STATISTICS – TOOLS FOR ANALYZING AND

INTERPRETING ASSESSMENT DATA

A. MEASURES OF CENTRAL TENDENCY

• It is a single value that is used to identify the center of the data.

• It is taught as the typical value in the set of scores.
• It tends to lie within the center if it is arranged from lowest to highest or vice
versa.
• It describes the performance of one group based on their scores.
A.1. MEAN
• Refers to the arithmetic average
• Used when the data are interval or in ratio level of measurement
• Used when the frequency distribution is regular, symmetrical, or normal
• Easily affected by extreme scores
• Very easy to compute
• Used to compute other measures such as standard deviation, coefficient of
variation, skewness, and z-score
• It measures stability
• It is the most reliable measure of central tendency only if it is a normal
distribution
A.1.1. STEP IN SOLVING THE MEAN VALUE USING RAW SCORES
1. Get the sum of all the scores in the distribution.
2. Identify the number of scores (n).
3. Substitute the given formula and solve the mean value.
A.1.2. FORMULA
𝚺𝒙
Mean = 𝒏

Example:
Scores of 15 students in the Mathematics 1 quiz consist of 25 items. The highest score
is 25 and the lowest is 10. Here are the raw scores: 25, 20, 18, 18, 17, 15, 15, 15, 14, 14,
13, 12, 12, 10, 10. Find the mean of the following scores.
Solution:
228
Mean = = 15.2
15

A.1.3. INTERPRETATION OF RESULTS

1. 15.2 is the average score in the Math 1 quiz
2. Students who scored above the mean (15.2) performed well in the said
examination
3. Students who scored below the mean (15.2) did not perform well in the said
examination
A.2. MEDIAN
• Refers to the centermost scores when the scores in the distribution are arranged
according to magnitude (from highest to lowest score or vice versa)
• Used when the middlemost score is desired
• Used when the data are in an ordinal level of measurement
• Used when the frequency distribution is irregular or skewed, asymmetrical, or
not badly skewed
• Used when there are extreme scores
• Not affected by extreme scores because it is a positional measure
• May not be an actual observation in the data set
• Most reliable if there are extreme scores
A.2.1. MEDIAN OF UNGROUPED DATA
1. Arrange the scores from lowest to highest or vice versa
2. Determine the middle score in a distribution if n is an odd number, and get the
average of the two middle scores if n is an even number.
Example:
Solve the Median (Ungrouped Data)
19, 17, 16, 15, 10, 5, 2
Median = 15

A.2.2. INTERPRETATION OF RESULTS

1. 50% of the scores are above 15
2. 50% of the scores are below 15
A.3. MODE
• Refers to the score/s that occurs most frequently in the score distribution
• Used when the data are in nominal level of measurement
• Used when a quick answer is need
• Used when the score distribution is normal, regular, symmetrical
• Can be used for quantitative, as well as qualitative data
• May not be unique
• Not affected by extreme values
• May not exist at times
A.3.1. TYPES OF MODE
1. Unimodal – One mode
2. Bimodal – Two Modes
3. Trimodal – Three Modes. Also considered multi-modal.
Examples:
Find the mode and identify its classification
1. Score of Section A: 25, 24, 24, 20, 20, 20, 16, 12, 10, 7
Mode: 20 Type of Mode: Unimodal

2. Score of Section B: 25, 24, 24, 20, 18, 18, 17, 10, 9, 7
Mode/s: 24 and 18 Type of Mode: Bimodal

3. Score of Section C: 25, 25, 25, 21, 21, 21, 22, 18, 18, 18
Mode/s: 18, 21, and 25 Type of Mode: Trimodal or Multi-modal
 B. MEASURES OF VARIABILITY

• It is a single value that is used to describe the spread of scores in a

distribution, that is above or below the measures of central tendency.
B.1. RANGE
• It is the difference between the highest score and the lowest score in the data
set.
• Used when the score distribution is normal, regular, symmetrical
• Easily affected by extreme scores
• Only two scores are needed to compute its value
• Easy to compute
• Rough estimation of variation/dispersion
B.1.1. FORMULA
Range = HS – LS
Example:
Below are the scores of 10 students in Mathematics and Science. Find the range and in
what subject has a greater variability.
Mathematics Science Range of Mathematics
35 35
HS – LS = 62 – 33 = 29
33 40
45 25
55 47
62 55 Range of Science
34 35 HS - LS = 57 – 25 = 32
54 45
36 57
47 39
40 52

Interpretation:
• The scores in Science have greater variability.
• The scores in Science are more scattered than the scores in Mathematics.

B.1.2. INTERPRETATION OF RESULTS: (RANGE, SD, AND QD)

• If the Range, Standard Deviation, or Quartile Deviation is SMALL
– then the scores in the distribution are CLOSER, CLUSTERED,
HOMOGENOUS, SCORES ARE LESS VARIED, OR HIGHLY BUNCHED
TOGETHER

• If the Range, Standard Deviation, or Quartile Deviation is LARGE

– then the scores in the distribution are DISPERSED, SCATTERED,
SPREAD APART, FAR FROM EACH OTHER, HETEROGENOUS, OR SCORES
ARE MORE VARIED
B.2. QUARTILE DEVIATION
• It is half of the difference between the third quartile (Q3) and the first quartile
(Q1).
• It is based on the middle 50% of the range, instead of the range of the entire set
of distribution.
• It reduces the influence of extreme scores
• It is not as easy to compute compared to range
• Its score distribution is irregular, asymmetrical, skewed
B.2.1. FORMULA
Quartile Deviation = (Q3-Q1)/2

B.3. STANDARD DEVIATION

• It is the most important and useful measure of variation.
• It is also the most commonly used.
• It is the square root of the variance.
• It is an average of the degree to which each set of scores in the distribution
deviates from the mean value.
• It is a more stable measure of variation because it involves all the scores in a
distribution compared to range and quartile deviation.
• Score distribution is regular, symmetrical, normal
• Interval or ratio level of measurement
B.3.1. PROCEDURE
1. Find the mean.
2. Subtract the mean from each score to get the deviation.
3. Square the deviation.
4. Get the sum of the squared deviations.
5. Divide the sum by the number of cases.
B.3.2. FORMULA

∑𝑑2
Standard Deviation = √𝑁−1

Example: Compute the standard deviation of the given data.

̅)
Mean (𝒙 d = (x – 𝒙
̅) d2
5 -4 16
7 -2 4
9 0 0
11 2 4
13 4 16
∑𝒅𝟐 = 𝟒𝟎

Solution:

40
SD = √ 4 = 3.16
 C. MEASURES OF SKEWNESS

• Describes the degree of departure of the scores from a symmetry

C.1. CLASSIFICATIONS
1. Positively Skewed Distribution – Sk > 0
2. Negatively Skewed Distribution – Sk < 0
3. Normally Distributed – Sk = 0

C.2. FORMULA
𝟑(𝑴𝒆𝒂𝒏−𝑴𝒆𝒅𝒊𝒂𝒏)
Sk = 𝒔𝒕𝒂𝒏𝒅𝒂𝒓𝒅 𝒅𝒆𝒗𝒊𝒂𝒕𝒊𝒐𝒏

C.3. POSITIVELY SKEWED - (Mean > Median > Mode)

• Skewed to the RIGHT
• Is a distribution where the thin end tail of the graph goes to the right part of the
curve
• This happens when most of the scores of the students are BELOW the mean
• Students are academically poor
• It tells you only on poor performance of takers but not the reasons why students
did poorly in the said examination
• Reasons of poor performance: (1) ineffective teaching method and instruction;
(2) students’ unpreparedness; (3) test items are very difficult; (4) not enough time
to answer test items

P positively skewed
R right
D difficult test
L low score

C.4. NEGATIVELY SKEWED – (Mode > Median > Mean)

• Skewed to the LEFT
• Is a distribution where the thin end tail of the graph goes to the LEFT part of the
curve
• This happens when most of the scores of the students are ABOVE the mean
• It tells you only on good performance of takers but not the reasons why students
did well in the said examination
• Possible reasons of high scores: (1) students are smart; (2) enough time to
finish examination; (3) test items are very easy; (4) effective instruction; (5)
students have prepared for the examination
 N negatively skewed
L left
E easy test
H high score

C.5. NORMAL DISTRIBUTION – (Mean = Median = Mode)

• Is a special kind of symmetric distribution
• Can be determined using the values of the mean and standard deviation
C.5.1. PROPERTIES OF NORMAL DISTRIBUTION
• The curve has a single peak, meaning the distribution is unimodal.
• It is a bell-shaped curve.
• It is symmetrical to the mean.
• The end tails of the curve can be extended indefinitely in both sides.
• The shape of the curve will depend on the value of the mean and standard
deviation.
 D. MEASURES OF RELATIVE POSITION

• Indicates where a score is in relation to all other scores in the distribution.

• They make it possible to compare the performance of an individual on two or more
different tests.
• Identifying the position of the score; comparing the performance between two tests

D.1. PERCENTILE RANK

• It is the percentage of the scores in the frequency distribution which are
lower.
• This means that the percentage of the examinees in the norm group who
scored below the score of interest.
• It is used to clarify the interpretation of scores on standardized tests.

Example:
Maria’s raw score in English Class is 82 which is equal to 90th percentile.

Interpretation:
1. This means that 90% of Maria’s classmates got a score lower than 82.
2. She surpassed 90% of her classmates.
3. She belongs to the upper 10% of the class.

D.2. Z-SCORE
• It tells how far from the mean the score is.
• The number of standard deviation units a score is above or below the mean
of a given distribution.
• A positive z-score = measures the number of standard deviations a score
is above the mean.
• A negative z-score = gives the number of standard deviations a score is
below the mean.
• Zero z-score = means the score is in the mean.

D.2.1. FORMULA
ZSCORE = (x – mean) / standard deviation

Example:
Jemar had a score of 59. He is compared with other examinees; the mean and
standard deviation of this group are 50 and 5, respectively. What is Jemar’s z-
score?
Solution:
Z = (x – mean) / sd = (59 – 50) / 5 = 9 / 5 = 1.8
Interpretation:
Jemar’s score is 1.8 SD above the mean.

NOTE: THE BIGGER THE Z-SCORE, THE BETTER THE PERFORMANCE.

D.3. T-SCORE
• It tells the location of a score in a normal distribution having a mean of 50
and a standard deviation of 10.
D.3.1. FORMULA
T = 10z + 50
Where z is the z-score. It has a mean of 50 and a standard deviation of 10.

NOTE: THE BIGGER THE T-SCORE, THE BETTER THE PERFORMANCE.

D.4. STANINE
• Also known as standard nine
• These are single digit scores ranging from 1-9.
• The distribution of raw scores is divided into nine parts.

Below Average Average Above Average

Percent in
4% 7% 12% 17% 20% 17% 12% 7% 4%
Stanines
Stanines 1 2 3 4 5 6 7 8 9

D.4.1. FORMULA

Stanine = 1.96z + 5
It has a mean of 5 and a standard deviation of 1.96.

Example:
Bryan has a z-score of 1.9. What is Bryan’s stanine score?
Solution:
Stanine = 1.96 (1.9) + 5 = 8.7 or 9 (Above Average)

• It tells the location of a raw score in a specific segment in a normal

distribution which is divided into 9 segments, numbered from low 1 through
high of 9.
• Scores falling within the boundaries of these segments are assigned ne of
these 9 numbers (standard nine)
 E. MEASURES OF RELATIONSHIP

• These describe the degree of relationship or correlation between the two

variables (example: academic achievement and motivation).
• The relationship between the results of two administration of tests would determine
the reliability of the instrument.
• The greater is the degree of relationship, the more reliable is the test.
±1 Perfect Correlation
±0.75 to ±0.99 Very High Correlation
±0.50 to ±0.74 High Correlation
±0.25 to ±0.49 Moderately Low Correlation
±0.10 to ±0.24 Very Low Correlation
±0.01 to ±0.09 Negligible Correlation
0 No Correlation

E.1. TYPES OF CORRELATION

1. Positive Correlation / Direct Correlation
• Rises from left to right
• As one quantity increases/decreases, the quantity increases/decreases
2. Negative Correlation / Indirect / Inverse
• Falls from left to right
• As one quantity increases/decreases, the quantity decreases/increases
3. No relation / Zero Correlation
• No discernable pattern
E.2. STATISTICAL TOOLS IN MEASURING RELATIONSHIP OF TWO VARIABLES
E.2.1. PEARSON R
• It is the most appropriate measure of correlation when sets of data are of
interval or ratio type.
• It is the most stable measure of correlation.
• It is used when the relationship between the two variables is a linear one.
E.2.2. SPEARMAN-RANK-ORDER CORRELATION OR SPEARMAN RHO
• It is most appropriate measure of correlation when the variable is expressed as
ranks instead of scores or when data represent an ordinal scale.
• Spearman Rho is also interpreted in the same ways as Pearson r.