Statistical Analysis of Data for Quantitative Research

Weighted Mean

● It is a form of specialized mean that is widely used in most of every quantitative

research. It values the weight of each given scale based on a particular research
statement.

Research Statements 5 4 3 2 1
Strongly Agree Neither Disagree Strongly
Agree Disagree
1. I can easily learn
mathematical concepts if
the teacher is providing
necessary examples for 12 3
the Math lessons.
2. I can excel in any Math
subject if our teacher can
instruct the concepts in a
very simplified way like 7 5 3
in online learning videos
in Math.
3. I think an innovative
teacher is the one who
always seek for
discovery and
improvement in assuring
that his/her students 12 3
learn.

4. Math is an interesting 11 4
subject.

n=15

Formula for Weighted Mean:

Weighted Mean:
X = ∑fx
n
∑= summation symbol
f= frequency for each given scale
x= value of each scale
n= number of frequency

Statistical Data Analysis 1

Solution:

1. X = ∑fx
n

X = ∑ 12 (5) + 3 (4)

15

X = 60 + 12
15

X = 72
15

X = 4. 80≈ 5 – Strongly Agree

2. X = ∑fx

X = ∑ 7 (5) + 5 (4) + 3 (3)

15

X = 35 + 20 + 9
15

X = 64
15

X = 4. 27≈ 4 - Agree

3. X = ∑fx

X = ∑ 12 (5) + 3 (4)

15

X = 60 + 12
15

X = 72
15

Statistical Data Analysis 2

X = 4. 80≈ 5 – Strongly Agree

4. X = ∑fx

X = ∑ 11(5) + 4 (4)

15

X = 55 + 16
15

X = 71
15

X = 4. 73≈ 5 – Strongly Agree

For the Overall Weighted Mean

(For each Category)

X = ∑fx

Total number of Questions in a Category

X = ∑ 4.80+ 4.27 +4.80 + 4.73

X = 18.60
4

X ( Overall ) = 4. 65 ≈ 5 – Strongly Agree

Statistical Data Analysis 3

Parametric Test for Known Samples ( Z - test)

Z – test for Known Sample

Z – test is used to determine the amount of effect of an Independent Variable to the
Dependent Variable.

This is one of the most fundamental statistical treatments in a quantitative research

specifically in descriptive research.

A hypothesis is an assumption or a proposed statement regarding a possible answer to a

research problem or question. To test the hypothesis, a research must be converted into a
statistical hypothesis.

Steps in Hypothesis Testing

1. State the null and alternative hypothesis

There are two types of hypothesis:

(Null) H0: µ =3.00 (indicating neutrality, not positive and not negative)
(Alternative)
H1: : µ > 3.00 (Right –Tailed Test) (higher than null hypothesis)
(indicating a significant effect)

2. Set a level of significance: 0.05

( indicating 95% success of research and 5% margin of error)
3. Compute for the z – value

z=x -µ

δ
√n

Where:

z= value of test statistics

x = sample mean
µ= population mean
δ= standard deviation
𝑛 = number of frequencies
4. Compare the z – test value to the table of values, to get the z- value, afterwards subtract it
from the critical value to get the p- value.
5. If p- value is < level of significance, then reject H0.
If p- value is > level of significance, then fail to reject H0.
6. Make a decision.

Statistical Data Analysis 4

Sample Problem:

The researchers assume that social media has no significant impact to students ‘behavioral
performance that is why they set 3.00 as a population mean.

Class Frequency Class Mark fx x-x

Interval (f) (x) (x-x)2 f(x-x)2
5 8 5 40 0.67 0.45 3.60
4 5 4 20 -0.33 0.11 0.55
3 1 3 3 -1.33 1.77 1.77
2 1 2 2 -2.33 5.43 5.43
1 0 1 0 -3.33 11.09 0
n= 15 ∑fx= 65 ∑(x-x )2 ∑f(x-x) =11.35
2

=18.85

x = 65 s2= ∑f(x-x)2 s2= 11.35 s2= 11.35 s2=0.81 √s2=√0.81 δ =0.90

15 n -1 15 -1 14

x = 4.33- sample mean based on the computation in the table

Hypothesis Testing for z – test

1.H0: µ =3.00 ( constant value for results based on rating scale)

H1: : µ >3.00

2. Level of Significance: 0.05

3. Compute for the test statistics

z=x -µ

δ (standard deviation)
√n

z = 4.33 - 3.00

0. 90
√ 15

Statistical Data Analysis 5

 z = 1.33

0. 90

3.87

z = 1.33

0.23

z= 5.78 ( test statistic result)

4. Look at the table of values for test statistics, and the z- value is 0.9999, subtract it from
the critical value which is 0.5.
p – value = critical value – z value

p - value = 0.5 – 0.9999

p – value = - 0.4999

5. Compare p –value with level of significance

-0.4999 <0.05, then reject H0

6. .: Reject H0
Decision : Social Media has a significant effect on students’ behavioral
performance.

Statistical Data Analysis 6

Statistical Table for Z - Value

http://www.z-table.com/

(If the computed value is more than 3.40 , the z – value will be 0.9999, exceeding the values on the table )

Statistical Data Analysis 7

Pearson –r

● It is used to indicate how strong the linear relationship of variables that are indicated in a
research.
● Formula:
r= n(∑xy) – (∑x) (∑y)
√[n(∑x )-(∑x)2] [n( ∑y2) – (∑y)2]
2

Where:
n= number of cases
∑x= sum of x-values
∑y=sum of y – values
∑xy= sum of x and y values
∑x2= sum of squared x – values
∑y2= sum of squared y – values

Given:
Here is a table that shows the scores of 10 selected students who have taken a 25 –
item test in Problem Solving and a 25 item test in English Proficiency. Find the value of
Pearson -r

x y xy x2 y2
20 19 380 400 361
21 22 462 441 484
22 23 506 484 529
19 21 399 361 441
17 19 323 289 361
16 17 272 256 289
16 20 320 256 400
23 24 552 529 576
20 18 360 400 324
18 20 360 324 400
2 2
∑x=192 ∑y=203 ∑xy= 3,934 ∑x =3,740 ∑y =4,165

Solution:
r= n(∑xy) – (∑x) (∑y)
√[n(∑x2)-(∑x)2] [n( ∑y2) – (∑y)2]
r= 10(3,934) – (192) (203)
√[10(3,740)-(192)2] [10(4165) – (203)2]
r= 39,340 – 38,976
√[37,400 – 36,864] [41,650–41,209 ]
r= 364
√[536] [441]

r= 364
√236,376

Statistical Data Analysis 8

 r= 364

486.19

r= 0.75 (strong positive correlation)

Pearson – r Strength of Correlation

Value
+1 Perfect Positive Correlation
+ 0.71 to + 0.99 Strong Positive Correlation
+ 0.51 to + 0.70 Moderately Positive Correlation
+ 0.31 to +0.50 Weak Positive Correlation
+ 0.01 to + 0.30 Negligible Positive Correlation
0 No Correlation
-0.01 to -0.30 Negligible Negative Correlation
-0.31 to -0.50 Weak Negative Correlation
-0.51 to – 0.70 Moderately Negative Correlation
-0.71to -0.99 Strong Negative Correlation
-1 Perfect Negative Correlation

Statistical Data Analysis 9

Chi – Square Test

It is used to find out if a particular set of data fits a claimed distribution. It shows the
comparison of observed frequencies and expected frequencies. It is used for nominal and ordinal data
set. It shows the significant differences of the categories of variables and their level of association. It
can be about the goodness of fit or the test of independence.

Formula:

x2= (O-E)2

where:

x2= chi – square values

O = observed values
E= expected values

Here is a table that shows the number of male and female respondents who have answered a particular
survey question.
Survey Question:

Social Media Usage affects one’s behavior and learning concentration.

Respondents 5 4 3 2 1 Total
Male 11 5 1 1 2 20
Female 10 4 2 1 3 20
Total 21 9 3 2 5 40

Statistical Data Analysis 10

To calculate for the Expected Value, (E),

(Total Value of the Scale for Each Column ) * (Total Value of the Category for Each Row)

Total Number of Respondents

Steps in calculating the Chi – Square Values

1. Construct a table values for each scale.

2. Calculate the summation of the nominal variables for each scale.
3. Calculate the grand total of the values based on the number of respondents.
4. Construct a chi –square table for each scale.
5. For each scale, calculate the observed values based on the indicated frequencies.
6. For each scale, calculate the expected values.
7. For each scale, calculate the difference of the observed values and expected values.
8. For each scale, calculate the squared difference of the observed values and expected values.
9. For each scale, calculate the quotient between (O- E)2 and E.
10. Solve for the summation of all values, to determine the chi – square value

5- Strongly Agree
Respondents O E O-E (O-E)2 (O-E)2/E
Male 11 10.5 0.5 0.25 0.02
Female 10 10.5 -0.5 0.25 0.02
4- Agree
Respondents O E O-E (O-E)2 (O-E)2/E
Male 5 4.5 0.5 0.25 0.06
Female 4 4.5 -0.5 0.25 0.06

3- Neutral
Respondents O E O-E (O-E)2 (O-E)2/E
Male 1 1.5 -0.5 0.25 0.17
Female 2 1.5 0.5 0.25 0.17

2- Disagree
Respondents O E O-E (O-E)2 (O-E)2/E
Male 1 1 0 0 0
Female 1 1 0 0 0
1- Strongly Disagree
Respondents O E O-E (O-E)2 (O-E)2/E
Male 2 2.5 -0.5 0.25 0.1
Female 3 2.5 0.5 0.25 0.1
∑x = 0.02+0.02+0.06+0.06+0.17+0.17+0.1+0.1 = 0.7
2

Statistical Data Analysis 11