FND 602/608

Problem Set 2 (Nonparametric Statistics)

Name: Ervin C. Malones Program/Degree: PhD Science Ed. (Biology) Score:

October 3, 2021

A. Using the DATA FILE STAT, answer the following problem/s using the steps in
hypothesis testing.

1. Are the students’ SATT scores normally distributed?

Solution:
1. Formulate the null hypothesis.
Ho: The distribution of students’ SATT scores does not deviate from
normality.
Ha: The distribution of students’ SATT scores deviates from
normality.
2. Set the level of significance and tailedness of the test.
α =0.05
tailedness: two-tailed
3. Determine the test to be used.
Test statistic: One-sample K-S Test
4. Compute the statistical test.

One-Sample Kolmogorov-Smirnov Test

SATT Score
N 45
Normal Parameters a,b
Mean 84.4667
Std. Deviation 6.74739
Most Extreme Differences Absolute .079
Positive .073
Negative -.079
Test Statistic .079
Asymp. Sig. (2-tailed) .200c,d
a. Test distribution is Normal.
b. Calculated from data.
c. Lilliefors Significance Correction.
d. This is a lower bound of the true significance.

5. Compare the significance/ probability obtained to the level of significance.

Make your
decision.
Reject H0 if p≤α, otherwise do not reject.
Decision: Since the p-value of .200 is higher than 0.05 alpha, accept
Ho
6. Make your conclusion.
Conclusion: The distribution of students’ SATT scores does not
deviate from normality, Z=0.079, p=0.200.

FND 602/608- GCC 1

2. Thirty students in a statistics class were exposed to an experiment by their
teacher. The teacher-researcher used application software, specifically the
use of SPSS and geogebra, as an intervention program in increasing their
performance in hypothesis testing. A pre-intervention and post-intervention
test duly validated and reliability tested were given to the student-
participants to measure their performance in hypothesis testing. This test is
following the six steps in hypothesis testing with a total score of 50 points.
Test whether the intervention is effective or not using the steps in hypothesis
testing.

Solution:
1. Formulate the null hypothesis.
Ho: There is no significant difference in the pre-intervention and
post intervention of students when exposed to a particular
intervention.
Ha: There is a significant difference in the pre-intervention and post
intervention of students when exposed to a particular intervention.
2. Set the level of significance and tailedness of the test.
α =0.05
tailedness: two-tailed
3. Determine the test to be used.
Test statistic: Wilcoxon Signed-Ranks Test
4. Compute the statistical test.

Ranks
N Mean Rank Sum of Ranks
a
Post-intervention - Negative Ranks 0 .00 .00
Pre-intervention Positive Ranks 42b 21.50 903.00
c
Ties 3
Total 45
a. Post-intervention < Pre-intervention
b. Post-intervention > Pre-intervention
c. Post-intervention = Pre-intervention

Test Statisticsa
Post-intervention - Pre-
intervention
Z -5.652b
Asymp. Sig. (2-tailed) .000
a. Wilcoxon Signed Ranks Test
b. Based on negative ranks.

5. Compare the significance/ probability obtained to the level of significance.

Make your
decision.

FND 602/608- GCC 2

 Reject Ho if p≤α, otherwise do not reject.
Decision: Since the p-value of .000 is less than 0.05 alpha, reject Ho
6. Make your conclusion.
Conclusion: There is a significant difference in the pre-intervention and
post intervention scores of students when exposed to a particular
intervention, Z=-5.652, p=0.000.

3. Determine whether students’ SATT score significantly differ across sex

assuming that the respondents were purposively chosen.
Solution:
1. Formulate the null hypothesis.
Ho: There is no significant difference in the students’ SATT scores
when
classified as to sex.
Ha: There is a significant difference in the students’ SATT scores
when
classified as to sex.
2. Set the level of significance and tailedness of the test.
α=0.05
tailedness: two-tailed
3. Determine the test to be used.
Test statistic: Mann-Whitney U-Test
4. Compute the statistical test.

Ranks
Sex N Mean Rank Sum of Ranks
SATT Score Male 23 23.63 543.50
Female 22 22.34 491.50
Total 45

Test Statisticsa
SATT Score
Mann-Whitney U 238.500
Wilcoxon W 491.500
Z -.330
Asymp. Sig. (2-tailed) .741
a. Grouping Variable: Sex

5. Compare the significance/ probability obtained to the level of significance.

Make your
decision.
Reject Ho if p≤α, otherwise do not reject.
Decision: Since the p-value of .741 is greater than 0.05 alpha, do not
reject Ho.
6. Make your conclusion.
Conclusion: There is no significant difference in the students’ entrance
test scores when classified as to sex, Z=-0.330, p=0.741.

FND 602/608- GCC 3

4. Determine whether students’ SATT score significantly differ across highest
educational attainment of the father assuming that the respondents were
purposively chosen.
Solution:
1. Formulate the null hypothesis.
Ho: There is no significant difference in the students’ SATT scores
when classified as to highest educational attainment of the father.
Ha: There is a significant difference in the students’ SATT scores
when classified as to highest educational attainment of the father.
2. Set the level of significance and tailedness of the test.
α=0.05
tailedness: two-tailed
3. Determine the test to be used.
Test statistic: Kruskal-Wallis Test
4. Compute the statistical test.

Ranks
HEA N Mean Rank
SATT Score Secondary 15 12.57
Bachelors 15 20.90
Masters 15 35.53
Total 45

Test Statisticsa,b
SATT Score
Chi-Square 23.639
df 2
Asymp. Sig. .000
a. Kruskal Wallis Test
b. Grouping Variable: HEA

Mann Whitney Comparison

Ranks
HEA N Mean Rank Sum of Ranks
SATT Score Secondary 15 12.07 181.00
Bachelors 15 18.93 284.00
Total 30

Test Statisticsa

FND 602/608- GCC 4

 SATT Score
Mann-Whitney U 61.000
Wilcoxon W 181.000
Z -2.148
Asymp. Sig. (2-tailed) .032
Exact Sig. [2*(1-tailed Sig.)] .033b
a. Grouping Variable: HEA
b. Not corrected for ties.

Ranks
HEA N Mean Rank Sum of Ranks
SATT Score Secondary 15 8.50 127.50
Masters 15 22.50 337.50
Total 30

Test Statisticsa
SATT Score
Mann-Whitney U 7.500
Wilcoxon W 127.500
Z -4.367
Asymp. Sig. (2-tailed) .000
Exact Sig. [2*(1-tailed Sig.)] .000b
a. Grouping Variable: HEA
b. Not corrected for ties.

Ranks
HEA N Mean Rank Sum of Ranks
SATT Score Bachelors 15 9.97 149.50
Masters 15 21.03 315.50
Total 30

Test Statisticsa
SATT Score
Mann-Whitney U 29.500
Wilcoxon W 149.500
Z -3.457
Asymp. Sig. (2-tailed) .001
Exact Sig. [2*(1-tailed Sig.)] .000b
a. Grouping Variable: HEA
b. Not corrected for ties.
5. Compare the significance/ probability obtained to the level of significance.
Make your
decision.
Reject H0 if p≤α, otherwise do not reject.
Decision: Since the p-value of .000 is less than 0.05 alpha, reject Ho.
6. Make your conclusion.

FND 602/608- GCC 5

 Conclusion: There is a significant difference in the students’ SATT
scores when classified as to highest educational attainment of the
father, x2(2)=23.639, p=0.000. Using Mann Whitney Comparison as
post hoc test, the significant differences existed among three
categories, between students whose father is a secondary graduate
and masters’ degree holder [Z=-4.367, p=0.000], between students
whose father is a bachelors’ degree holder and masters’ degree
holder [Z=-3.457, p=0.001], and lastly between students whose
father is a secondary graduate and bachelors’ degree holder [Z=-
2.148, p=0.032].

5. Determine whether students’ grades in Math and grades in English are

significantly related assuming that the respondents were purposively chosen
and the data are not approximately normally distributed.
Solution:
1. Formulate the null hypothesis.
Ho: There is no significant relationship between students’ grades in
Math and English.
Ha: There is a significant relationship between students’ grades in
Math and English.
2. Set the level of significance and tailedness of the test.
α=0.05
tailedness: two-tailed
3. Determine the test to be used.
Test statistic: Spearman rho
4. Compute the statistical test.

Correlations
Grades in Grades in
English Math
Spearman's Grades in Correlation
1.000 .367*
rho English Coefficient
Sig. (2-tailed) . .013
N 45 45
Grades in Math Correlation
.367* 1.000
Coefficient
Sig. (2-tailed) .013 .
N 45 45
*. Correlation is significant at the 0.05 level (2-tailed).

5. Compare the significance/ probability obtained to the level of significance.

Make your
decision.
Reject H0 if p≤α, otherwise do not reject.
Decision: Since the p-value of .000 is less than 0.05 alpha, reject Ho

FND 602/608- GCC 6

6. Make your conclusion.
Conclusion: There is a significant relationship between students’
grades in Math and English, rs=0.367, p=0.013.

FND 602/608- GCC 7