INFERENTIAL STATISTICS

INFERENTIAL STATISTICS
Inferential statistics enables one to make descriptions of data and draw inferences
and conclusions from the respective data. Through inferential statistics, an individual can
conclude what a population may think or how it’s been affected by taking sample data.
Inferential statistics is mainly used to derive estimates about a large group (or
population) and draw conclusions on the data based on hypotheses testing methods.
Inferential statistics uses sample data because it is more cost-effective and less
tedious than collecting data from an entire population. It allows one to come to reasonable
assumptions about the larger population based on a sample’s characteristics. Sampling
methods need to be unbiased and random for statistical conclusions and inferences to be
validated.

TYPES OF INFERENTIAL STATISTICS

1. t-test (Pooled Estimate)
-comparative study
-2 independent samples
-interval and ratio data
2. t-test (Paired)
-comparative study
-1 set of dependent sample
-interval and ratio data
3. F-test
-comparative study
-3 or more independent samples
-interval and ratio data
4. Chi-square
-comparative/correlational study
-groups vs. Categories
-nominal data
-(one way) 1 group vs. Categories
5. Spearman’s Rho
-correlational data
-1 independent variable vs, 1 dependent variable
-ordinal data
6. Pearson’s r
-correlational data
-1 independent variable vs.1 dependent variable
-interval and ratio data
7. Multiple Linear Regression
-correlational study of F
-2 or more independent variable vs. 1 dependent variable
-interval and ratio data
 t-test (Pooled Estimate)
I. Statement of the problem
Is there a difference between the achievement score in statistics of section A
and section B of DWCC?
Statement of null hypothesis
There is no difference between the achievement score in statistics of section A
and section B of DWCC.
II. Level of significance
Alpha is equal to 5% level of significance
III. Critical value / Tabular value
df = n1 + n2 − 2 tabular value = 2.120
= 8 + 10 − 2
= 16
IV. Computation
A �2 B �2
35 1125 36 1296
38 1444 37 1369
30 900 31 961
36 1296 35 1225
41 1681 42 1764
42 1764 40 1600
39 1521 38 1444
40 1600 41 1681
- - 36 1296
- - 34 1156
Σ=301 Σ=11431 Σ=370 Σ=13792
 301 370
= 8
= 37.62 = 10
= 37

8(11431) − (301)2 91448 − 90601 847

= = = = 15.12
8(8 − ) 56 56
10(13792)−(370)2 137920−136900 1020
= = = = 11.33
90 90 90

37.62−37 .62
= = = 0.36
1(15.12)=9(11.33) 1 1 207.81
( + ) (.225)
16 8 10 16

V. Interpretation
Since the computed t value of 0.36 is less than the tabular value of 2.120 df of
16 at 5% level of significance. The null hypothesis is therefore accepted. Thus, there is
no difference between the achievement score in statistics of section A and section B of
DWCC.
 t-test(Paired)
I. Statement of the problem
Is there a difference between the responses of the students before and
after listening to a seminar?
Statement of null hypothesis
There is no difference between the responses of the students before
and after listening to a seminar
II. Level of significance
Alpha is equal to 5% level of significance
III. Critical value / tabular value
df = n − 1 = 7 − 1 = 6
Tabular value = 2.447
IV. Computation
pre post d �2
27 30 -3 9
28 31 -3 9
25 26 -1 1
26 24 2 4
21 25 -4 16
22 24 -2 4
23 27 -4 16
Σ=172 Σ=187 Σ=-15 Σ=59
 −15
= =− 2.14
7

7(59)−(15)2 413−225
= = = 4.47 = 2.11
7(7−1) 42

−2.14 −2,14 −2.14

= = = =− 2.70
2.11 7 2.11/2.64 0.79

V. Interpretation
Since the computed t value of -2.70 is greater than the tabular value of 2.447 df
of 6 at 5% level of significance. The null hypothesis is therefore rejected. Thus there is
difference between the responses of the students before and after listening to a seminar.
 F-Test
I. Statement of the Problem
Is there a difference between the responses of sections A,B and C regarding the
survey conducted for environmental awareness?
Statement of null hypothesis
There is no difference between the responses of sections A,B and C regarding
the survey conducted for environmental awareness.
II. Level of significance
Alpha is equal to 5% level of significance
III. Critical value / tabular value
df = k − 1, n − k tabular value = 3.354
= 3 − 1, 30 − 3
= 2,27
IV. Computation

ANOVA table
variance SS df MS F-value
treatment 5.12 2 2.56 0.16
error 425.84 27 15.77
total 430.96
 A B C
88 90 92
89 92 81
91 88 90
87 87 93
85 85 94
90 83 85
93 93 86
88 82 88
87 - 93
84 - 83
- - 90
- - 82
t.j = 882 700 1047 T=2629
n.j = 10 8 12 N=30

Σ�2 = 77858 61364 91597 ΣΣ�2 = 230819

V. Interpretation
Since the computed F-value of 0.16 is less than the tabular value of 3.354 at 5%
level of significance, df of 2,27. Therefore we accept null hypothesis. Thus there is no
difference between the responses of sections A,B and C regarding the survey
conducted for environmental awareness.
 Chi-square
I. Statement of the problem
Is there a difference between the responses of roman catholic, Iglesia
Ni Cristo and Muslim regarding divorce with agree, disagree, and no
comment as categories?

Statement of null hypothesis

There is no difference between the responses of roman catholic,
Iglesia Ni Cristo and Muslim regarding divorce with agree, disagree, and
no comment as categories.

II. Level of significance

Alpha at 5% level of significance

III. Critical value / tabular value

df = (R − 1)(C − 1) = (3 − 1)(3 − 1) = 4
tabular value = 9.488

IV. Computation
AGREE DISAGREE NC TOTAL
RC 10 13 18 41
INC 15 9 9 33
MUSLIM 9 20 11 40
TOTAL 34 42 38 114
 AGREE DISAGREE NC TOTAL
RC 12.24 15.10 13.66 41
INC 9.84 12.15 11.01 33
MUSLIM 11.92 14.75 13.33 40
TOTAL 34 42 38 114

AGREE DISAGREE NC TOTAL

RC 0.4099 0.2920 1.3788 2.0807
INC 2.7058 0.8166 0.3669 3.8893
MUSLIM 0.7153 1.8686 0.4072 2.9911
TOTAL 3.831 2.9772 2.1529 8.9611

V. Interpretation
Since the computed chi-square value of 8.9611 is less than the tabular chi
square value of 9.488 df of 4 at 5% level of significance. The null hypothesis is
therefore accepted. Thus, there is no difference between the responses of roman
catholic, Iglesia Ni Cristo and Muslim regarding divorce with agree, disagree, and no
comment as categories.
 Spearman’s rho
I. Statement of the problem
Is there a relationship between the cost and sale of an Automation Company?
Statement of null hypothesis
There is no relationship between the cost and sale of an Automation Company.
II. Level of significance
Alpha is equal to 5% level of significance
III. Critical value / tabular value
df = 10 tabular value = 0.648
IV. Computation
Cost(x) Sale(y) Rx Ry d �2
23500 301250 3 2 1 1
30850 180690 1 9 -8 64
21990 275850 6 5.5 0.5 0.25
23500 262980 3 7 -4 16
16350 180690 8 9 -1 1
21990 301250 6 2 4 16
15998 275850 9 5.5 3.5 12.25
23500 301250 3 2 1 1
21990 299180 6 4 2 4
11875 180690 10 9 1 1
Σd2 = 116.5

6(116.5) 699
=1− =1− = 0.294
10(102 −1) 990

V. Interpretation
Since the computed rs value of 0.294 is less than the tabular value of 0.648 df of
10 at 5% level of significance. Therefore the null hypothesis is accepted. Thus there is
no relationship between the cost and sale of an Automation Company.
 Pearsons r
I. Statement of the problem
Is there a relationship between the cost and sale of ABC company?
Statement of null hypothesis
There is no relationship between the cost and sale of ABC company.
II. Level of significance
Alpha is equal to 5% level of significance
III. Critical value / tabular value
df = n − 2 = 10 − 2 = 8
tabular value = 0.6319
IV. Computation
Cost(x) Sale(y) �2 �2 xy
23 16 529 256 368
16 23 256 529 368
11 15 121 225 165
21 16 441 256 336
15 21 225 441 315
18 30 324 900 540
13 29 169 841 377
9 8 81 64 72
26 9 676 81 234
30 11 900 121 330
Σ=82 Σ=178 Σ=3722 Σ=3714 Σ=3105
 (10)3105−(182)(178)
=
[10(3722)−(182)2 ][10(3714)−(178)2 ]
31050−32396
=
(37220−33124)(37140−31684)

−1346 −1346
= = 4727.34 =− 0.2847
(4096)(5456)

V. Interpretation
Since the computed r value of -0.2847 is less than the tabular value of
0.6319 df of 8 at 5% level of significance. The null hypothesis is therefore
accepted. Thus there is no relationship between the cost and sale of ABC
company.
The negative r value of -0.2847 indicates a inverse relationship
between the variables, which means as the cost decreases the sale increases,
as the cost increases the sale decreases.
The r value of -0.2847 indicates a low relationship between the cost
and sale of ABC company.
 Multiple linear regression
I. Statement of the problem
Is there a relationship between the tv cost and advertising cost with regards the
sale of ABC Company?
Statement of null hypothesis
There is no relationship between the tv cost and advertising cost with regards
the sale of ABC Company.
II. Level of significance
Alpha is at 5% level of significance
III. Critical value / tabular value
df = k − 1, n − k = 2,7 tabular value = 4.737
IV. Computation
1. ��0 + �1 �1 + �2 �2 = �
2. �0 Σ�1 + �1 Σ�21 + �2 Σ�1 �2 = Σ�1 �
3. �0 Σ�2 + �1 Σ�1 �2 + �2 Σ�22 = Σ�2 y

�1 �2 y �21 �22 �2 �1 �2 �1 y �2 �
1 1 3 1 1 9 1 3 3
3 2 6 9 4 36 6 18 12
1 2 5 1 4 25 2 5 10
2 3 6 4 9 36 6 12 18
2 3 6 4 9 36 6 12 18
1 1 4 1 1 16 1 4 4
3 2 7 9 4 49 6 21 14
2 1 4 4 1 16 2 8 4
1 2 4 1 4 16 2 4 8
1 2 5 1 4 25 2 5 10
17 19 50 35 41 264 34 92 101
 50 17 19 50 17
1. 10�0 + 17�1 + 19�2 = 50 1.
92 35 34 92 35
2. 17�0 + 35�1 + 34�2 = 92
101 34 41 101 34
3. 19�0 + 34�1 + 41�2 = 101
10 17 19 10 17

17 35 34 17 35
(71750+58378+59432)−(67165+57800+64124)
1. �0 = (14350+10982+10982)−(12635+11,560+11849) 19 34 41 19 34

189560−189089 471
= = = 1.7444
36314−36044 270

10 50 19 10 50
(37720+32300+32623)−(33212+34340+34850)
2. �1 = 270
2.
17 92 34 17 92
102643−102402 241
= 270
= 270 = 0.8925 19 101 41 19 101

270

3. 10(1.74) + 17(0.89) + 19b0 = 50

17.4 + 15.13 + 19b2 = 50
50 − 17.4 − 15.13 = 19b2
17.47 19�2
19
= 19

�2 = 0.9194
(50)2
= 264 − = 264 − 250 = 14
10
(50)2
= [1.74(50) + 0.89(92) + 0.91(101)] −
10

= 87 + 81.88 + 91.91 − 250 = 10.79

= 14 − 10.79 = 3.21

ANOVA Table
SS df MS F-value = 10.79/14 = 0.7707

treatment 10.79 2 5.395 11.76

error 3.21 7 0.4585 = 0.7707 = 0.8779

total 14
V. Interpretation
Since the computed f-value of 11.76 is greater than the tabular value of 4.737 df
of 2,7 at 5% level of significance. Therefore the null hypothesis is rejected. Thus there
is difference between the tv cost and advertising cost with regards the sale of ABC
Company.
The R2 value or the coefficient of determination of 0.7707 indicates that the tv
cost and advertising cost could affect the sales up to 77.07%, and the remaining
22.93% is for other factors no included in the study.
The computed multiple r of 0.8779 indicates a very high relationship between
the tv cost, advertising cost, and sales
The beta coefficient 1 or b1 is a positive contributor to the sales while the beta
coefficient 2 is also a positive contributor to the sales
The tv cost and advertisement cost is directly proportional to sales. This means
that as the tv and advertising cost increase, the sales increase, vice versa.
.
.