INFERENTIAL STATISTICS

(PARAMETRIC DATA)
LEARNING OBJECTIVES
At the end of the course, students should be able to:
1. discuss the characteristics of a parametric data;
2. identify appropriate parametric tests that will answer inferential
questions;
3. illustrate how to calculate t-test for paired sample and independent
sample;
4. illustrate how to calculate one-way ANOVA with Post Hoc test and PPM
using SPSS software; and
5. analyze parametric research data to determine the findings of the study.
INFERENTIAL STATISTICS
 Generally, there are two types of Statistics; Descriptive and
Inferential
 Inferential questions compares between at least two sets of data
 Compares whether the difference or the relationship is
significant or not by using a desired degree of confidence (1 – ά )
 starts with “ Is there a significant difference between . . .? ”
 testing the Null Hypothesis
 works on the concept of Probability of the Events to happen
PARAMETRIC DATA
 Assumes that the data is taken from a normally distribution
population
 A test of normality can be used if required
 Large data are assumed to be normally distributed
 Small samples are normally assumed non-parametric, thus a test is
necessary to declare that the data are parametric
 Samples should be collected using a Probability Sampling
technique
includes t-test for paired sample data, t-test for independent sample
data, one-way ANOVA, PPM
NORMAL DISTRIBUTION
 The distribution approaches normal since normal distribution is ideal
The distribution is symmetrical; 50% of the variable is located above
the mean and 50% below the mean
The total area under the curve is 100% (complete)
The are six Standard Deviation (δ) in a normal distribution
 Empirical Study says that 68.2% is located 1 SD below and 1 SD
above the mean
 99.7% of all the data is located within 6 SDs
NORMAL DISTRIBUTION
(HTTPS://WWW.SPSS-TUTORIALS.COM/NORMAL-DISTRIBUTION/
EXAMPLE 1:
 The graph represents the distribution of all the first year students
of USLS who took the IQ test prior to admission to the university.
The admission committee decided that only 50% of the entire
applicants are to be accepted for enrollment.
 Only those with IQ Score of 100 and above will be accepted to
enroll
 If the are 12,000 applicants who took the IQ test, only 6,000
enrollees are accepted
EXAMPLE 2:
 The graph represents the distribution of all the first year students
of USLS who took the IQ test prior to admission to the university.
The admission committee noticed that 50% of the entire applicants
are too small for the enrollment and decided to all more enrollees.
 At 1 SD below the mean, the enrollees will increase by another
34.1% or a total of 84.1%
 If the are 12,000 applicants who took the IQ test, then 10,092
enrollees will be accepted.
 This means that all those with IQ test scores of 85 and above
will be accepted
SIGNIFICANCE LEVEL ( ά )
 indicates the percentage of sample means that is outside the
critical value (threshold point)
 refers to the probability of rejecting the null hypothesis when it
is true (Type I error)
SIGNIFICANT AT 95% ALPHA
NOT SIGNIFICANT AT 95%
ALPHA
TEST FOR NORMALITY (SPSS)
 Shapiro Wilk
The Shapiro-Wilk Test is more appropriate for small sample sizes
(< 50 samples), but can also handle sample sizes as large as 2000.
 Kolmogorov-Smirnov

If the Sig. value of the Shapiro-Wilk Test is greater than 0.05, the data is
normal. If it is below 0.05, the data significantly deviate from a normal
distribution.
SAMPLE: SHAPIRO WILK TEST
Participant Mean Participant Mean Participant Mean
s s s
1 3.3 12 2.7 23 4.9
2 4.0 13 1.9 24 3.5
3 3.4 14 4.1 25 4.2
4 4.1 15 2.7 26 4.1
5 3.1 16 5.0 27 5.0
6 2.2 17 1.8 28 3.6
7 4.2 18 3.5 29 2.8
8 3.3 19 3.2 30 4.6
9 3.7 20 3.4 31 3.7
10 4.1 21 4.2 32 3.6
11 3.2 22 4.0 33 3.9
SAMPLE: SHAPIRO WILK TEST
STEPS:
1. Encode label (Resiliency)
in the variable view
2. Encode the means in the
data view
3. Press Analyze, the
Descriptive Statistics, the
Explore
 SAMPLE: SHAPIRO WILK TEST

4. Transfer the variable

to the Dependent List
using arrow;
5. Press Plots and
checks Normality plots
with test;
6. Press continue, then
OK.
SAMPLE: SHAPIRO WILK TEST
Notice that Shapiro
Wilk test is
significant at 0.332
which is higher
that 0.05 which
means that the data
despite only 33 is
taken from a
normally
distributed
population. Mean
the data is
PARAMETRIC.
PARAMETRIC TEST
 Descriptive Questions:
What is the level of resiliency among Grade 10 students in this
time of pandemic when group according to section?
 Continuous Data ( either Interval or Ratio )
 Appropriate statistical tool: Mean and Standard Deviation

The above test also provides the Mean = 3.606 and the Standard
Deviation = 0.7968.
PARAMETRIC TEST
 Inferential Questions:
Is there a significant difference on the level of resiliency among
Grade 10 students during this pandemic when group according to
section?
 Hypothesis: There is no significant difference on the level of
resiliency among Grade 10 students during this pandemic between
sections.
Data: Independent Variable (2 groups only)
 Appropriate statistical tool:
T-test for Independent Sample
 T-TEST FOR INDEPENDENT SAMPLES
USING SPSS Assuming that the means from 1 – 22
belongs to Section A and means from 23 – 33
belongs to Section
1. Encode the label in the
variable view (Resiliency)
and another level as Groups.
2. Press Values in the Groups
row; Value Labels will pop-
up.
3. Encode 1 for value and
Section A for label, then press
ADD.
4. Do the same for value 2 and
section B, then press ADD,
then OK.
T-TEST FOR INDEPENDENT
SAMPLES USING SPSS
5. Encode all the means under
the column “Resiliency”
6. Encode 1under the column
“Groups” for means 1 – 22.
7. Encode 2 under the column
“Groups” for means 23 –
33.
8. Press Analyze, then
Compare Means, the
Independent Sample T-
test.
T-TEST FOR INDEPENDENT
SAMPLES USING SPSS
9. A new window will pop-up.
10. Transfer “Resiliency” to
test variable using arrow
and “Groups” to the
grouping variable.
11. Press Define Groups, new
window pop-up.
12. Encode 1 for Group 1 and
encode 2 for Group 2.
13. Press OK
14. Output page pop-up.
T-TEST FOR INDEPENDENT
SAMPLES USING SPSS
RESULTS AND
INTERPRETATION
Table 1
Level of Resiliency when group according to Sections

Section n Mean SD Interpretation

Section A 22 3.414 0.8031 High

Section B 11 3.991 0.6580 High

As a Whole 33 3.606 0.7968 High

Scale
1.0 – 1.80 very low 2.61 – 3.40 moderate
1.81 – 2.60 low 3.41 – 4.20 high
4.21 – 5.00 very high
RESULTS AND
INTERPRETATION
Table 2
Significant Difference when group according to Sections

Section n Mean df t-value Sig-value Interpretation

Section A 22 3.414
31 - 2.059 0.048 Significant at 0.05
Section B 11 3.991 alpha

Significant mean reject the Null Hypothesis;

There is a significant difference on the level of resiliency among Grade 10
students when group according to sections. There is a strong evidence to
prove that students in section B is more resilient than students in section A.
PARAMETRIC TEST
 Inferential Questions:
Is there a significant difference on the level of academic
performance of Grade 10 students in Mathematics before and after
the online review classes?
 Hypothesis: There is no significant difference on the level of
academic performance of Grade 10 students in Mathematics
between before and after the online review classes.
Data: Dependent Variable (2 groups only)
 Appropriate statistical tool:
T-test for Dependent Sample / Paired-sample T-test
EXAMPLE: PAIRED-SAMPLE T-
TEST USING SPSS
Respondents Before After Respondents Before After
Resp 1 23 25 Resp 12 17 20
Resp 2 33 30 Resp 13 20 19
Resp 3 26 27 Resp 14 31 30
Resp 4 19 20 Resp 15 22 25
Resp 5 21 23 Resp 16 32 30

Resp 6 30 27 Resp 17 28 30
Resp 7 25 25 Resp 18 22 20
Resp 8 31 33 Resp 19 18 20
Resp 9 27 29 Resp 20 26 25
Resp 10 22 25 Resp 21 20 22
Resp 11 30 28 Resp 22 17 19
EXAMPLE: PAIRED-SAMPLE T-
TEST USING SPSS
STEPS:
1. Encode Labels in the
variable view “Before”
and “After”.
2. Encode scores in the
data view for before and
after respectively.
3. Press Analyze,
Compare means, and
Paired-Sample T-test.
EXAMPLE: PAIRED-SAMPLE T-
TEST USING SPSS
STEPS:
4. Window pop-up.
5. Transfer both variables
to the Paired Variable
box.
6. You may check options
for the confidence level.
The default value is 95%.
7. Press OK.
EXAMPLE: PAIRED-SAMPLE T-
TEST USING SPSS
RESULTS AND
INTERPRETATION
Table 1
Level of Mathematics Performance Before and After Online Review
Class
Section n Mean SD Interpretation

Section A 22 25.09 4.264 Moderate

Section B 22 24.55 5.143 Moderate

Scale
0 – 10 very low 21 – 30 moderate
11 – 20 low 31 – 40 high
41 – 50 very high
RESULTS AND
INTERPRETATION
Table 2
Significant Difference when group according to Sections

Section n Mean df t-value Sig-value Interpretation

Section A 22 25.09
21 1.254 0.224 Not Significant at
Section B 22 24.55 0.05 alpha

Not Significant means accept the Null Hypothesis;

There is no significant difference on the level of mathematics performance
before and after the online review classes. There is no evidence to prove that
students perform better after an online review classes.
PARAMETRIC TEST
 Inferential Questions:
Is there a significant difference on the level of academic
performance of Grade 10 students in Mathematics when group
according to section?
 Hypothesis: There is no significant difference on the level of
academic performance of Grade 10 students in Mathematics among
sections.
Data: Independent Variable (3 or more groups)
 Appropriate statistical tool:
One-way Analysis of Variance (ANOVA)
 To determine which two are significant: Schiffe Post Hoc Test
SAMPLE: ANALYSIS OF VARIANCE (ANOVA) USING SPSS

Section A Mean Section B Mean Section C Mean

1 3.3 1 2.7 1 4.9
2 4.0 2 1.9 2 3.5
3 3.4 3 4.1 3 4.2
4 4.1 4 2.7 4 4.1
5 3.1 5 3.0 5 5.0
6 2.2 6 1.8 6 3.6
7 4.2 7 3.5 7 2.8
8 3.3 8 3.2 8 4.6
9 3.7 9 3.4 9 3.7
10 4.1 10 4.2 10 3.6
11 3.2 11 3.9
12 4.7 12 4.8
STEPS:
1. Encode Labels in the
variable view
“Resiliency” , adjust
Decimal to 1 and for row
2 “Groupings”.
2. Press Values in the
Group row, pop-up new
window.
3. Enter 1 for value and
“Section A” for label,
ADD. Enter 2, then
Section B, ADD, enter 3
and Section C, and ADD.
4. Press OK.
EXAMPLE: ANALYSIS OF VARIANCE
(ANOVA) USING SPSS

STEPS:
5. Encode all data in the data
view under “Resiliency”.
6. Under Grouping, encode 1
for section A (1-12), encode 2
for section B (13-22), encode
3 for section 3 (23-34).
7. Press Analyze, Compare
Means, One-way ANOVA.
Pop-up new window.
EXAMPLE: ANALYSIS OF VARIANCE
(ANOVA) USING SPSS

STEPS:
8. Transfer “Resiliency” to
the Dependent List using
arrow and “Grouping” to
factor.
9. Press Post Hoc, check
Scheffe among the list.
Significant level at 0.05. Press
Continue.
10. Press Options, and check
Descriptive, press Continue.
11. Press OK.
ANALYSIS OF VARIANCE
(ANOVA) USING SPSS
ANALYSIS OF VARIANCE
(ANOVA) USING SPSS
RESULTS AND
INTERPRETATION
Table 1
Level of Resiliency when group according to Sections

Section n Mean SD Interpretation

Section A 12 3.60 0.6481 High
Section B 10 3.05 0.8100 High
Section C 12 4.06 0.6694 High

As a Whole 34 3.60 0.7977 High

Scale
1.0 – 1.80 very low 2.61 – 3.40 moderate
1.81 – 2.60 low 3.41 – 4.20 high
4.21 – 5.00 very high
RESULTS AND
INTERPRETATION
Table 2
Significant Difference when group according to Sections

Section n Mean df F Sig-value Interpretation

Section A 12 3.60
Significant at 0.05
Section B 10 3.05 33 5.562 0.009 alpha
Section C 12 4.06

Significant means reject the Null Hypothesis;

There is significant difference on the level of resiliency among Grade 10 students
when group according to sections. Post hoc test shows that there is a strong
evidence to prove that students Sections A and C has higher resiliency than students
in section B.
PARAMETRIC TEST
 Inferential Questions: Relationship
Is there a significant relationship on the level of academic
performance of Grade 10 students in Mathematics before and after
the online review classes?
 Hypothesis: There is no significant relationship on the level of
academic performance of Grade 10 students in Mathematics
between before and after the online review classes.
Data: Dependent/ Independent Variable (2 groups only)
 Appropriate statistical tool:
Pearson Product Moment Correlation (PPM)
EXAMPLE: PEARSON PRODUCT
MOMENT (PPM) CORRELATION
USING
Respondents
SPSS
Before After Respondents Before After
Resp 1 23 25 Resp 12 17 20
Resp 2 33 30 Resp 13 20 19
Resp 3 26 27 Resp 14 31 30
Resp 4 19 20 Resp 15 22 25
Resp 5 21 23 Resp 16 32 30

Resp 6 30 27 Resp 17 28 30
Resp 7 25 25 Resp 18 22 20
Resp 8 31 33 Resp 19 18 20
Resp 9 27 29 Resp 20 26 25
Resp 10 22 25 Resp 21 20 22
Resp 11 30 28 Resp 22 17 19
PEARSON PRODUCT MOMENT
(PPM) CORRELATION USING SPSS
STEPS:
1. Encode label in the variable
view “Before” and “After”.
2. Encode scores under Before
and After respectively.
3. Press Analyze, Correlate,
Bivariate.
4. Transfer both Before and After
to the Variable Box. Make sure
Pearson is check.
5. Press Option, check Means and
Standard Deviation, and continue.
6. Press OK.
PEARSON PRODUCT MOMENT
(PPM) CORRELATION USING SPSS
RESULTS AND
INTERPRETATION
Table 2
Significant Relationship on Math Performance Before and After Online Review
Classes
Section n Mean SD r Sig-value Interpretation

Before 22 24.55 5.143

0.923 0.000 Significant at 0.05
After 22 25.09 4.264 alpha

Significant means reject the Null Hypothesis;

There is a significant relationship on the level of mathematics performance
before and after the online review classes. There is very strong evidence to
prove that students perform better after an online review classes.
THANK YOU
FOR LISTENING