CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

To present a well-balanced picture of the students’ performance in the assessment

procedures, teachers present, describe, analyze, and interpret test scores. From these
results, they draw conclusions or make predictions or forecasts about the group being
studied, and eventually plan for enrichment and intervention activities to ensure that
students attain the pre-set learning outcomes.

To perform the abovementioned tasks, teachers need to be equipped with the

necessary background and expertise in educational statistics and measurement. According
to Smart (2019), statistics in education with special considerations to measurement and
evaluation are essential parts of the teaching and learning process. In this process, scores
are analyzed, interpreted, and utilized to make better-informed instructional decisions.
Through statistics, the teachers can see and study test scores more objectively. Scores can
provide a more accurate representation of the acquired competency and can serve as a
basis in the appraisal of the student progress, in determining what works well, and what
areas or aspects in the instructional process need to be improved.

Learning Outcomes

At the end of the chapter, the students are able to:

1. construct frequency distribution based on a given set of scores;

2. apply statistics to calculate and compare individual and group scores;
3. distinguish the measures of central tendency from the measures of variation; and
4. describe and interpret assessment results;

Performing Statistical Treatment and Analysis to Describe Performance

Statistics is a highly interdisciplinary field that applies a systematic method of

collecting, analyzing, and interpreting numerical data. These data convey meaning and/or
information where conclusions, decisions, and succeeding actions can be based upon. In
this particular context, the students’ scores in a test or the assessment results are the data
the teachers can use to obtain information about what students have learned, how well
they have learned, and what more could be done to ensure learning according to the
desired standards or outcomes.

Frequency Distribution illustrates in a table or a graph the distribution of

observations within a particular test. It makes the presentation of test scores or data more
interpretive or simple to understand.

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 94
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Example:

Scores of 30 Students in a 10-item Math Test

Scores Tally
8 - 10 IIII-II
4 - 7 IIII-IIII
0 - 3 IIII-IIII-III

Components of the frequency distribution Class mark is the

midpoint of the lower
Class Limit is defined by the lower Class boundaries separate one and upper class limits.
class in a grouped frequency ¿+UL
(LL) and upper limits (UL). Formula: XM =
distribution from another. 2

Lower class limit is Upper class limit is

the smallest number LL UL the highest number Lcb - Ucb XM
in each group. 41 - 46 in each group. 40.5 - 46.5 43.5
36 - 40 35.5 – 40.5 38
31 - 35 30.5 – 35.5 33
26 - 30 25.5 – 30.5 28
5
21 - 25 20.5 – 25.5 23

Class interval width is the difference

between the lower endpoint of an
interval and the lower endpoint of
the next interval (Statistics Canada,
2017).

Steps in constructing a grouped frequency distribution

1. Sort the data in ascending order.

46 46 45 45 43 43 43 42 41 40 38 38 36 34
34 33 27 26 26 25 24 24 24 24 23 23 22 21

2. Determine the range of scores.

Range (R) = Highest Score (H) – Lowest Score (L)
= 46 – 21 = 25
3. Decide on the number of intervals in the frequency distribution
Number of intervals selected = 5 (this can vary)

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 95
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

4. Determine the class interval width by dividing the range by the number of
intervals selected.

25
Classinterval width= =5
5
41 - 46
36 - 40
31 - 35
26 - 30
5
21 - 25

Scores Tally frequency

41-46 IIII-IIII 9
36-40 IIII 4
31-35 III 3
26-30 III 3
21-25 IIII-IIII 9
Total 28
5. Tally and count the observations under each category.
6.

Graphical Representation of Scores in a Frequency Distribution

10
9
8
7
6
5 Series1
4
3
2
1
0
41-46 36-40 31-35 26-30 21-25

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 96
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

The data show that 9 scores belong to the highest score category ( 41-46) and
similarly 9 scores belong to the lowest score category (21-25). The remaining 10
scores can be categorized in the middle score categories (26-30, 31-35 and 36-40).

Measures of Central Tendency are some of the most basic statistical

functions we can use to describe a summary of data set using a single value that
represents the center or the middle point of the distribution. These measures
present a summary statistics the teachers can use to conveniently describe the
performance of the individual or the entire group. The measures of central tendency
include the mean, median, mode, and quantile.

The mean or the average is the commonly used statistic to measure the
center of a data set. It is determined by adding all values in the data set divided by
the number of values in the data set.
Properties of Mean
o It is simple to calculate.
o It is the most reliable among all measures of central tendency.
o It is determined by adding all values of the data divided by the sum of
all values.
o It is easily affected by the magnitude of (extremely high or low)
scores.
o The sum of each score’s deviation from the mean is zero.
o It is used to compute other statistics such as standard deviation,
variance, t-ratio, critical ratio, coefficient of variation, and skewness.
o It can be applied to interval and continuous data but cannot be used
for nominal or categorical data.

Mean for Ungrouped Data

This application utilizes raw data that are not sorted or grouped into
categories. In example 1, we compute for the mean of ungrouped data by
determining the sum of grades divided by the number of subjects taken.
Formula:
∑x
¿
n
Where:

= mean (grade point average)

x = grade earned for every subject
∑x = total grade earned in all subjects (x1 + x2 + x3 + ...)
n = number of subjects

Example 1: Kitz took 8 subjects in grade 10 class. Determine his grade point
average based on the following grades he earned in different subjects.

Learning Areas Final ∑x

Grades Solution: ¿
n
701
This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
¿
permission from the authors. 8 Page 97
¿ 87.63
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Filipino 90
English 84
Mathematics 87
Science 86
Araling Panlipunan 89
Edukasyong Pagpapakatao 88
Edukasyong Pantahanan at 90
Pangkabuhayan
MAPEH 87
∑x = 701

To interpret and analyze the computed mean value of 87.63, we use the
guideline for grading system issued by the Department of Education as presented in
Table 11 below:

Table 11. Descriptors, Grading Scale, and Remarks

DESCRIPTORS GRADING REMARKS Interpretation:

SCALE
Outstanding 90-100
Passed Kitz had a very satisfactory
Very Satisfactory 85-89
Passed academic performance as
Fairly Satisfactory 80-84
Passed indicated by the grade point
Satisfactory 75-79
Passed average (¿ 87.63 ¿ he earned
Did Not Meet Below 75
Failed in the 8 subjects in grade 10.
Expectations
Source: Professional Learning Online
Community of Teachers and for Teachers (2020)

Example 2: The following are the scores of the 12 students in a 50-item unit
test in Science IV. Determine the mean or the average of the scores.

Specific to this context, we calculate for the mean by adding all scores divided
by the number of test-takers.

∑x
Formula: ¿ n
Where:

= mean
x = score
∑x = Sum of scores (x1 + x2 + x3 + ...)
n = number of test takers

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 98
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Students Score Interpretation

s (x)

1 34 Above the mean

2 43 Above the mean
3 12 Below the mean
4 26 Below the mean
5 17 Below the mean
6 19 Below the mean
7 24 Below the mean
8 25 Below the mean
9 24 Below the mean
10 42 Above the mean
11 16 Below the mean
12 4 Below the mean

∑x = 286 Interpretation:
The average performance of 12
students who took the 50-item test is 23.83.
The data show that 3 out of 12 students
scored above the mean while 9 out of 12
students scored below the mean. Three
students performed well in the test while the
remaining 9 students showed poor
performance in the test.

Mean for Grouped Data

Grouped data are data that are arranged in a frequency distribution.

∑ f xm
Formula: =
n

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 99
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Where:
f = frequency in each class category
Xm = midpoint of each class or category
∑fXm = summation of the frequency (f) multiplied by the class midpoint (Xm).

Example: Find the mean score of 45 students in a reading test of 45 items based
on the tabulated scores below.

x f xm fxm So
37-45 3 41 123
30-36 9 33 297
23-29 12 26 312
16-22 13 19 247
9-15 8 12 96
45 1075

Interpretation:

The mean performance of 45 students in reading test is 23.89 rounded off to

24. Students who achieved scores lower than 24 did not perform well in the test.

Median is a point in a scale that divides it into two equal parts. A scale is a
succession of numbers, classes, degrees, gradations, or categories with a fixed
interval (Calderon, 2006).

Properties of Median

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 100
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

o It is the exact midpoint of the score distribution.

o It is a stable measure of central tendency that is not affected by
extreme values.
Scores o It is used to determine whether the data values fall into the upper half
49 or lower half of the distribution.
49 o It is robust to skewness and outliers.
46
44 Median of Ungrouped Data
42
41 To find the median of ungrouped data, arrange scores in
38 descending order and locate for the middlemost score. If there are
37 two middlemost scores, the median is the average of the two scores.
36
36
32
29
27
24
21
N=15

Example:

Scores
50
48
47
36
The median ( ) is 37, the The 35
middlemost scores are 29
middlemost score. 32
and 30. = 30+29 = 59 = 29.5
31 2 2
Interpretation:
30
The median score is 37. This
means that 50% of the test 29
Interpretation:
takers (7 students) scored 29
The median score is 29.5.
higher than 37 and 50% (7 28
This means that 50% of the
students) scored lower than 37. 24 in the distribution are
scores
higher
26 than 29.5 and 50% are
lower
23than 29.5.
21
19
Median of Grouped Data N=16

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 101
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

( )
n
−cfp
Formula: = 2
LB + ci
fm

Where:
= median value
n
MC = median class category containing the
2
LB = Lower boundary of median class
cfp = cumulative frequency before the median class (scores are arranged
from the lowest to the highest)
fm = frequency of the median class
ci = size of the class interval

x f ncf<
Solution: 2
23-29
30-36 8
=
LB + 2
2
−cfp
fm
10
( )
ci

( )
37-43 7 17 40
cfp −17
44-50 11 fm = 43.5+ 282 7
51-57 6 34 11
58-64
65-70
4
2
= 43.5+ 38
20−17
40 11 ( )
7

40
= 43.5+
3
11
7 ( )
= 43.5+
21
11 ( )
40 = 43.5+¿ 1.91
=20
2
0.5 = 43.5

Interpretation:
The median is 45.41. This indicates the 50%
or 20 students scored lower than 45.41

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 102
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

The mode is the most frequently observed value in a set of data. It is the
score that occurred the most number of times, therefore with the highest frequency.
If there is only one mode in a distribution of scores, it is unimodal. If it consists of
two modes, it is called bimodal. If there are three or more modes in a distribution, it
is either called trimodal or multimodal.

Properties of Mode
o It is the most probable and/or typical value.
o The value of mode is based on the predominant frequency and not by
value in the distribution.
o It is unstable and influenced by grouping procedures.

Mode for Ungrouped Data

To easily find the mode for ungrouped data, sort the scores (from least to
greatest or greatest to least) and count the number times each score appears.
The score that appears the most is the mode.

Example: Teacher Marsh gave a 30-item test in Filipino. The following were the scores of
the students in 3 sections of the first-year level. Find the mode.

Scores in Filipino Test

SECTION A SECTION B SECTION C
30 29 30
29 27 28
28 27 25
Interpretation
The
28 25 23
number of ti
28 23 23
distribution i
27 22 23
one mode. T
26 22 20
22 since both
23 22 19
bimodal dist
23 20 19
three scores
20 18 19
13, 19, and 2
19 18 13
multimodal d
18 18 13
14 14 13

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 103
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Mode for Grouped Data

Formula: ( 1 d
x̂ = LB + d + d ci
1 2
)
Where:
x̂ = mode
LB = lower boundary of the modal class
MC = modal class (category containing the highest frequency)
d1 = difference between the frequency of modal class and the frequency
above it (scores are arranged from lowest to highest)
d2 = difference between the frequency of modal class and the frequency
below it (scores are arranged from lowest to highest)
ci = size of the class interval

Example: The table below shows the scores obtained by 48 students in the
periodic test in Technology and Livelihood Education. Find the mode.
x f Solution:
20-24 1

Modal
25-29
30-34
2
12
Given:
1
(1

2
d
x̂ = LB + d + d ci )
Class
35-39
40-44
5
8
Modal Class = 30-34
LB = 29.5 (30-0.5= 29.5) ( )
10
x̂ = 29.5+ 10+ 7 5
d1 = 10 (12-2=10)
x̂ = 29.5+ ( 17 ) 5
45-49 6 10
d2 = 7 (12-5=7)
50-54 5 ci = 5
x̂ = 29.5+ ( 17 ) 5
55-59 5 10
60-64 3
x̂ = 29.5+ ( 17 )
65-70 1 50
48
x̂ = 29.5+2.94

Interpretation:
The mode of the score distribution of 48 students is 32.44 . This means
that the score of 32.44 occurred the most number of times.

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 104
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Quantiles are cut points that divide a dataset into equal groups. It
determines how many scores in a distribution are above or below a certain limit.
It can be a quartile, decile, or percentile as illustrated below.

divides the scores in a divides the scores in a divides the scores in a

distribution into distribution into distribution into
4 equal parts 10 equal parts equal 100 parts

Figure 13. Quartile, Decile, and Percentile

Properties of Quartiles
o There are three quartiles (Q1, Q2, and Q3). Q1 denotes that
about 25% of the numbers in the data set lie below the
number in Q1; about 50% of the numbers below the number in
Q2; and about 75% of the numbers below the number in Q3.
o The second quartile (Q2) is the median.
o Quartiles divide the number of data points into quarters or
four equal parts.
o They are used to compute for quartile deviation.
o They are used to show which scores are below (below Q 2) or
above the median (above Q2).

Properties of Deciles

o There are 9 deciles (D1, D2, D3, D4, D5, D6, D7, D8, and D9).
o D1 indicates the about 10% of the scores in a distribution are
lower than the score in D1; about 20% of the scores lower than
the score in D2; about 30% of the scores lower than the score
in D3; ... and about 90% of the scores lower than the score in
D 9.
o Deciles divide the number of data points into 10 equal parts.

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 105
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

o The 5th decile is the median.

o They are used to show which scores are below (below D5) or
above the median (above D5).

Properties of Percentiles

o There are 99 percentiles (P1, P2, P3,...P97, P98, and P99).

o P1 indicates the about 1% of the scores in a distribution are
lower than the score in P1; about 2% of the scores lower than
the score in P2; about 3% of the scores lower than the score
in P3; ... and about 99% of the scores lower than the score in
P99.
o Percentiles divide the number of data points into 100 equal
parts.
o The 50th percentile is the median.
o They are used to show which scores are below (below P50) or
above the median (above P50).
o They are useful when there is a large number of scores.

Formulas: Quantiles for Ungrouped Data

Quartile Decile Percentile
k
( ( ))
Qk = n+ 1−
4
k
4
Dk =
k
10 (
n+ 1−
k
10 ( ))
Pk =
k
100
n+ 1−
k
100 ( ( ))
Where: Where: Where:
Qk = is the indicated Dk = is the indicated Pk = is the indicated quartile
quartile quartile k = 1,2, 3, ... 97, 98, 99
k = 1, 2, 3 k = 1, 2, 3,...7, 8, 9 n = number of cases
n = number of cases n = number of cases

Example: Solve for the Q2, D6, and P73 using the given data below:
Q2 = 13th score P73 = 19th score

14, 15, 16, 17, 17, 19, 21, 21, 24, 26, 27, 31, 36, 36, 41, 43, 44, 48, 51, 52, 53, 53, 54, 57, 59

D6 = 15th score

Solution:

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 106
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Q 2=
( 2
4 ( ))
25+ 1−
2
4
D 6=
( 6
10 (
25+ 1−
6
10 )) P73=
( 73
100
25+ 1− (
73
100 ))
Q 2=
( 2
4 ( ))
25+ 1−
2
4
D 6=
( 6
10 (
25+ 1−
6
10 )) P73=
( 73
100
25+ 1− (
73
100 ))
Q 2= ( 504 + 24 ) D 6= ( 150 + )
4
10 10
P73= ( 1825
100 100 )
+
27

52 154 1852
Q 2= D 6= P73=
4 10 100

Q2=12.5 or the D6=15.4 or the P73=18.52 or the

13th score 15th score 19th score
Interpretation:
Q2 is the13th score in the D6 is the 15th score in the P73 is the 19th score in the
distribution (36). This distribution (41). Therefore distribution (51), hence, 73%
means that 50% of the 60% of the scores are lower of the scores are less than 51.
scores are below 36. than 41.

Formula: Quantiles for Grouped Data

Quartile Decile Percentile

( ) ( ) ( )
kn kn kn
−cfp −cfp −cfp
4 10 100
Qk =L B + ci D k = LB + ci Pk =L B + ci
fq fd fp
Where: Where: Where:
Qk = the indicated quartile Dk = the indicated decile Pk = the indicated percentile
k = 1, 2, and 3 k = 1, 2, 3, 4, 5, 6, 7, 8, and k = 1, 2, 3, ...97, 99, and 100
QC = class category 9 PC = class category
n DC = class category n
containing for Q1, n containing for P1,
4 containing for D1, 100
2n 10 2n
4 2n 100
3n 10 3n
for Q2, for Q3 3n for P2, for P3,... ,
4 for D2, for D3,... , 100
LB = Lower boundary of the 10 99 n
quartile class 9n for P99
for D9 100
cfp = cumulative frequency 10 LB = Lower boundary of the
before the quartile class

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 107
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

fq = frequency of the LB = Lower boundary of the percentile class

quartile class decile class cfp = cumulative frequency
ci = size of the class interval cfp = cumulative frequency before the percentile
before the decile class class
fd = frequency of the fd = frequency of the
decile class percentiile class
ci = size of the class interval ci = size of the class interval

Note: Scores are arranged in ascending order

Example: The Table below shows the scores of 60 students in a long test in Social Studies.
Find the values of Q2, D3, and P56.
x f cf<
25-29 3 3
30-34 4 7
Quartile (Q 2)
35-39 Decile (D
7 3) Percentile
14 (P56)
Given: 40-44 Given: 7 Given: 21
Qk = Q2 45-49 Dk = D3 9 Pk = P56 30
k = 2 50-54 k = 3 12 k = 56 42
2(60) 120
55-59 13 = 180 = 18
3(60) 56(60)
55 3360
Q2C = 45-49 = = 30 D3C = 40-44 P56C = 50-54 = =
4 60-644 101 10 100
56 100
LB = 44.5 45-0.5=44.5 LB = 39.5 40-0.5 33.6
65-69 2 = 39.5 LB = 49.5
58
50-0.5 = 49.5
cfp = 21 cfp = 14
70-74 2 cfp = 30 60
fq = 9 fd = 7
ci = 5 ci = 5 n = 60 fp = 12

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 108
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

ci = 5

( ) ( ) ( )
2 ( 60 ) 3 (60) 56 ( 60 )
−21 −14 −30
4 10 100
Q2=44.5+ 5 D 3=39.5+ 5 P56=49.5+ 5
9 7 12

( ) ( )
180 3360
−14 −30

( )
120 10 100
−21 D3=39.5+ 5 P56=49.5+ 5
4 7 12
Q2=44.5+ 5
9

Q 2=44.5+
30−21
5( )
D 3=39.5+ ( 18−14
7 )
5 P56=49.5+ ( 33.60−30
12 )5
9

Q2=44.5+
9
5 ()
D3=39.5+ ( 47 ) 5 P56=49.5+ ( 3.60
12 )
5
9
D3=39.5+ 2.86 P56=49.5+1.5
Q2=44.5+5
D3=42.36 P56=51
Q2=49.5
Interpretation:
Fifty percent (50%) of the Thirty percent (30%) of Fifty-six percent (56%) of
scores in the distribution the scores in the the scores in the
are less than 49.5. distribution are below distribution are lower
42.36. than 51.

Measures of variation or the measures of dispersion are used to describe the

amount of variability or spread of scores in a data set. The foremost measures of
variability are range, inter-quartile range, quartile deviation, mean deviation,
standard deviation, and variance.
The range is the difference between the highest and the lowest
values in a data set. It is useful when we do a rough estimate of variation. It
is easily affected by extreme scores and gives no information as to where
other values lie.
Example 1: Compute for the range of the following ungrouped scores.

Group A Group B
Lowest scores
13 10
16 12
17 15
19 17
19 19
20 20
23 20

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 109
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

26 24
Highest scores 29 30

Group A Group B
Solution: Solution:
RA = HS – LS RB = HS – LS
RA = 29-13 RB = 30-10
RA = 16 RB = 20

Interpretation:
The range of Group A (RA = 16) is less than the range of Group B (R B =
20). This means that the scores in Group A have smaller variability or less
scattered than the scores in group B.

When finding for the range of grouped data, apply the formula below

R=HSUB−LS LB

Where:
HSUB = Upper boundary of the highest score
LSLB = Lower boundary of the lowest score

Example 2. Find the range of the scores of 40students in Science

diagnostic test.
X f
LSLB = 15-.5= 14.5 15-19 7
20-24 9
25-29 11
30-34 6
35-39 4
40-45 HSUB = 45 + .5
3 = 45.5
40

Solution:

R=HSUB−LS LB
R=45.5−14.5
R=31

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 110
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

When the computed value for the range is high, the scores in the distribution
are more dispersed, scattered, or heterogeneous. In contrast, when the range is
small, it means that the scores in the distribution are less dispersed, closer to
each other, or more homogeneous.
Inter-quartile Range (IQR) is a measure of variability that uses the difference
between the third (Q3) and the first quartile (Q1). It covers the middle fifty
percent of the scores in a data set as the point of dispersion. The lower and the
upper points of this middle 50% area are quartile 1 and quartile 3, respectively.
So, the IQR is the difference between the two quartiles.
Formula:

IQR=Q 3−Q1

Example: Determine the interquartile range using the scores below.

Solution:
Q1 Q3

Q 1=
( 1
4 ( ))
25+ 1−
1
4
Q 3=
( 3
4 ( ))
25+ 1−
3
4
Q1=( 6.25+ ( .75 ) ) Q3=( 18.75+ ( . 25 ) )
Q1=7 th score Q3=19 th score

IQR=51−21

IQR=30

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 111
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Median

14, 15, 16, 17, 17, 19, 21, 21, 24, 26, 27, 31, 36, 36, 41, 43, 44, 48, 51, 52, 53, 53, 54, 57, 59

Q1 = 7th score Q3 = 19th score

Interpretation:

The IQR of 30 reflects the extent of dispersion or how spread out the middle
50% of scores in a data set. Taking each quartile, the first and the third quartiles are
equally distant from the median which covers the middle fifty percent of the
scores in a data set as the point of dispersion.
Quartile Deviation, also called as the semi-quartile range, is a measure of
variability that is based on the range of middle 50% of the scores (IQR) divided by
two. It summarizes the amount by which every value within a dataset varies or
deviates from the mean.

Formula:
IQR
QD=
2

Quartile Deviation for Ungrouped Data

Example: Using the preceding value for IQR, find the QD for the given set of
scores.

Q3−Q1
QD=
2

51−21
QD=
2

30
QD=
2

QD=15

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 112
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Interpretation:
The amount that deviates from the median is 15.

Quartile Deviation for Grouped Data

x f cf<
25-29 3 3
30-34 4 7
35-39 7 14 cfp
40-44 7 fq 21
45-49 9 30
50-54 12 42 cfp
55-59 13 fq 55
60-64 1 56
65-69 2 58
70-77 2 60
n = 60

Q1 Q3
Given: Given:
Q k = Q1 Q k = Q3
k = 1 k = 3
1(60) 60 3(60) 180
Q1C = 40-44 = = 15 Q3C = 55-59 = = 45
4 4 4 4
LB = 39.5 40-0.5=39.5 LB = 54.5 55-0.5=54.5
cfp = 14 cfp = 42
fq = 7 fq = 13
ci = 5 ci = 5

( ) ( )
1(60) 3 (60)
−14 −42
4 4
Q1=39.5+ 5 Q3=54.5+ 5
7 13
Q 1=39.5+
15−14
7 ( 5 ) Q 3=54.5+
45−42
13
5 ( )
5 15
Q1=39.5+ Q3=54.5+
7 13
Q1=39.5+ 0.71 Q3=54.5+ 1.15
Q1=40.21 Q3=55.65
55.65−40.21
QD=
2
15.44(IQR)
QD=
2
QD=7.72

Interpretation:
This material is originally prepared and
The amount thatcontains digitalfrom
deviates signatures. Further reproduction
the median is 7.72. is prohibited without prior
permission from the authors. Page 113
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

The higher the value of the IQR and the QD, the more dispersed or spread
out the scores in the middle 50% of the score distribution. To determine which
group of distribution is more or less dispersed, there should be two or more
groups of distribution being compared since there is no standard value of a small
or large value of IQR and QD (Gabuyo, 2012).

Mean Deviation is the average deviation of all values in a distribution from

the arithmetic mean. It answers the question: How far, on average, all values
are from the middle (mean)? This is derived by computing the mean of the
absolute values of the differences between observed values of a variable and the
variable's mean divided by the number of cases.
Mean Deviation of Ungrouped Data

Formula:
∑| x−x̅ |
MD=
n
Where:
MD = mean deviation value
x = individual score
x̅ = mean value
n = number of cases

Example: Find the mean deviation of the scores of 15 students in a

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 114
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

unit-test in T.L.E. using the given scores below.

x x-x̅ |x−x̅ |
41 12 12
39 10 10
Solution:
38 9 9
38 9 9
36 7 7
35 6 6
31 2 2
29 0 0
27 -2 2
26 -3 3
26 -3 3
24 -5 5
19 -10 10
14 -15 15
12 -17 17
∑x=435 ∑ |x−x̅ | = 110
∑x
x=
n

435
x= =29
15

MD=
∑ |x−x|
15

110
MD=
15

MD=7.33

Interpretation: On average, the value that deviated from the mean ( x=29 ¿ is 7.33.

Mean Deviation for Grouped Data

Formula:
This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 115
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

f |X m −x̅ |
MD=
n

Where:
MD = mean deviation value
f = class frequency
Xm = class mark or midpoint of each category
x̅ = mean
n = number of cases

Example: Find the mean deviation of the given scores.

.
x f xm fx m X m−x̅ | X m−x̅ | f | X m−x̅ |
9-15 8 12 96 -11.89 11.89 95.12
16-22 13 19 247 -4.89 4.89 63.57
23-29 12 26 312 2.11 2.11 25.32
30-36 9 33 297 9.11 9.11 81.99
37-45 3 41 123 17.11 17.11 51.33
N = 45 ∑ fx m=¿ 107 ∑ f | X m−x̅ |=¿ 317.33
5

Solution:
To compute for mean:
∑ fx m
x=
n

1075
x=
45
1075
x=
45
x=23.88

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 116
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

To compute for mean deviation:

f |X m −x|
MD=
n
317.33
MD=
45
MD=7.05

Interpretation:

On average, the value that deviated from the mean ( x=23.88 ¿ is

7.05.

Variance and Standard Deviation

Standard Deviation is a quantity that indicates the average deviation

(distance) of all the observed values from the mean (how spread out a group of
numbers is from the mean). It is the square root of the variance. Variance, on
the other hand, measures how spread out the scores in a set of data (the average
degree to which each point differs from the mean—the average of all data
points). The higher the variance, the more scattered or spread out the scores are
from the mean.

Variance and Standard Deviation for Ungrouped Data

Formulas: Variance Standard Deviation

√
Population ∑ (x−µ)
2
∑ (x−µ)2
σ 2= σ=
N N

√
Sample ∑( x−x̅ )
2
∑ ( x−x̅ )2
s2= sd=
N −1 N−1

Where:
sd = standard deviation for the sample
σ = standard deviation for the population
S2 = variance for the sample
This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 117
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

σ2= variance for the population

x = score
x̅ = mean value for sample group
µ = mean value for population group
N = number of scores

Example: Find the variance and standard deviation of the scores of sampled
12 students in a short test in Geography using the given scores below.

Students x x- x̅ ( x−x̅ )2 Solution:

1 19 5.42 29.38
2 17 3.42 11.70 Variance
3 16 2.42 5.86 2
∑( x−x̅ )
4 16 2.42 5.86 s2=
N −1
5 15 1.42 2.02
6 14 0.42 0.18 2 122.92
s=
7 14 0.42 0.18 12−1
8 13 -0.58 0.34 2 122.92
9 12 -1.58 2.50 s=
11
10 10 -3.58 12.82 2
11 9 -4.58 20.98 s =¿ 11.17
12 8 -5.58 31.14
N Standard Deviation
∑x 163 122.92 (calculate for the square root of the
13.58 variance)

√
∑ ( x−x̅ )2
2
∑ (x−x̅ ) sd=
163 N−1
x̅ =
12 sd=√ 11.17
sd=3.34

Interpretation:

The average degree to which the scores of 12 students deviate or differ from the
This material is originally
meanprepared
is 3.34and contains
(sd). On thedigital signatures.
other hand, Further reproduction
the extent is prohibited
of dispersion of without prior (variability
their scores
permission from the authors. 2
Page 118
of scores) from the mean is 11.17 (sd ).
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Variance and Standard Deviation for Grouped Data

Formulas: Variance Standard Deviation

√
Population 2 ∑ f ( x m−µ)
2
∑ f (x m−µ)2
σ = σ=
N N

√
Sample 2∑ f ( x m−x̅ )
2
∑ f ( x m −x̅ )2
s= sd=
N −1 N−1
Where:
sd = standard deviation for the sample
σ = standard deviation for the population
S2 = variance for the sample
σ2= variance for the population
f = frequency
xm=class mark
x = score
x̅ = mean value for sample group
µ = mean value for population group
N = number of scores

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 119
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Example: Compute for the variance and standard deviation of the scores of
30 students (population) in a 45-item quarterly test in P.E.

x f xm fxm x̅ (xm - x̅ ) (xm - x̅ ¿2 f(xm - x̅ )2

22-25 3 23.5 70.5 32.6 -9.1 82.81 248.43
26-29 5 27.5 137.5 32.6 -5.1 26.01 130.05
30-33 9 31.5 283.5 32.6 -1.1 1.21 10.89
34-37 8 35.5 284 32.6 2.9 8.41 67.28
38-43 5 40.5 202.5 32.6 7.9 62.41 312.05
30 978 768.7

978 ∑ f(xm - x̅ )2
x=
30

Solution:
2
∑ f ( x m−µ)
Variance σ = 2
Interpretation:
N
2 768.70 The extent of dispersion of
σ =
30 scores from the mean is 25.62 (
2
σ ¿. On the other hand, the
2
σ =25.62
variation between the scores
Standard Deviation relative to the mean is 5.06
(calculate for the square root of the variance) (sd).
sd=
√
∑ ( x−x̅ )2
N
sd=√ 25.62
sd=5.06

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 120
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Measures of Skewness is a measure of the symmetry in a distribution (SPC

for Excel, 2016). It provides numerical measures of the shape of data to present
a more precise description of performance.

Formula:
3(x−x̃ )
sk=
s

Where:
x=mean value
~
x=median value
s=standard deviation

Measures of Skewness is a measure of the symmetry in a distribution (SPC

for Excel, 2016). It provides numerical measures of the shape of data to present
a more precise description of performance.

Formula:
3(x−x̃ )
sk=
s

Where:
x=mean value
~
x=median value
s=standard deviation

Broadly speaking, there are two types of skewness which are: (a) positive
skewness and (b) negative skewness. The positively skewed distribution is
skewed to the right where the longer or flatter end tail goes to the right part of
the curve. The sk > 0. In this distribution, the mean value is higher than the
median which means that most of the scores are lower than the mean. This
happens when the students either found the test to be difficult or performed
poorly in the test and consequently achieved scores lower than the mean.
On the other hand, the distribution is negatively skewed when the thinner
end tail is the left of the curve. The sk < 0. The shape of the data implies that
most of the students’ scores are above the mean. The mean value is less than
the median.

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 121
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Positively Skewed Distribution (sk > Negatively Skewed Distribution (sk < 0)
0)

Figure 14. Graphical Representation of Positively and Negatively Skewed

Distribution

Example: Find the coefficient of skewness of 20 scores of the students who

took a reading test with the following given values: mean value= 15.35,
median value = 17.50, and sd= 5.

Solution:
3(15.35−17.5)
sk=
5
3(−2.15)
sk=
5
−6.45
sk=
5
sk=−¿1.29

Interpretation:
The sk coefficient is -1.29. The yielded value presents that the score
distribution is negatively skewed. It could be inferred from the result that the
students showed a very good performance in the reading test. This could
also mean that the students found the test easy.

Normal Distribution

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 122
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Data can be distributed or spread out either more on the right (positively
skewed) or more on the left (negatively skewed). There are likewise cases that
data tend to be around the central values and form the “bell curve.” The normal
distribution is a symmetric distribution where most of the observations cluster
around the central peak and the probabilities for values further away from the
mean taper off equally in both directions (Frost, 2020).

The data are normally distributed when the mean is equal to the median and
the mode (mean=median=mode). In this distribution, 50% of the scores or
values are less than the mean and 50% are greater than the mean. Generally,
there are 4 standard deviations (sd) on either side of the mean. The area within
the 4 standard deviations of the mean covers 99.98 of the values, 99.72% of the
values are within 3 standard deviations of the mean, 95.44% of the values are
within 2 standard deviations of the mean, and 68.26% of the values are within 1
standard deviation of the mean. A normal distribution is a symmetrical dataset
which has a skewness equal to 0.

40 44 48

Figure 15. The Normal Distribution Curve

Example: A quarterly test in Science was given to 42 grade four pupils. The
class achieved an average score of 40. Find the number of scores that lie within 2
standard deviations above the mean.

Given:
x=40
Score found 2 sd above the mean = 48

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 123
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

No. of scores/values = 42

Area for scores lower than 48 (2 2 SD above the

Sd above the mean) mean

28 32 36 40 44 48 52

Solution:
Number of scores within 2 SD above the mean
= 34.13 %+ 13.59% = 47.72%
= (.4772) (42)
= 20 scores

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 124
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Interpretation:
There are, more or less, about 20 scores in the distribution that lie 2
SD above the mean (between 40 and 48).

Standard Scores

Teachers can come up with further and more meaningful analysis and
interpretation of students’ performance. To realize this, they can make use of
the actual or raw scores that are based on the number of items that were
answered correctly by the students. Another way of doing this is by transforming
or converting the raw scores to standard scores.

A standard score is a set of observations that have been standardized around

a given mean and standard deviation (Morrow, et al. (2016). This is commonly
expressed in number that represents the performance of the students with
reference to the average norm or the criterion. The fundamental standard
scores include the z-scores, t-scores, stanines, and percentile ranks.

The z-scores tell how far from the mean a score or value is in terms of the
quantity of standard deviation. The z-scores can present the exact location of
scores above and below the mean value.

Formula:
x−x
z=
s
Where:
x=score
x=mean value
s=standard deviation

Example: Compare the scores earned by Chris in Math and Science with 50
items each using the z-scores. Refer to the matrix table below for the given
values for scores, mean values, and standard deviations. Based on the average
performance of the whole class, in which test Chris performed better?

Given Chris’s Score in Math Chris’s Score in Science

x 38 40
x 35 42
s 3 2
Solution:
38−35 40−42
z= z=
3 2
3 −2
z= z=
3 2

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 125
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

z=1 z=−1

Interpretation:
Chris’s score in Math is 1 unit standard deviation (z=1) above the mean.
Her score in Science is 1 unit standard deviation below the mean (z=-1).
Based on the average performance of the class, Chris performed better in
Math than in Science test.

T-score is another commonly used standard score. A score of 50 represents

the mean and 10 represents the standard deviation. A difference of 10 from the
mean indicates a difference of one standard deviation.

Formula:

10 ( x−x )
T =50+ or T =50+10 z
s
Where:
x=score
x=mean value
s=standard deviation
z=z−score

Example: Using the above example, determine the t-score equivalent of

Chris’s scores in Math and Science tests.

Given Chris’s Score in Math Chris’s Score in Science

X 38 40
x 35 42
s 3 2
z 1 -1
Solution:
T =50+10 ( 1 ) T =50+10 (−1 )
T =50+10 T =50−10
T =60 T =40

Interpretation:
Chris’s score in Math is 1 unit standard deviation (T=60) above the mean;
while her score in Science is 1 unit standard deviation below the mean
(T=40). Based on the average performance of the class, Chris performed
better in Math than in Science test.

Note: A difference of 10 from the mean indicates a difference of one standard

deviation

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 126
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Raw score - 38 (Math) 40 (Science)

Z-scores - 1 sd above the mean 1 sd below the mean
t-scores - 60 40
Stanine or standard nine is a way to rescale raw scores using a nine-
point scale ranging from 1 (indicating very poor ability) to 9 (indicating very
superior ability) relative to the norm of the reference group. Similar to z-
score and T-score, it can be used to assign a number that will represent the
performance of a student relative to the performance of all test-takers in a
test.

Table 12. Stanine Scores and the Corresponding Percentage of

Scores and Descriptions
Calculation of Stanine Scores
Stanine Percentage Description
9 4% Very Superior
8 7% Superior
7 12% Considerably Above Average
6 17% Slightly Above Average
5 20% Average
4 17% Slightly Below Average
3 12% Below Average
2 7% Poor
1 4% Very Poor

Example: A total of 100 applicants took an entrance test administered

by a prestigious university in the locality. For a student to qualify for
enrolment in the university, he or she must obtain at least a stanine
rating of 6. The scores of the 100 test-takers arranged from lowest to
highest are reflected in the matrix below. The scores in the table are
from a population that is normally distributed with a mean of 400 and
a standard deviation of 25. If Andrew, one of the student applicants,
obtained the actual score of 399, will he be admitted for enrolment in
the university? How do we appropriately describe his performance in
the test?

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 127
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Scores of the 100 Test-takers in the Entrance Test

350 380 394 410 422 Andrew’s
351 380 394 410 422 Score
353 381 395 411 422
354 382 397 412 423
356 383 398 412 424
356 384 398 412 425
360 385 399 412 425
362 386 401 413 426
364 386 401 414 427
365 387 401 414 428
365 387 402 414 432
366 387 403 414 433
368 387 403 415 436
370 388 404 416 440
371 388 405 416 441
371 390 405 417 445
372 390 405 417 450
372 390 406 417 452
376 392 408 418 452
379 393 409 421 456

Stanine 1

Stanine 2

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 128
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Solution:
The first 4% ranked scores (raw scores 350-354) are assigned stanine 1. The
next 7% ranked scores (raw scores 356-365) are assigned stanine 2. The next 12%
ranked scores (raw scores 366-381) are assigned stanine 3. The next 17% ranked
scores (raw scores 382-393) are assigned stanine 4. The next 20% ranked scores
(raw scores 394-409) are assigned stanine 5. Andrew’s score is 399 that comes
between 394 and 409 with a given stanine score of 5 interpreted as average. He
failed to obtain the cut-off rating of stanine 6 set by the university; hence, he can’t
be accommodated to enroll in the university.

The percentile rank indicates the percentage of a score in a frequency

distribution that is equal or lower a given score.

Formula:

PR= ( CF +.5n F ) x 100

b g

Where:
PR= percentile rank
CF b=cumulative frequency below the given score
F g=frequency of the given score
n=number of scores∈the distribution
Example:
Using the data below, find the percentile rank of 37 (raw score).

Scores f cf Solution:

49
47
2
1
52
50
PR=( CF +.5n F ) x 100
b g

43 3 49
PR=( ) x 100
34+.5(2)
41 4 46
52
38 6 Fg 42
CFb

( 34+1
52 )
37 2 36
PR= x 100
35 3 34
32 5 31
PR=( ) x 100
29 7 26 35
27 11 19 52
25 3 8
PR=( ) x 100
35
24 5 5 52
n=52
PR=.67 x 100

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 129
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Interpretation:
The raw score of 37 has the equivalent of 67% percentile rank. This means that
67% of the test-takers had scores of equal to or lower than 37.

To illustrate the relationship between the standard scores (z-score, T-score,

and stanine), examine Figure 16.

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 130
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

For example, if a student has a raw score of 65 with an SD of 5, the equivalent z-

score is 1, T-score is 60, and stanine is 6.

Correlation analysis examines the strength of the linear association between

two continuous variables. A correlation coefficient symbolized by r measures the
extent of the relationship that can range from -1.0 to + 1.0. The closer the
coefficient to +1.0, the stronger the relationship.

The direction of correlation determines whether the relationship exists

(positive or negative correlation) or does not exist (zero correlation).

A positive correlation depicts a linear relationship between two variables

such that when one increases, there is a corresponding increase in the other
variable. They increase and decrease together. The value of r is greater than 0.

Student Math Scores English Scores 30

s 25
1 25 24
2 23 23 20
Math Scores

4 22 21 15
5 21 19
10
6 19 17
7 18 15 5
8 15 14 0
9 14 12 8 10 12 14 16 18 20 22 24 26
10 10 10 English Scores
11 8 9

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 131
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Interpretation:
As Math scores increase, the English scores also increase. The data points
make a straight line going from the origin out to high x- and y-values.

A negative correlation exists when a decrease in one variable predictably

meets with a comparable increase in the other. The value of r is less than 0.

Students Math English 30

Scores Scores
1 25 9 25
2 23 10 20
Math Scores

4 22 12
5 21 14 15
6 19 15 10
7 18 17
5
8 15 19
9 14 21 0
10 10 23 6 8 10 12 14 16 18 20 22 24 26
11 8 24 English Scores

Interpretation:
As Math scores increase, English scores decrease. When the points in the
coordinate plane are connected, the line goes from a high-value on y-axis down to
low-value on x-axis.

A zero correlation shows the absence of a relationship between two

variables. The value of r is equal to 0.

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 132
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Students Math English 30

Scores Scores
25
1 25 13 20
2 23 10

Math Scores
4 22 15 15
5 21 0
10
6 19 8
7 18 2 5
8 15 0
0
9 14 25 8 10 12 14 16 18 20 22 24 26
10 10 5
English Scores

Interpretation:
No linear relationship exists between the Math and English scores. The points
representing the Math and English scores could not be connected by a line. that is
best fitted to all values.

Determining the Relationship Using Pearson r.

Pearson’s product-moment correlation coefficient is the test statistic that

measures the strength of a linear association between two continuous variables. It is
denoted by r. It attempts to draw a line of best fit through the data of two variables
indicates how far away all these data points are to this line of best fit (Laerd
Statistics, 2020).

Example:
Teacher Deanna wanted to know if the scores of the students during the
entrance test correlate with their current average performance in first-year
subjects. She used a sample of 21 students for analysis.
The formula for Pearson r is:

n ∑ xy−( ∑ x ) (∑ y )
r xy =
√( n ∑ x −( ∑ x ) )(n ∑ y −(∑ y ) )
2 2 2 2

Where:

r xy =correlation coefficient
n=number of cases
∑ x=summation of x scores
∑ y=summation of y scores
∑ xy=summation of the product of x ∧ y scores
∑ x =∑ of squared x scores
2

∑ y =∑ of squared y scores
2

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 133
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

The more efficient and easier way of calculating for the Pearson r is using the
Microsoft Excel application. The steps are the following:

Step 1. Encode all data in Microsoft Excel Spreadsheet.

Step 2. To open the data analysis in Excel, click the File tab > options

Step 3. Click the Add-Ins category.

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 134
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Step 4. In the Manage box, select Excel Add-ins and then click Go.

Step 5. In the Add-Ins available box, select the Analysis ToolPak check box, and then
click OK.

(Note: Skip steps 2, 3, 4, and 5 if the data analysis is already on the top
menu.)

Step 6. On the top menu, click data analysis > correlation > OK.

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 135
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 136
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Step 7. The correlation dialogue box will appear. Click the input range > new
worksheet ply > OK.

Step 8. The generated output will appear in a new sheet.

Step 9. Analyze and interpret the correlation coefficient using Table 13

below.

Table 13. Correlation Coefficient Value

Correlation Coefficient Interpretation

0.00 No Correlation
±0.01 to ±0.20 Slight or negligible correlation
±0.21 to ±0.40 Low correlation
±0.41 to ±0.60 Moderate correlation
±0.61 to ±0.80 High correlation
±0.81 to ±0.99 Very high correlation
1.00 Perfect correlation

Interpretation:
The r-value of 0.4678 or 0.47 (above ±0.41 but lower than ±0.60)
shows that there is a moderate correlation between the scores of 21
students during the entrance test and their current average grades.

Determining the Relationship Using Spearman rho

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 137
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Spearman rho correlation also measures the strength of association

between two variables but unlike the Pearson r, it uses data in rank or ordinal
scale.

Formula:
2
6∑ D
ρ=1−
n ( n2−1 )

Where:
ρ=Spearman rho coefficient value
D=difference between pairs of rank
N=number of cases

Example:
Teacher Abby wanted to determine if there is a correlation
between the written and practical performance of the 17 students in
TLE class using their ranked scores in both assessments.

Students Written Test Practical Test

Rank Rank D D2
1 3.5 3.5 0.0 0.00
2 10.5 15 -4.5 20.25
3 2 6 -4.0 16.00
4 12 9.5 2.5 6.25
5 16.5 16 0.5 0.25
6 1 3.5 -2.5 6.25
7 13 8 5.0 25.00
8 3.5 1 2.5 6.25
9 5 2 3.0 9.00
10 7 5 2.0 4.00
11 7 12 -5.0 25.00
12 7 7 0.0 0.00
13 9 9.5 -0.5 0.25
14 10.5 11 -0.5 0.25
15 14 13 1.0 1.00
16 15 14 1.0 1.00
17 16.5 17 -0.5 0.25
∑D2 = 121.00

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 138
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Solution: Interpretation:
6 (121) The ρ-value of 0.85
ρ=1−
17 ( 17 −1 )
2 indicates a very high positive
correlation between the
726 written and practical tests in
ρ=1−
17 ( 289−1 ) the TLE class. It could be
726 gleaned from the results
ρ=1−
17 ( 288 ) that the students who
achieved high scores in the
726 written tests also exhibited
ρ=1−
4896 good performance in the
ρ=1−.15 practicum test.
ρ=0.85

To perform this in Microsoft Excel, perform the following steps:

Step 1. Click data analysis > correlation > OK

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 139
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Step 2. The correlation dialogue box will appear. Click the input range > new
worksheet ply > OK.

Step 3. The generated output will appear in a new sheet.

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 140
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

The formula for manual computation and the Excel yielded a similar result
(ρ0.85) showing a very high level of a positive correlation between the
written and practical test scores of the 17 students in the TLE class.

Activity 1. Calculating the Measures of Central Tendency. Find the mean, median,
mode, Q3, D2, and P65 of the ungrouped scores below. Analyze and interpret each
result. (3 points each).

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 141
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

35 34 34 33 31 29 27 27 27 25
23 21 21 20 19 18 17 17 16 14

Mean

Median

Mode

Quartile 3

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 142
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Decile 2

Percentile 65

Activity 2. Illustrating Skewness. Illustrate in a graph the kind of distribution the

above scores in Activity 1 show (normal distribution, positively skewed,
negatively skewed). Explain briefly what the shape of the distribution implies. (5 pts)

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 143
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Activity 3. Constructing Frequency Distribution. Arrange the scores in the data

matrix below into groups using the class interval size (ci) of 7. Construct a
a frequency distribution with columns for class boundaries (c b) and class marks (xm).
(10 points)

28 42 48 25 30 41
37 16 17 26 32 29
32 21 23 31 24 18
32 23 22 36 23 13
33 25 35 32 21 36
34 32 26 27 35 29

x f cb Xm

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 144
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Activity 3. Measuring Variation. Find the range, mean deviation, and variance of
the scores of the population group in Activity 2. (3 points each) (Use the grouped
data).

Range

Mean Deviation

Variance

Activity 4. Standardizing Scores. Using the data matrix below, convert the raw
score 24 to its equivalent z-score, T-score, and standard nine, and percentile rank.
(3 pts each)
x f
30 1
28 3
27 4
25 7
24 2
22 2
20 5

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 145
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

19 6
17 3
16 4
14 2
11 1

N= 40

Z-score

T-score

Stanine

Percentile Rank
x y
1 27 22
2 35 25
3 65 38
4 64 37
5 60 41
6 65 50
7 36 24
8 24 22
Activity 5. Describing Relationship. The data
9 29 23
in the x-column show the scores of the 23
10 35 21
students in the comprehensive test in English.
11 60 37
The y- column indicates their scores in the
12 29 26
English achievement test they took at the end of the
13 34 30
school year. Determine if the two scores are
14 56 34
correlated. State your interpretation
15 55 41
of the result.
16 58 27
17 59 26
18 60 24 Please write your answer here.
This material is originally19
prepared and27contains digital
23 signatures. Further reproduction is prohibited without prior
permission from the authors.
20 17 22 Page 146
21 34 19
22 29 20
23 31 23
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

Exercises.
I. True or False. Write true if the statement is correct and false if it is wrong.
Write your answer on the space provided before each number. (1 point each)

_______ 1. The z-score of 2 is equal to the T-score of 70.

_______ 2. When the range is small, the scores in the distribution are more
dispersed.
_______ 3. The distribution is positively skewed when most of the

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 147
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

scores are below the mean.

_______ 4. The third quartile (Q3) is the central point that lies between the
median and the highest number of the distribution.
_______ 5. The mode is the best measure of central tendency when
determining for the most frequent score in the distribution.

II. Matching Type. Match the statistical tool/measure specified in column B

with the type of analysis it serves in column A. Write the CAPITAL LETTER of
the correct answer on the space provided before each number. (1 point
each)
A B
___ 1. To determine the grade-point average (GPA) A. Mean
of a student B. Mean deviation
___ 2. To show how strong the two variables are C. median
related D. Measure of
___ 3. To determine the average distance of all correlation
values from the middle E. Measure of
___ 4. To describe the degree of departure of the skewness
scores from the symmetry F. Range
___ 5. To calculate for the difference between the G. Variance
highest and the lowest scores

III. Multiple Choice. From the listed options, choose the best answer that
would address each question/item below. Write the CAPITAL LETTER that
corresponds to your answer on the space provided before the number. (1
point each).

___ 1. Which measure of central tendency is not affected by extreme scores?

A. Mean C. Mode
B. Median D. Quantile

___ 2. When the skewness coefficient for the whole class is greater than 0 (sk > 0),
it means that ______________________.
A. most students in the class performed well in the test.
B. most students in the class did not perform well in the test.
C. 50% of the students scored above the mean and 50% scored below
the mean.
D. the number of students who scored above the mean is greater than
those who scored below the mean.

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 148
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

___ 3. Which statement reflects the property of the inter-quartile range?

A. It is easier to calculate than range.
B. Its point of dispersion is the mean value.
C. It increases the influence of extreme values.
D. It only considers the middle 50% of the scores in the distribution

___ 4. Ten students took a 50-item Science test. Scores were analyzed using the
mean deviation. The obtained results were: Mean = 33.63, MD = 10.
Which of the analyses stated below is most appropriate?
A. On average, 10 students scored higher the mean of 33.63.
B. On average, the scores deviated from the mean by 10 points.
C. On average, 10 students got scores lower than the mean of 33.63.
D. On average, the value that deviated from the mean of 33.63 is 10.

___ 5. Which comparison of measures of central tendency shows a normal

distribution of scores?
A. mean = median = mode
B. mean ˂ median ˂ mode
C. mode ˃ median ˃ mode
D. mode ˂ median = mode

___ 6. Forty students took a test in History class. After using the formula, the value
of sk was -0.1. What does this mean?
A. The students performed poorly in the said test.
B. The students had scores hat lie below the average.
C. Most students obtained scores lower than the median.
D. The students showed excellent scores in the test.

___ 7. Emma scored 50 in the English examination which equals to 88th percentile.
What does this mean?
A. Emma’s score is less than 88% of her classmates.
B. 88% of Emma’s classmates scored higher than 50.
C. Emma’s score is 12% lower than the perfect score.
D. 88% of Emma’s classmates had scores equal to or lower than 50.

___ 8. Sarah score is 1 standard deviation above the mean of her class in Filipino
and 3.5 standard deviations above in English. What does this mean?
A. She is better in English than in Filipino.
B. She is better in Filipino than in English.
C. She scored well both in English and Filipino.
D. She scored poorly both in English and Filipino.

___ 9. A national achievement test is given yearly to 3rd graders. The test has a
mean score of 80 standard deviations of 4. If Katie’s z-score is 2, what was
her score on the test?
A. 80 C. 85

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 149
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

B. 83 D. 88

___ 10. When high scores in distribution x are associated with low scores in
distribution y, the two sets of scores (in x and y) are
____________________.
A. not associated C. negatively correlated
B. positively correlated D. indirectly correlated

IV. Restricted Essay. Distinguish the measures of central tendency from the
measures of variation in NOT more than 100 words. (5 points each)

Scoring Rubric
Level Score Description
Exemplary 5 points  All main points are included with good
organization and a clear flow of ideas.
 Sentences are well-structured and free
from grammatical and/or syntactic errors
Very Good 4 points  One or two points is/are missing.
 There are minor problems in logic and
construction.
 Few grammatical/syntactic errors are
found.
Good 3 points  The answer is generally accurate but the
quality of writing is inferior.
Needs 2 points  The answer does convey a full
Improvement understanding of the lesson.
 Sentences contain major
grammatical/syntactic error
Unsatisfactory 1 point  The answer is inaccurate or deviates from
what is asked.
 Sentences are disorganized and contain
major grammatical/syntactic errors.

Please write your answer here.

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 150
 CHAPTER 4: INTREPRETATION AND UTILIZATION OF ASSESSMENT RESULTS

References:

Calderon, J., and Gonzales, E. (2006). Measurement and Evaluation. Manila, Philippines:
National Book Store.

Frost, J. (2020). Normal distribution in statistics. Retrieved September 1, 2020,from

https://statisticsbyjim.com/basics/normal-distribution/.

Gabuyo, Y. (2012). Assessment of Learning 1. Manila, Philippines: Rex Book Store.

Laerd Statistics. (n.d.) Pearson Product-Moment Correlation. Retrieved September 17,

2020, from https://statistics.laerd.com/statistical-guides/pearson-correlation-
coefficient-statistical-guide.php.

Morrow, J. (2016). Measurement and Evaluation in Human Performance. 5 th edition. USA:

Human Kinetics.

Professional Learning Online of Teachers and For Teachers. (2020). DepEd K to 12 Grading
System Steps for Computing Grades. Retrieved September 13, 2020, from
https://www.teacherph.com/deped-grading-system/.

Romijn, Jan-Willem (2014). "Philosophy of statistics". Stanford Encyclopedia of Philosophy.

Smart, P. (2019). The importance of statistics in education. Retrieved September 10,

2020, from https://machinep.com/importance-of-statistics-in-education/.

SPC for Excel. (2016). Are the skewness and kurtosis useful statistics? Retrieved September
17, 0202, from https://www.spcforexcel.com/knowledge/basic-statistics/are-
skewness-and-kurtosis-useful-statistics.

Statistics Canada. (2017). Frequency distribution tables. Retrieved September 11, 2020,
from https://www150.statcan.gc.ca/n1/edu/power-pouvoir/ch8/5214814-eng.htm.

This material is originally prepared and contains digital signatures. Further reproduction is prohibited without prior
permission from the authors. Page 151