpretation is:

 23

Anal
ysis: Computed X² value is lesser than the tabular X² value.

Decision for Ho: Accepted

Conclusion: Attitudes of the Student Teachers in their off-campus or internship year

training are not associated with gender.

In other words, attitudes of the students have nothing to do with their gender. Each is
independent.

E. Fischer’s Exact Test

or Yate’s Correction Formula

This is a Chi-Square test when the degree of freedom is equal to 1, (df=1) and the
contingency table has a cell frequency of less than 5. The formula is:

∑(¿ fo−fe /−0.5)²

X² =
fe
Example:
Ho: There is no association between gender and performance in dance sports among
the contestants during Palarong Bicol.

Gender
Male Female Total
Performance
Very Good 9 4 13
Good 3 7 10
Total 12 11 23

Cell fo fe /fo-fe/-.50 〔/fo-fe/-.50〕² 〔/fo-fe/-.50〕²/fe

A 9 6.78 1.72 2.9584 0.4363
B 4 6.22 -2.72 7.3984 1.1895
C 3 5.22 -2.72 7.3984 1.4173
D 7 4.78 1.72 2.9584 0.6189
X² = 3.662
df=1; level of significance=0.05; critical value=3.84;

Analysis : The computed value is lesser than the critical value.

Decision: Ho is accepted.
 24

Conclusion: There is no association between gender and performance in dance

sports among the contestants during the Palarong Bicol.

(You will learn the electronic computation when we meet).

F. Kendall Coefficient of Concordance(W)

The symbol used for this tool is W, that is why it is called Kendall’s W. It is a non-
parametric test used to determine the relationship among 3 or more variables using ranks.
Hence, it is ordinal. When the degree of correlation is obtained, the test of significance
uses the Chi-Square test. The formula is:

12∑ Tj ² 3(N +1)

W= 2
m² N (N −1)
- N−1

Where: W = is the value of concordance

m = is the number of groups of preceptors or respondents
N = is the number of items correlated
∑Tj² = is the sum of the ranks squared

Just like in the Spearman’s r, W uses ranks of frequencies, percentages, and

weighted mean previously obtained.

Example: Perceptions of adult males on the causes of tuberculosis.

Age Group
Young Adults Middle Adults Late Adults
Perceptions f Rank f Rank f Rank
A. Being a heavy smoker. 11 1 15 2 25 3
B. Poor nutrition. 22 2.5 38 4 42 4
C. Untreated respiratory tract 35 4 2 1 7 1
infection
D. Exposure to a person with 22 2.5 25 3 16 2
tuberculosis.

To make the calculation easier, make a chart like this:

N
 25

Perceptions A B C D
Groups
Young Adults 1 2.5 4 2.5
Middle Adults 2 4 1 3
Late Adults 3 4 1 2
∑Tj 6 10.5 6 7.5
∑Tj² 36 110.25 36 56.25 238.5

Null Hypothesis (Ho): The concordance of the ranks of the perceptions of the
male adults on the causes of tuberculosis is not significant.

To Calculate:

12∑ Tj ² 3( N +1) 12(238.5) 3(5) 2862 15

W= 2 – = 2 - = - = 5.3 – 5 =
m² N ( N −1) N−1 3²(4 )(4 −1) 3 540 3
0.3

Low correlation
To test the Ho. Use the formula: X² = m(N-1)W

So, X² = 3(4-1)0.3 = 2.7 Computed value

Now, consider these: df=3; critical value at 0.05=7.82

Analysis: The computed value is lower than the critical value.

Decision for Ho. Accepted.
Conclusion: The concordance of the ranks of the perceptions of the male adults
on the causes of tuberculosis is not significant.
 26

Score
Practicum Paper Fischer’s Exact Test

Name: _______________________________________________________ 10

Part I. Calculate the value of X² using Fischer’s Exact Test then give the analysis and
interpretation of the result of computation.

Ho. There is no relationship between socio-economic level and general health status of
males beyond 40 years old.

SES
Health Status High Low Total
Good 15 4 19
Poor 5 3 8
Total 20 7 27

__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
__________________________________________
X²=____________
_

df=1; level of significance = 0.05; critical value= 3.841;

computed value= __________

Analysis:_________________________________________________________.
 27

Decision: _______________________________________________________.

Conclusion:___________________________________________________.
____________________________________________________.

Part II. Make a 2 x 2 contingency table; put the values in the corresponding cell then
determine the value of the X² and make an analysis and interpretation of results obtained.

Twenty-eight samples were taken in a study in Division Z to

determine the relationship between performance rating and
educational attainment of elementary teachers who were non-MA and
MA degree holders. Data showed that among the MA degree holders,
12 received Outstanding while 5 were Very Satisfactory. Among the
non-MA degree holders, 3 were Outstanding while 8 were Very
Satisfactory.

Ho.
__________________________________________________________________
__________________________________________________________________
____________.

__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
____________________________________

Analysis: _______________________________________________________________.

Decision for Ho: ________________________________________________________.

 28

Conclusion: ____________________________________________________________.

_______________________________________________________________.
Score
Practicum Paper
Kendall Coefficient of Concordance (W) 25
Name: ___________________________________________________

Part I. Complete the statistical processes needed to determine the value of W then test the
null hypothesis (Ho).

Determine the weighted mean values of the following using the weights and
interpretations as indicated below:

1= Low 2 = Moderate 3 = High

A. Children (N=50)

Beliefs on Food Habits

1 2 3 ∑fx Wm Rank Interpretation
1.Drinking coffee doesn’t
make you sleep. 9 11 30
2.Eating plenty of rice
makes you fat. 27 14 9
3.Frequent drinking of
carbonated drinks is one 8 37 5
cause of hyperacidity.
4.If you want to lose
weight, do not eat anything 42
after 6:00 pm. 5 3
5.Fair skin is the effect of
eating fresh fruits and 22 18 10
vegetables.
Overall Weighted Mean=
 29

B. Teenagers (N=40)

Beliefs on Food
Habits 1 2 3 ∑fx Wm Rank Interpretation
1.Drinking coffee
doesn’t make you 1 10 29
sleep.

2.Eating plenty of
rice makes you fat. 12 12 16
3.Frequent drinking
of carbonated drinks
is one cause of 21 11 8
hyperacidity.
4.If you want to lose
weight, do not eat 35
anything after 6:00 2 3
pm.
5.Fair skin is the
effect of eating fresh 7 26 7
fruits and vegetables.
Overall Weighted Mean=
C. Young Adults (N=45)

Beliefs on Food Habits 1 2 3 ∑fx Wm Rank Interpretation

1.Drinking coffee doesn’t
make you sleep. 31 10 4
2.Eating plenty of rice
makes you fat. 15 18 12
3.Frequent drinking of
carbonated drinks is one
cause of hyperacidity. 7 11 27
4.If you want to lose
weight, do not eat anything 2 29 14
after 6:00 pm.
5.Fair skin is the effect of
eating fresh fruits and 19 18 8
vegetables.
Overall Weighted Mean =
Part II. Answer the following questions:

1. What belief is Rank 1 among the:

a. Children? _______________________________________________
b. Teenagers? _____________________________________________
c. Young Adults? __________________________________________
2. What belief is the last rank among the:
a. Children? _______________________________________________
 30

b. Teenagers? _____________________________________________
c. Young Adults?__________________________________________
3. Which among the groups has the highest of belief based on the overall weighted
mean value? _________________________________.
4. Which among the groups has the lowest level of belief based on the overall
weighted mean value? _____________________________

Part III. Put the data in a matrix and determine the coefficient of correlation using
Kendall Coefficient of Concordance W.

__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
______________________________________________________.

Part IV. Test the Ho using X².

Ho. The correlation of the ranks given by the three groups of respondents on their belief on food
habits is not significant.

Analysis: ___________________________________________________________________________.

Decision for Ho: ____________________________________________________________________.

Conclusion: ________________________________________________________________.

Score
Practicum Paper Chi-Square

Name ___________________________________________________
5
Calculate the following and supply the necessary information.

There were 120 samples pupils diagnosed with reading difficulties. Determine whether
this problem is associated with their socio-economic status.
 31

Ho. The stage of tuberculosis among the 120 samples is not associated with their
socio-economic status.

SES
Reading High Average Low Total
Proficiency
Approaching 9 17 32 58
Beginning 7 40 15 62
Total 16 57 47 120
df= 2 Critical value: 0.05 = 5.991

Cell Fo Fe fo – fe (fo-fe)² (fo-fe)²/fe

X² =___________

Analysis: ______________________________________________________________.

Decision for Ho: _______________________________________________________.

Conclusion: ___________________________________________________________.

___________________________________________________________.

___________________________________________________________.

Score
Practicum Paper
Kendall’s Tau
Name: _________________________________________________
10

Instruction: Find the degree of correlation between weight gain of male and female Grade
4 pupils with severely wasted nutritional status during the first 6 months of nutrition
therapy.

Weight gain Males Females

(in kilos) Frequency Rank frequency Rank
0.50 7 9
 32

0.75 5 11
1.00 9 12
1.25 13 7
1.50 12 8
1.75 11 10

Males Females C D Test the significance of correlation using

t-value

Total

Ho. There is no significant correlation between weight gain of male and female Grade 4
pupils with severely wasted nutritional status during the first 6 months of nutrition
therapy.

df = 4 level of significance at 0.05= 2.776

Analysis: ______________________________________________________________.

Decision for Ho: ______________________________________________________.

Conclusion: __________________________________________________________.

___________________________________________________________.
Score
Practicum Paper Spearman’s rho

10

Name: ______________________________________________________

Instruction: Calculate the degree of correlation of the given situation below using
Spearman’s rho then test the significance of correlation.

Common Causes of Sick

Leaves among Teachers Elementary Teachers High School Teachers
Frequency Rank frequency Rank

Heart diseases 54 85

Tuberculosis 16 51
 33

Renal failure 27 18

Anorexia 32 59

Anemia 79 35

Scurvy 52 27

Diabetes 48 63

Colon cancer 94 41

Hepa-A 12 9

Hepa-B 80 72

Elementary Teachers High School Teachers D D²

∑D²=

Calculate the value of Spearman’s rho here.

Meaning: ________________________

Test of Significance using t-value

t-value =
 34

df= 8 level of significance at 0.05 = 2.896

Write the null hypothesis

(Ho).___________________________________________________________________
________________________________________________________________________
________________________________________________________________________
____________________________________________________________________.

Analysis: ______________________________________________________________.

Decision for Ho: _______________________________________________________.

Conclusion: ___________________________________________________________.

___________________________________________________________.

___________________________________________________________.