Quantitative Analysis in Evaluation

Descriptive Data Analysis

Measures of Central Tendency
Suppose, senior high school students were asked how many
hours they spent on the computer, and in what subject they
often used it for. Results of the survey could indicate that on
average, the senior high school students spent two (2) or more
hours with a range of one (1) to four (4) hours. A typical senior
high school student spent more than two hours studying
his/her research subject using the computer.
 Measures of Central Tendency: Mean
Mean: arithmetic average of a set of data

𝒔𝒖𝒎 𝒐𝒇 𝒐𝒃𝒔𝒆𝒓𝒗𝒂𝒕𝒊𝒐𝒏𝒔
ഥ) =
𝑴𝒆𝒂𝒏(𝑿
𝒏𝒖𝒎𝒃𝒆𝒓 𝒐𝒇 𝒐𝒃𝒔𝒆𝒓𝒗𝒂𝒕𝒊𝒐𝒏𝒔
The formula is: 𝒏
𝑿𝒊
ഥ=෍
𝑿
𝒏
𝒊=𝟏
or
σ𝒙
ഥ=
𝑿
𝒏
 Measures of Central Tendency: Mean
For Ungrouped Data
Example 1: Find the mean of the measurements 18, 26, 27, 29, and 30.

Example 2: Find the mean of the following:

Scores in the National Achievement Test (NAT)

90 95 96 87 110
102 95 98 87 117
115 96 91 95 95
93 105 86 103 106
 Measures of Central Tendency: Mean
For Grouped Data
When the observations are grouped into classes, the formula for grouped
data is as follows:

𝒇𝒓𝒆𝒒𝒖𝒆𝒏𝒄𝒚 𝒐𝒇 𝒆𝒂𝒄𝒉 𝒄𝒍𝒂𝒔𝒔

ഥ =
𝑴𝒆𝒂𝒏 𝑿 𝒙 𝒄𝒍𝒂𝒔𝒔 𝒎𝒊𝒅𝒑𝒐𝒊𝒏𝒕
𝒕𝒐𝒕𝒂𝒍 𝒏𝒖𝒎𝒃𝒆𝒓 𝒐𝒇 𝒐𝒃𝒔𝒆𝒓𝒗𝒂𝒕𝒊𝒐𝒏𝒔

The Weighted Mean

The formula for Weighted Mean is
σ 𝒇𝒙
ഥ𝑾 =
𝑿
𝒏
Where 𝑓 (frequency), 𝑥 (numerical value or item in a set of data), and n
number of observations.
 Measures of Central Tendency: Mean
For Grouped Data
Example 1: Find the mean of the heights of 50 senior high school
students.
Heights (inches) Frequency Height x Frequency
56 6 336
57 15 855
58 12 696
59 8 472
60 5 300
61 2 122
62 2 124

෍ 𝑓 = 50 ෍ 𝑓𝑥 = 2905
 Measures of Central Tendency: Mean
Solution:

𝒇𝒓𝒆𝒒𝒖𝒆𝒏𝒄𝒚 𝒐𝒇 𝒆𝒂𝒄𝒉 𝒄𝒍𝒂𝒔𝒔

𝑾𝒆𝒊𝒈𝒉𝒕𝒆𝒅 𝑴𝒆𝒂𝒏 = 𝒙 𝒄𝒍𝒂𝒔𝒔 𝒎𝒊𝒅𝒑𝒐𝒊𝒏𝒕
𝒕𝒐𝒕𝒂𝒍 𝒏𝒖𝒎𝒃𝒆𝒓 𝒐𝒇 𝒐𝒃𝒔𝒆𝒓𝒗𝒂𝒕𝒊𝒐𝒏𝒔

σ 𝒇𝒙
ഥ𝑾
𝑿 =
𝒏

𝟐𝟗𝟎𝟓
ഥ𝑾
𝑿 = = 𝟓𝟖. 𝟏 𝒊𝒏𝒄𝒉𝒆𝒔
𝟓𝟎
 Measures of Central Tendency: Mean
For Grouped Data
Example 2: Solve for the mean of the data below.
Class Frequency Class Midpoint (x) fx

76-80 3 78 234
71-75 5 73 365
66-70 6 68 408
61-65 8 63 504
56-60 10 58 580
51-55 7 53 371
46-50 7 48 336
41-45 3 43 129
36-40 1 38 38
 Measures of Central Tendency: Median
Median: midpoint of the distribution (best for ordinal data)

𝑛+1 𝑡ℎ
A. For Ungrouped Data: The median has rank if it is odd;
2
and the median is the average of the two middle values if it is even.

B. For Grouped Data: If the data are grouped into classes, the
𝑛 𝑡ℎ
median will fall into one of the classes as the value.
2
𝑛
2
−𝐹
𝑀𝑒𝑑𝑖𝑎𝑛 = 𝐿 + 𝑖( ).
𝑓
Measures of Central Tendency: Median
Example 1: Consider these numerical values: 12, 15, 18,
22, 30, 32; solve for the median.

Example 2: Consider these numerical values: 15, 20,

12, 26, 3, 30, and 14, and find the median.

Example 3: Find the median for the set of

measurements 7, 8, 8, 9, 9, 10, 23.
 Measures of Central Tendency: Median
Median (For Grouped Data)
𝑛
−𝐹
𝑀𝑒𝑑𝑖𝑎𝑛 = 𝐿 + 𝑖(2 )
𝑓
where
L = exact lower limit of the class containing the median class
i = interval size
n = total number of items or observations
F = cumulative frequency in the class preceding the median
class
f = frequency of the median class
 Measures of Central Tendency: Median
Example 4: The following data show the distribution of the ages of people
interviewed for a survey on climate change.

Class Interval Frequency (𝒇) [Link] (𝑭)

(𝒙)
11-20 20
21-30 14
31-40 22
41-50 18
51-60 14
61-70 12
𝒇 = 𝟏𝟎𝟎
 Measures of Central Tendency: Median
Solution:
Since the number of values or the frequency is 𝑓 = 100,
𝑛 𝑡ℎ
then the median or the median class falls between value,
2
100 𝑡ℎ
or the item, that is the 50𝑡ℎ largest value.
2
Determine in which class the 50𝑡ℎ value falls. The first two
classes have a cumulative frequency of 34 classes.
We need another 16 values to reach 50. Thus, the 50𝑡ℎ value
falls in the next class which contains 22 values. The median
class then is 31-40.
Thus, 𝐿=30.5, 𝑛=100, 𝐹=34, 𝑓=22, and 𝑖=10.
 Measures of Central Tendency: Mode

The Mode: most frequently occurring value in a set of

data
➢Example 1: The ages of 15 persons assembled in a room are
as follows:
16, 18, 18, 18, 25, 25, 25, 25, 30, 30, 30, 34, 34, 36, and 38.

➢Example 2: The number of hours spent by ten (10) students

in an internet café was as follows:
2, 2, 2, 3, 3, 4, 4, 4, 5, and 5.
 Measures of Dispersions
The Range: the difference between a data set’s largest
and smallest values.
➢Example 1: Consider the following scores obtained by ten
(10) students participating in a mathematics contest:
6, 10, 12, 15, 18, 18, 20, 23, 25, and 28.

The Average Mean Deviation: the absolute difference

or deviation between values in a data set and the mean,
divided by the total number of values in a data set.
 The Average Mean Deviation
For Ungrouped Data: The definition-based formula
is:
𝒙−𝒙 ഥ
𝑨𝑫 = ෍
𝒏
➢Example 1: Consider the data set which consists of 20, 25,
35, 40, and 45; calculate the average mean deviation.

➢Example 2: Find the AD for the data set: 22, 60, 75, 85, and
98.
 The Standard Deviation
The Standard Deviation: a measure of the spread or
variation of data about the mean.

❑ For Ungrouped Data: The formula is:

σ 𝒙−𝒙
ഥ 𝟐
𝑺𝑫 =
𝒏−𝟏
 The Standard Deviation
➢Example 1: Let us consider the same data used in the
illustration in the illustration for using the range. The
values are 6, 10, 12, 15, 18, 18, 20, 23, 25, and 28.
Calculate the standard deviation of the given data set.

oSolution: Compute the mean.

 The Standard Deviation
Score (x) ഥ)
(𝒙 − 𝒙 ഥ
𝒙−𝒙 𝟐

6 6-17.5=
10 10-17.5=
12 12-17.5=
15 15-17.5=
18 18-17.5=
18 18-17.5=
20 20-17.5=
23 20-17.5=
25 25-17.5=
28 28-17.5=

ഥ
෍ 𝒙−𝒙 2 =
 The Standard Deviation
Interpretation of the Standard Deviation

1. Approximately 68% of the samples in the sample fall within

one standard deviation of the mean.

2. Approximately 95% of the scores in the sample fall within

two standard deviations of the mean.

3. Approximately 99% of the scores in the sample fall within

three standard deviations of the mean.
 Test of Significance of Difference (T-test)
Between Means
❑ For independent samples (i.e., when the respondents consist of
two different groups as boys and girls, working mothers and non-
working mothers, healthy and malnourished children, and the likes).

❖Case 1: 𝛿1 𝛿2 unknown or 𝒏𝟏 > 𝟑𝟎, and 𝒏𝟐 > 𝟑𝟎.

ഥ𝟏 − 𝒙
𝒙 ഥ𝟐 − 𝝁𝟏 − 𝝁𝟐
𝒛=
𝜹𝟐𝟏 𝜹𝟐𝟐
+
𝒏𝟏 𝒏𝟐
❖Case 2: 𝛿1 ≠ 𝛿2 and 𝑛1 < 30, and 𝑛2 < 30.

ഥ𝟏 −ഥ
𝒙 𝒙𝟐 − 𝝁𝟏 −𝝁𝟐
𝒕= ; (𝑑𝑓=smaller of 𝑛 − 1 or 𝑛 − 2)
𝒔𝟐 𝒔𝟐
𝟏+ 𝟐
𝒏𝟏 𝒏𝟐

❖Case 3: 𝛿1 = 𝛿2 and 𝑛1 < 30, and 𝑛2 < 30.

ഥ𝟏 − 𝒙
𝒙 ഥ 𝟐 − 𝝁𝟏 − 𝝁𝟐
𝒕=
𝒔𝟐𝒑 𝒔𝟐𝒑
+
𝒏𝟏 𝒏𝟐
❖Case 3: 𝛿1 = 𝛿2 and 𝑛1 < 30, and 𝑛2 < 30.

ഥ𝟏 −ഥ
𝒙 𝒙𝟐 − 𝝁𝟏 −𝝁𝟐
𝒕= ; 𝑑𝑓 = 𝑛1 + 𝑛2 − 2
𝒔𝟐
𝒑 𝒔𝟐
𝒑
+
𝒏𝟏 𝒏𝟐

where
𝟐 𝟐
𝟐
𝒏 𝟏 − 𝟏 𝒔𝟏 + 𝒏 𝟐 − 𝟏 𝒔𝟐
𝒔𝒑 =
𝒏𝟏 + 𝒏𝟐 − 𝟐
Assumptions when conducting a Test for 2
Means from Independent Samples
1. We do not know the population standard
deviations, and we do not assume they are equal.
2. The two samples or groups are independent.
3. Both samples are simple random samples
4. Both populations are Normally distributed OR
both samples are large (𝒏𝟏 > 𝟑𝟎 and 𝒏𝟐 > 𝟑𝟎).
Two Means: Independent Samples

Example 1: A survey found that the average

hotel room rate in an upscale area is $88.42 and
the average in a downtown area is $80.61. Each
sample contained 50 hotels. Assume that the
populations’ standard deviations are $5.62 and
$4.83, respectively. At α = 0.05, can it be
concluded that there is a significant difference in
the rates?
➢Solutions:
o Given:
1. Group 1 (Upscale) Step 1: 𝐻0 , 𝐻1 , claim, and Tails.
❑ 𝑥1 = $88.42 Step 2: TS. Calculate TS.
❑ 𝑛1 = 50 Step 3: CV using 𝜶.
❑ 𝜎1 = $5.62 Step 4: Make the decision to
a. Reject or Not the 𝐻0 .
2. Group 2 (Downtown) b. The claim is True or False.
❑ 𝑥ҧ2 = $80.61 c. Restate this decision:
❑ 𝑛2 = 50 “There is/is not sufficient
❑ 𝜎2 = $4.83 evidence to support the
claim that...”
➢Solutions:
o Given:
1. Group 1 (Upscale) Step 1: 𝐻0 , 𝐻1 , claim, and Tails.
❑ 𝑥1 = $88.42 Step 2: TS. Calculate TS.
❑ 𝑛1 = 50 Step 3: CV using 𝜶.
❑ 𝜎1 = $5.62 Step 4: Make the decision to
a. Reject or Not the 𝐻0 .
2. Group 2 (Downtown) b. The claim is True or False.
❑ 𝑥2 = $80.61 c. Restate this decision:
❑ 𝑛2 = 50 “There is/is not sufficient
❑ 𝜎2 = $4.83 evidence to support the
claim that...”
Step 1: 𝐻0 , 𝐻1 , claim, and
Tails.
Step 1:
Step 2: TS. Calculate TS. 𝐻0 : 𝜇1 = 𝜇2
Step 3: CV using 𝜶.
Step 4: Make the decision to 𝐻1 : 𝜇1 ≠ 𝜇2
a. Reject or Not the 𝐻0 .
b. The claim is True or
Claim: The means are not equal.
False. Tails: 2TT
c. Restate this decision:
“There is/is not
Step 2: TS. Calculate TS.
sufficient evidence to 𝑥ҧ1 − 𝑥ҧ2 88.42 − 80.61
support the claim 𝑧= = = 𝟕. 𝟒𝟓
5.62 2 4.83 2
that...”
𝜎12 𝜎22 +
+ 50 50
𝑛1 𝑛2
Step 3: CV using 𝜶.
CV: 𝛼 = 0.05 → 𝒛 = ±𝟏. 𝟗𝟔.
Step 1: 𝐻0 , 𝐻1 , claim, and
Tails.
Step 1:
Step 2: TS. Calculate TS. 𝐻0 : 𝜇1 = 𝜇2
Step 3: CV using 𝜶.
Step 4: Make the decision to 𝐻1 : 𝜇1 ≠ 𝜇2
a. Reject or Not the 𝐻0 .
b. The claim is True or
Claim: The means are not equal.
False. Tails: 2TT
c. Restate this decision:
“There is/is not
Step 2: TS. Calculate TS.
sufficient evidence to 𝑥ҧ1 − 𝑥ҧ2 88.42 − 80.61
support the claim 𝑧= = = 𝟕. 𝟒𝟓
5.62 2 4.83 2
that...”
𝜎12 𝜎22 +
+ 50 50
𝑛1 𝑛2
Step 3: CV using 𝜶.
CV: 𝛼 = 0.05 → 𝒛 = ±𝟏. 𝟗𝟔.
Step 1: 𝐻0 , 𝐻1 , claim, and
Tails. Step 4: Decision
Step 2: TS. Calculate TS.
Step 3: CV using 𝜶.
a. Reject 𝐻0 .
Step 4: Make the decision to
a. Reject or Not the 𝐻0 .
b. The claim is TRUE.
b. The claim is True or c. There is enough
False.
c. Restate this decision: evidence to support the
“There is/is not
sufficient evidence to claim that the means are
support the claim
that...”
not equal. Hence, there
is a significant difference
in the rates.
Step 1: 𝐻0 , 𝐻1 , claim, and
Tails. Step 4: Decision
Step 2: TS. Calculate TS.
Step 3: CV using 𝜶.
a. Reject 𝐻0 .
Step 4: Make the decision to
a. Reject or Not the 𝐻0 .
b. The claim is TRUE.
b. The claim is True or c. There is enough
False.
c. Restate this decision: evidence to support the
“There is/is not
sufficient evidence to claim that the means are
support the claim
that...”
not equal. Hence, there
is a significant difference
in the rates.
Test of Significance of Difference (T-
test)
❑ For correlated/dependent samples (i.e., when the
same set of respondents or paired sets of respondents are
involved)

ഥ 𝒅
𝒅−𝝁
𝒕= 𝒔𝒅 ; (𝑑𝑓 = 𝑛 − 1)
𝒏
 Test of Significance of Difference (T-
test)
Between Proportions or Percentages
❑ For independent samples

𝒑𝟏 − 𝒑𝟐 − 𝒑𝟏 − 𝒑𝟐
𝒛=
𝒑𝒒 𝒑𝒒
+
𝒏 𝟏 𝒏𝟐
❑ For correlated/dependent samples
𝑫−𝑨 𝒑𝟏 −𝒑𝟐
𝒛= or 𝒛 =
𝑨+𝑫 𝒂+𝒅
𝑵
 Test of Significance of Difference (T-
test)
Analysis of Variance (ANOVA): is used when
the significance of the difference of means of two
or more groups is to be determined at one time.
❑ One-Way ANOVA
ANOVA relies on the F-ratio to test the hypothesis that
the two variances are equal; the subgroups are from the
same population. “Between groups” refers to the
variation between each group mean and the grand or
overall mean.
 Test of Significance of Difference (T-
test)
A typical ANOVA table:
Source of Degree of Sum 𝒇 Mean F-ratio p
variation 𝒇 Squares Square
Freedom

Between
groups

Within
groups

Total
 Test of Relationship
Spearman Rank-Order Correlation or Spearman rho.
This is used when data available are expressed in terms of ranks
(ordinal variable).
6 σ 𝐷2
𝜌 =1−
𝑁 𝑁2 − 1
Chi-Square Test for Independence. This is used when data
are expressed in terms of frequencies or percentages (nominal
variables).
𝑶 − 𝑬 𝟐
𝒙𝟐 = ෍ 𝒅𝒇 = 𝒓 − 𝟏 𝒄 − 𝟏
𝑬