EFD 502 Advanced Statistics with Software Application

Exam

Name: ARIATE, JOVILOU V.

Program: MAEd - LT

General Instruction: Do as indicated. Copy the questions for each item followed by your
answers. Attach only the tables you need from the SPSS to support your claims, excluding
unnecessary ones. Arrange your tables and explanations as orderly and neatly as possible. No
two papers must have the exact same answers. If identical answers are found, the points
will be divided among those who shared the same answers . Submit the pdf file of your
exam in the Google Classroom with the filename in the following format: EXAM - SURNAME,
FIRST NAME, MI. Ensure submission on or before December 29, 2024 (Sunday).

Please note: Display all the required tables and graphs first before you make/enumerate your
answers as a support to your claims.
1. (15 points) The data set consists of mathematics scores ranging from 1 to 100. Determine
whether the average mathematics score of a sample of 1000 students. Based on the
historical information, the hypothesized average score of mathematics is 75. Determine
whether the average math score is significantly different from the hypothesized score.
a. Display the essential descriptive statistics. Describe the mean, median, standard
deviation and IQR.

Descriptive Statistics
math_score

Median 66.000
Mean 66.089
Std. Deviation 15.163
IQR 20.000

 The average mathematics score of 1,000 students is 66.089.

 A median of 66.000 means that half of the students scored below 66, and
half scored above 66.
 The mean is slightly higher than the median. The result implies that there
are fewer higher scores pulling the average up.
 The standard deviation implies that the mathematics scores of the students
deviate from the mean by 15.163 points. The scores are diverse and spread
out over a wide range.
 The IQR indicates that the middle 50% of the students' scores fall within a
range of 20 points. It reveals a significant spread of the scores and
variation of scores.
EFD 502 Advanced Statistics with Software Application
Exam

b. Display the graphs (histogram, Q-Q plot, and boxplot) of the data. What can be
said about the data using these graphs? Enumerate your observations.

HISTOGRAM

 The mathematics scores between 60 and 70 have the highest frequency,

meaning the largest number of students' population scored within this
range. In contrast, the scores between 20 and 30 have the smallest
population, indicating few students scored in this range. Additionally,
there are outliers, with a few extremely low scores from a small portion of
the population.
 The histogram shows a left-skewed (negative) distribution, even though
the mean is slightly higher than the median. This is due to the presence of
extreme low outliers which the distribution of the data.

BOXPLOT


EFD 502 Advanced Statistics with Software Application
Exam

c. Compute the Shapiro-Wilk Test for Normality of the data. What can be said about
the data? Discuss your answer.

Descriptive Statistics
math_score

Valid 1000
Missing 0
Median 66.000
Mean 66.089
Std. Deviation 15.163
IQR 20.000
Shapiro-Wilk 0.993
P-value of Shapiro-
< .001
Wilk
Minimum 0.000
Maximum 100.000

d. Using your results in b and c, if you are to decide a statistical test, what type of
test (parametric or nonparametric) and what test will you choose to analyze the
data?
Non-parametric test

e. Implement the data in JASP and discuss the result.

EFD 502 Advanced Statistics with Software Application
Exam

One Sample T-Test

V p

math_score 96795.500 < .001

Note. For the Wilcoxon test, the alternative hypothesis specifies that the median is different from
75.
Note. Wilcoxon signed-rank test.

2. (15 points) Literatures say that there is a relationship between reading and math skills.
Strong reading skills can enhance understanding of math problems, especially word
problems. Students with better reading skills often perform better in math because they
comprehend the questions more effectively. On the other hand, math skills can also
support reading developments. Math promotes logical thinking and problem-solving
skills which are beneficial in reading comprehension and critical analysis. Determine
whether there is a significant difference between the math and reading scores of students.
a. Display the descriptive statistics for each data and describe mean, median,
standard deviations and IQR of both data.

Descriptive Statistics
reading_score math_score

Valid 1000 1000

Missing 0 0
Median 70.000 47.000
Mean 69.169 47.145
Std.
14.600 14.981
Deviation
IQR 20.000 20.000
Minimum 17.000 1.000
Maximum 100.000 81.000

b. Display the graphs (histogram, Q-Q plot, and boxplot) of the data. What can be
said about the data using these graphs? Enumerate your observations.
EFD 502 Advanced Statistics with Software Application
Exam

c. Compute the Shapiro-Wilk Test for Normality of the data. What can be said about
the data? Discuss your answer.

Descriptive Statistics
reading_scor
math_score
e

Valid 1000 1000

Missing 0 0
Mean 69.169 66.089
Std. Deviation 14.600 15.163
Shapiro-Wilk 0.993 0.993
P-value of Shapiro-
< .001 < .001
Wilk
Minimum 17.000 0.000
Maximum 100.000 100.000

d. Using your results in b and c, if you are to decide a statistical test, what type of
test (parametric or nonparametric) and what test will you choose to analyze the
data?
Non-parametric
e. Implement the data in JASP and discuss the result.
EFD 502 Advanced Statistics with Software Application
Exam

Paired Samples T-Test

Measure 1 Measure 2 W z df p

reading_scor math_scor
- 313786.500 10.177 < .001
e e

Note. Wilcoxon signed-rank test.

3. (20 points) Course preparation before a test is crucial for many reasons. Thorough
preparation not only enhances your understanding of the subject but also boosts your
confidence and reduces anxiety. Accordingly, good preparation can significantly improve
student’s performance. Determine whether the effect of test preparation for a course in
the writing scores of students conforms with the literature.
a. Display the descriptive statistics for each data and describe mean, median,
standard deviations and IQR of both data.

Descriptive Statistics
writing_score
completed none

Valid 358 642

Missing 0 0
Median 76.000 65.000
Mean 74.419 64.505
Std. Deviation 13.375 15.000
IQR 17.000 20.000
Minimum 36.000 10.000
Maximum 100.000 100.000

b. Display the graphs (histogram, Q-Q plot, and boxplot) of the data. What can be
said about the data using these graphs? Enumerate your observations.
EFD 502 Advanced Statistics with Software Application
Exam

c. Compute the Shapiro-Wilk Test for Normality of the data. What can be said about
the data? Discuss your answer.

Descriptive Statistics
writing_score
complete
None
d

Valid 358 642

Missing 0 0
Median 76.000 65.000
Mean 74.419 64.505
Std. Deviation 13.375 15.000
IQR 17.000 20.000
Shapiro-Wilk 0.986 0.995
P-value of Shapiro-
0.001 0.042
Wilk
Minimum 36.000 10.000
Maximum 100.000 100.000

d. Compute the Levene’s Test of Homogeneity. What can be said about the data?
Discuss your answer.

Test of Equality of Variances (Levene's)

F df1 df2 P

5.73 99
writing_score 1 0.017
9 8
EFD 502 Advanced Statistics with Software Application
Exam

e. Using your results in b, c and d, if you are to decide a statistical test, what type of
test (parametric or nonparametric) and what test will you choose to analyze the
data? Non-parametric
f. Implement the data in JASP and discuss the result.

Kruskal-Wallis Test
Factor Statistic df p

test_preparation_cours
100.524 1 < .001
e

Dunn's Post Hoc Comparisons - test_preparation_course

Comparison z Wi Wj rrb p pbonf pholm

completed - 623.10 < .00 < .00

10.026 432.134 0.382 < .001
none 1 1 1

Note. Rank-biserial correlation based on individual Mann-Whitney tests.

ANOVA - writing_score
Sum of
Cases df Mean Square F p
Squares

test_preparation_cours 108.35 < .00

22591.447 1 22591.447
e 1 1

Residuals 208085.637 998 208.503

Note. Type III Sum of Squares

Standard (HSD)
Post Hoc Comparisons - test_preparation_course
Mean
SE df t ptukey
Difference

complete 99 < .00

none 9.914 0.952 10.409
d 8 1

4. (20 points) Determine whether race or ethnicity affects math scores of students.
a. Display the descriptive statistics for each data and describe mean, median,
standard deviations and IQR of both data.
EFD 502 Advanced Statistics with Software Application
Exam

Descriptive Statistics
math_score
group
group A group C group D group E
B

Valid 89 190 319 262 140

Missing 0 0 0 0 0
Median 61.000 63.000 65.000 69.000 74.500
Mean 61.629 63.453 64.464 67.363 73.821
Std.
14.523 15.468 14.853 13.769 15.534
Deviation
IQR 20.000 20.000 19.000 18.000 20.250
Minimum 28.000 8.000 0.000 26.000 30.000
100.00
Maximum 100.000 97.000 98.000 100.000
0

b. Display the graphs (histogram, Q-Q plot, and boxplot) of the data. What can be
said about the data using these graphs? Enumerate your observations.
EFD 502 Advanced Statistics with Software Application
Exam

c. Compute the Shapiro-Wilk Test for Normality of the data. What can be said about
the data? Discuss your answer.

Descriptive Statistics
math_score
group A group B group C group D group E

Valid 89 190 319 262 140

Missing 0 0 0 0 0
Median 61.000 63.000 65.000 69.000 74.500
Mean 61.629 63.453 64.464 67.363 73.821
EFD 502 Advanced Statistics with Software Application
Exam

Descriptive Statistics
math_score
group A group B group C group D group E
Std. Deviation 14.523 15.468 14.853 13.769 15.534
IQR 20.000 20.000 19.000 18.000 20.250
Shapiro-Wilk 0.992 0.981 0.989 0.990 0.977
P-value of Shapiro-
0.855 0.010 0.017 0.059 0.018
Wilk
Minimum 28.000 8.000 0.000 26.000 30.000
Maximum 100.000 97.000 98.000 100.000 100.000

d. Compute the Levene’s Test of Homogeneity. What can be said about the data?
Discuss your answer.

Test for Equality of Variances (Levene's)

F df1 df2 p

0.56 4.00
995.000 0.686
8 0

e. Using your results in b, c and d, if you are to decide a statistical test, what type of
test (parametric or nonparametric) and what test will you choose to analyze the
data? Non-parametric
f. Implement the data in JASP and discuss the result.

Kruskal-Wallis Test
Factor Statistic df p

race_ethnicit < .00

57.079 4
y 1

Dunn's Post Hoc Comparisons - race_ethnicity

Comparison Z Wi Wj rrb p pbonf pholm

405.07 453.25
group A - group B -1.299 0.101 0.194 1.000 0.388
3 0
405.07 468.38
group A - group C -1.829 0.133 0.067 0.674 0.202
3 2
405.07 528.12 < .00
group A - group D -3.473 0.252 0.005 0.003
3 4 1
405.07 646.77 < .00
group A - group E -6.174 0.453 < .001 < .001
3 5 1
453.25 468.38
group B - group C -0.572 0.033 0.567 1.000 0.567
0 2
453.25 528.12
group B - group D -2.721 0.153 0.007 0.065 0.033
0 4
group B - group E -6.017 453.25 646.77 0.376 < .00 < .001 < .001
EFD 502 Advanced Statistics with Software Application
Exam

Dunn's Post Hoc Comparisons - race_ethnicity

Comparison Z Wi Wj rrb p pbonf pholm
0 5 1
group C - group 468.38 528.12
-2.481 0.125 0.013 0.131 0.052
D 2 4
468.38 646.77 < .00
group C - group E -6.094 0.355 < .001 < .001
2 5 1
528.12 646.77 < .00
group D - group E -3.925 0.257 < .001 < .001
4 5 1

Note. Rank-biserial correlation based on individual Mann-Whitney tests.

5. (20 points) Reading and writing are complementary skills. Improving one often leads to
improvements in the other. For instance, strong reading skills can provide the foundation
for effective writing, while frequent writing practice can enhance reading comprehension
and analytical abilities. How are the reading and writing scores of the students related?
a. Display the descriptive statistics for each data and describe mean, median,
standard deviations and IQR of both data.

Descriptive Statistics
reading_scor
writing_score
e

Valid 1000 1000

Missing 0 0
Median 70.000 69.000
Mean 69.169 68.054
Std. Deviation 14.600 15.196
IQR 20.000 21.250
Minimum 17.000 10.000
Maximum 100.000 100.000

b. Display the graphs (histogram, Q-Q plot, and boxplot) of the data. What can be
said about the data using these graphs? Enumerate your observations.
EFD 502 Advanced Statistics with Software Application
Exam

c. Compute the Shapiro-Wilk Test for Normality of the data. What can be said about
the data? Discuss your answer.

Descriptive Statistics
reading_score writing_score

Valid 1000 1000

Missing 0 0
Median 70.000 69.000
Mean 69.169 68.054
Std. Deviation 14.600 15.196
IQR 20.000 21.250
Shapiro-Wilk 0.993 0.992
P-value of Shapiro-
< .001 < .001
Wilk
Minimum 17.000 10.000
Maximum 100.000 100.000
EFD 502 Advanced Statistics with Software Application
Exam

d. Compute the Levene’s Test of Homogeneity. What can be said about the data?
Discuss your answer.

e. Show the scatterplot of the data. What can be said about the data in terms of their
linear relationship? Enumerate your answers.
f. Using your results in b, c, d, and e, if you are to decide a statistical test, what type
of test (parametric or nonparametric) and what test will you choose to analyze the
data?
g. Implement the data in JASP and discuss the result.
6. (20 points) Students with strong reading comprehension skills tend to perform better in
math because they can better understand and solve word problems. Moreover, students
who can clearly express their mathematical reasoning in writing often have a deeper
understanding of math concepts, leading to better performance. Identify if reading and
writing skills are essential predictors of math skills by analyzing the reading, writing, and
math scores of the students.
a. Display the descriptive statistics for each data and describe mean, median,
standard deviations and IQR of both data.
b. Display the graphs (histogram, Q-Q plot, and boxplot) of the data. What can be
said about the data using these graphs? Enumerate your observations.
c. Compute the Shapiro-Wilk Test for Normality of the data. What can be said about
the data? Discuss your answer.
d. Compute the Levene’s Test of Homogeneity. What can be said about the data?
Discuss your answer.
e. Show the scatterplot of the data. What can be said about the data in terms of their
linear relationship? Enumerate your answers.
f. Using your results in b, c, d, and e, if you are to decide a statistical test, what type
of test (parametric or nonparametric) and what test will you choose to analyze the
data?
g. Implement the data in JASP and discuss the result.