CARLOS HILADO MEMORIAL STATE UNIVERSITY

Talisay City, Negros Occidental

COLLEGE OF EDUCATION

END-TERM EXAMINATION IN MFC 502-EDUCATIONAL STATISTICS

Advanced Education Programs
First Semester, S.Y. 2024-2025

Name: JOY MARK L. NERVEZ Score: _________

Course/Yr./Section: MAED ENG 1-C

I. Discuss the following:

1. What is hypothesis testing? How is it important in real life?
Hypothesis testing is a fundamental concept and a very important step in statistics and research. It
is a way we can formally test our hypothesis whether it needs to be rejected or accepted based on
the data gathered.
2. Differentiate parametric tests and nonparametric tests.
Parametric tests assume data is normally distributed, use interval or ratio level measurements,
require random sampling, assess group means, and are generally used with larger sample sizes
while non-parametric tests are used when data is not normally distributed (skewed), use nominal
or ordinal level measurements, may involve non-random sampling, assess group medians, and are
often employed with smaller sample sizes.
3. Enumerate the parametric tests and their alternative methods (counterparts in nonparametric
tests) and briefly discuss each of their uses.
Parametric Test Non-Parametric Test
1. One Sample T-Test Wilcoxon Signed Rank Test
The One Sample t-Test is a parametric The Wilcoxon Signed Rank Test is a
procedure that tests whether a sample mean nonparametric counterpart of the one-
is statistically significantly different than a sample t-test. It compares your sample
population mean or other known value. median against a hypothetical median.

2. Student’s Independent Samples T-Test Mann-Whitney U test

The Student’s Independent samples t-test is The Mann-Whitney U test, a non-
used to test the null hypothesis that two parametric counterpart to the independent
groups have the same mean. It also assumes samples t-test, facilitates comparison
that the data from each group are from a between two independent groups. It is
normal distribution, and that the variances of particularly valuable when analyzing ordinal
these groups are equal. data or continuous data that violates the
normality assumptions required for the t-
test.
3. Student’s Paired Samples T-Test Wilcoxon Matched-Pairs Signed Rank Test
The paired samples t-test checks if the average The Wilcoxon matched-pairs signed rank
difference between paired measurements is test computes the difference between each
zero. A low p-value suggests this difference is set of matched pairs.
statistically significant. This indicates the null
hypothesis (no difference) is likely false.

pg. 1
 4. One-way ANOVA (Analysis of Variance) Kruskal-Wallis Test
The One-way ANOVA (Analysis of Variance) is a The Kruskal-Wallis Test is a nonparametric
statistical test that determines if the means of alternative to the one-way ANOVA. It is
three or more independent groups are used to compare more than two
significantly different. 1 It examines the independent groups with ordinal data. It is
influence of a single categorical variable (with used to determine whether or not there is a
three or more levels) on a continuous outcome statistically significant difference between
variable. the medians of three or more independent
groups.

5. Pearson Correlation Coefficient (Pearson’s r) Spearman rank correlation coefficient

Pearson's r measures the strength and (Spearman’s Rho)
direction of a linear relationship between two Spearman’s Rho is used to understand the
continuous variables. It quantifies how well a strength of the relationship between two
straight line fits the data points in a variables. Your variables of interest can be
scatterplot. A correlation coefficient of +1 continuous or ordinal and should have a
indicates a perfect positive linear relationship, monotonic relationship.
-1 a perfect negative linear relationship, and 0
no linear relationship. Pearson's r is typically
used when the data is normally distributed.

II. A. Use the ATTACHED FILE “DATA FOR ENDTERM EXAM”, and answer each of the following
problems using the 5-step solution in hypothesis testing.

1. Is there a significant difference in the students' problem-solving skills before the interventions
when they are grouped according to the school they graduated from?
Step 1: Ho: There no significant difference in the students' problem-solving skills before the
interventions when grouped according to the school they graduated from.
Ha: There is significant difference in the students' problem-solving skills before the
interventions when grouped according to the school they graduated from.
Step 2: Level of significance: α = 0.05
Step 3: Normality test: Private school p-value = 0.192
Public school p-value = 0.791
Distribution of data is normal
Descriptive Statistics
PRETEST
PRIVATE PUBLIC

Shapiro-Wilk 0.881 0.959

P-value of Shapiro-Wilk 0.192 0.791

Note. Excluded 1 rows from the analysis that correspond to the missing values of the split-by variable SCHOOL THEY
GRADUATED FROM

Type of Test: Parametric Test

Statistical Tool: Independent Samples T-Test
Decision rule: If the obtained p-value is < 0.05, reject Ho. Otherwise, do not reject Ho.

pg. 2
Step 4: Computation: (Insert JASP table result)
Independent Samples T-Test
t Df p

1
PRETEST -0.823 0.423
5

Note. Student's t-test.

Step 5: Decision: The obtained p-value 0.423 is more than the a = 0.05, therefore we do not reject
the null hypothesis.
Conclusion: There no significant difference in the students' problem-solving skills before the
interventions when grouped according to the school they graduated from.

2. Is there a significant difference in the students' problem-solving skills after the first intervention
when they are grouped according to the school they graduated from?
Step 1: Ho: There is no significant difference in the students' problem-solving skills after the first
intervention when grouped according to the school they graduated from.
Ha: There is a significant difference in the students' problem-solving skills after the first
intervention when grouped according to the school they graduated from.

Step 2: Level of significance: α = 0.05

Step 3: Normality test: p-value is < 0.05
Private school p-value = 0.007
Public School p-value = 0.011
Data distribution is not normal.
Descriptive Statistics
POSTTEST (INTERVENTION 1)
PRIVATE PUBLIC

Mean 88.625 91.222

Std. Deviation 3.701 2.906
Shapiro-Wilk 0.743 0.778
P-value of Shapiro-Wilk 0.007 0.011
Minimum 80.000 89.000
Maximum 92.000 97.000

Note. Excluded 1 rows from the analysis that correspond to the missing values of the split-by variable SCHOOL THEY
GRADUATED FROM

Type of Test: Non-Parametric Test

Statistical Tool: Mann- Whitney U Test
Decision rule: If the obtained p-value is < 0.05, reject Ho. Otherwise, do not reject Ho.
Step 4: Computation: (Insert JASP table result)
Independent Samples T-Test
U df p

24.50
POSTTEST (INTERVENTION 1) 0.277
0

Note. Mann-Whitney U test.

Step 5: Decision: The obtained p-value 0.277 is more than the a = 0.05, therefore we do not reject
the null hypothesis.

pg. 3
 Conclusion: There is no significant difference in the students' problem-solving skills after the
first intervention when grouped according to the school they graduated from.

3. Is there a significant difference in the students' problem-solving skills after the second
intervention when they are grouped according to the school they graduated from?
Step 1: Ho: There is no significant difference in the students' problem-solving skills after the second
intervention when they are grouped according to the school they graduated from.
Ha: There is a significant difference in the students' problem-solving skills after the second
intervention when they are grouped according to the school they graduated from.
Step 2: Level of significance: α = 0.05
Step 3: Normality test: p-value is < 0.05
Private school p-value = 0.007
Public School p-value = 0.024

Descriptive Statistics
POSTTEST (INTERVENTION 2)
PRIVATE PUBLIC

Shapiro-Wilk 0.744 0.806

P-value of Shapiro-
0.007 0.024
Wilk

Type of Test: Non-Parametric

Statistical Tool: Mann-Whitney U Test
Decision rule: If the obtained p-value is < 0.05, reject Ho. Otherwise, do not reject Ho.

Step 4: Computation:
Independent Samples T-Test
U df p

24.50
POSTTEST (INTERVENTION 2) 0.286
0

Note. Mann-Whitney U test.

Step 5: Decision: The obtained p-value 0.286 is more than the a = 0.05, therefore we do not reject
the null hypothesis.
Conclusion: There is no significant difference in the students' problem-solving skills after the
second intervention when they are grouped according to the school they graduated from.

4. Do the problem-solving skills of the students after the first intervention significantly differ from
their problem-solving skills after the second intervention?
Step 1: Ho: The problem-solving skills of the students after the first intervention do not significantly
differ from their problem-solving skills after the second intervention.
Ha: The problem-solving skills of the students after the first intervention do significantly differ
from their problem-solving skills after the second intervention.
Step 2: Level of significance: α = 0.05
Step 3: Normality test: Post Test p-value= 0.005

pg. 4
Descriptive Statistics
PRETEST POSTTEST (INTERVENTION 1)

Variance 35.765 12.000

Shapiro-Wilk 0.945 0.825
P-value of Shapiro-
0.384 0.005
Wilk

Type of Test: Non-parametric Test

Statistical Tool: Wilcoxon Signed Rank Test
Decision rule: If the obtained p-value is < 0.05, reject Ho. Otherwise, do not reject Ho.
Step 4: Computation:
Paired Samples T-Test
Measure
Measure 2 W z df p
1

18.00
PRETEST - POSTTEST (INTERVENTION 1) -2.385 0.018
0

Note. Wilcoxon signed-rank test.

Step 5: Decision: The obtained p-value 0.018 is less than the a = 0.05, therefore we reject the null
hypothesis.
Conclusion: The problem-solving skills of the students after the first intervention do
significantly differ from their problem-solving skills after the second intervention.

5. Is there a significant difference in the problem-solving skills of the students after the first
intervention when they are grouped according to socio-economic status?
Step 1: Ho: There is no significant difference in the problem-solving skills of the students after the
first intervention when grouped according to socio-economic status.
Ha: There is a significant difference in the problem-solving skills of the students after the first
intervention when grouped according to socio-economic status.
Step 2: Level of significance: α = 0.05
Step 3: Normality test: p-value (average) = 0.146
p-value (high) = 0.091
p-value (low) = 0.678
Distribution of Data is normal.

Descriptive Statistics
POSTTEST (INTERVENTION 1)
AVERAGE HIGH LOW

Shapiro-Wilk 0.833 0.822 0.942

P-value of Shapiro-Wilk 0.146 0.091 0.678

Note. Excluded 1 rows from the analysis that correspond to the missing values of the split-by variable
SES

Type of Test: Parametric Test

Statistical Tool: One-Way ANOVA
Decision rule: If the obtained p-value is < 0.05, reject Ho. Otherwise, do not reject Ho.

pg. 5
Step 4: Computation:
ANOVA - POSTTEST (INTERVENTION 1)
Mean
Cases Sum of Squares df F p
Square

SES 1.333 2 0.667 0.049 0.952

Residual
190.667 14 13.619
s

Note. Type III Sum of Squares

Step 5: Decision: The obtained p-value 0.952 is more than the a = 0.05, therefore we do not reject
the null hypothesis.
Conclusion: There is no significant difference in the problem-solving skills of the students
after the first intervention when grouped according to socio-economic status.

6. Is there a significant difference in the students' problem-solving skills after the second
intervention when they are grouped according to socio-economic status?
Step 1: Ho: There is no significant difference in the students' problem-solving skills after the second
intervention when they are grouped according to socio-economic status.
Ha: There is a significant difference in the students' problem-solving skills after the second
intervention when they are grouped according to socio-economic status.
Step 2: Level of significance: α = 0.05
Step 3: Normality test: p-value (average) = 0.656
p-value (high) = 0.167
p-value (low) = 0.743
Distribution of Data is normal.

Descriptive Statistics
POSTTEST (INTERVENTION 2)
AVERAGE HIGH LOW

Shapiro-Wilk 0.939 0.853 0.950

P-value of Shapiro-Wilk 0.656 0.167 0.743

Note. Excluded 1 rows from the analysis that correspond to the missing values of the split-by variable
SES

Type of Test: Parametric

Statistical Tool: One-Way ANOVA
Decision rule: If the obtained p-value is 0.05, reject Ho. Otherwise, do not reject Ho.
Step 4: Computation:
ANOVA - POSTTEST (INTERVENTION 2)
Mean
Cases Sum of Squares df F p
Square

SES 4.249 2 2.125 0.103 0.903

Residual
289.633 14 20.688
s

pg. 6
ANOVA - POSTTEST (INTERVENTION 2)
Mean
Cases Sum of Squares df F p
Square
Note. Type III Sum of Squares

Step 5: Decision: The obtained p-value 0.903 is more than the a = 0.05, therefore we do not reject
the null hypothesis.
Conclusion: There is no significant difference in the students' problem-solving skills after the
second intervention when they are grouped according to socio-economic status.

7. Is there a significant difference in the students' problem-solving skills before and after the first
Intervention?
Step 1: Ho: There is no significant difference in the students' problem-solving skills before and after
the first Intervention
Ha: There is a significant difference in the students' problem-solving skills before and after
the first Intervention.
Step 2: Level of significance: α = 0.05
Step 3: Normality test: p-value (Post Test) = 0.005
p-value is < 0.05
Distribution of data is not normal.
Descriptive Statistics
PRETEST POSTTEST (INTERVENTION 1)

Shapiro-Wilk 0.945 0.825

P-value of Shapiro-
0.384 0.005
Wilk
Minimum 75.000 80.000
Maximum 96.000 97.000

Type of Test: Non-Parametric Test

Statistical Tool: Wilcoxon-Signed Rank Test
Decision rule: If the obtained p-value is 0.05, reject Ho. Otherwise, do not reject Ho.
Step 4: Computation:
Paired Samples T-Test
Measure
Measure 2 W z df p
1

18.00
PRETEST - POSTTEST (INTERVENTION 1) -2.385 0.018
0

Note. Wilcoxon signed-rank test.

Step 5: Decision: The obtained p-value 0.018 is less than the a = 0.05, therefore we reject the null
hypothesis.
Conclusion: There is a significant difference in the students' problem-solving skills before
and after the first Intervention.

8. Is there a significant difference in the students' problem-solving skills before and after the second
Intervention?
Step 1: Ho: There is no significant difference in the students' problem-solving skills before and after
the second Intervention?
Ha: There is a significant difference in the students' problem-solving skills before and after
the second Intervention.

pg. 7
Step 2: Level of significance: α = 0.05
Step 3: Normality test: p-value of pretest = 0.384
p-value (Post Test 2) = 0.043
p-value (Post Test 2) is less than 0.05
Distribution of Data is not normal

Descriptive Statistics
POSTTEST (INTERVENTION 2) PRETEST

Shapiro-Wilk 0.888 0.945

P-value of Shapiro-
0.043 0.384
Wilk

Type of Test: Non-Parametric Test

Statistical Tool: Wilcoxon-Signed Rank Test
Decision rule: If the obtained p-value is 0.05, reject Ho. Otherwise, do not reject Ho.

Step 4: Computation:
Paired Samples T-Test
Measure
Measure 2 W z df p
1

17.00
PRETEST - POSTTEST (INTERVENTION 2) -2.442 0.015
0

Note. Wilcoxon signed-rank test.

Step 5: Decision: The obtained p-value 0.015 is less than the a = 0.05, therefore we reject the null
hypothesis.
Conclusion: There is a significant difference in the students' problem-solving skills before
and after the first Intervention.

9. Is reading comprehension significantly related to the student's problem-solving skills in posttest

2?
Step 1: Ho: Reading comprehension is not significantly related to the student's problem-solving
skills in posttest 2
Ha: Reading comprehension is significantly related to the student's problem-solving skills in
posttest 2

Step 2: Level of significance: α = 0.05

Step 3: Normality test: p-value for Posttest is 0.043 is less than 0.05.
The Distribution of Data is not normal
Descriptive Statistics
READING COMPREHENSION POSTTEST (INTERVENTION 2)

Shapiro-Wilk 0.957 0.888

P-value of Shapiro-
0.583 0.043
Wilk

Type of Test: Non-parametric test

Statistical Tool: Spearman’s Rho
Decision rule: If the obtained p-value is 0.05, reject Ho. Otherwise, do not reject Ho.
Step 4: Computation:

pg. 8
Spearman's Correlations
Spearman's rho p

POSTTEST (INTERVENTION 2) - READING COMPREHENSION 0.870*** < .001

* p < .05, ** p < .01, *** p < .001

Assumption checks
Shapiro-Wilk Test for Bivariate Normality
Shapiro-Wilk p

POSTTEST (INTERVENTION 2) - READING COMPREHENSION 0.907 0.045

Step 5: Decision: The obtained p-value 0.001 is less than the a = 0.05, therefore we reject the null
hypothesis.
Conclusion: Reading comprehension is significantly related to the student's problem-solving
skills in posttest 2.

B. Consider the problem below and answer the following questions:

If “reading comprehension” (x) and “problem-solving skills of the students in the posttest 2” (y) are
significantly related…
Model Summary - POSTTEST (INTERVENTION 2)

RM
Model R R² Adjusted R²
SE

M₀ 0.000 0.000 0.000 4.286

M₁ 0.824 0.679 0.657 2.509

Note. M₁ includes READING COMPREHENSION

Coefficients
Model Unstandardized Standard Error Standardized t p

M₀ (Intercept) 89.647 1.039 86.245 < .001

M₁ (Intercept) 53.712 6.414 8.374 < .001

READING COMPREHENSION 0.450 0.080 0.824 5.628 < .001

1. What will be the regression equation for the two variables, x and y?
y = 0.450x + 53.712
2. Determine the standard error of prediction. RMSE = 2.509
3. Predict the POSTTEST 2 result of a student with a reading comprehension of 78.
y = 0.450 x 78 + 53.712
y = 35.1 + 53.712

pg. 9
y = 88.812
4. Predict the POSTTEST 2 result of a student with a reading comprehension of 83.
y = 0.450 * 83 + 53.712
y = 37.35 + 53.712
y = 91.062
5. Predict the POSTTEST 2 result of a student with a reading comprehension of 94.
y = 0.450 * 94 + 53.712
y = 42.3 + 53.712
y = 96.012

GOD BLESS 

Prepared by:

Novemay C. Makilan, PhD

pg. 10