You are on page 1of 8

FINAL EXAMINATION

a. Find the relationship between the Physics and Chemistry score of a


mathematics student examination
using the data at the right. Test the Studen
significance of r at the 0.05 level. t Physics Chemistry
No.
Step 1.
1 87 75
Null Hypothesis: There is no significant
2 69 85
relationship between
Physics and Chemistry 3 91 79
score of the mathematics
4 88 80
students’ examination.
5 84 72
6 90 80
Alternative Hypothesis: There is a significant
relationship between 7 85 88
Physics and Chemistry
score of the mathematics 8 76 83
students’ examination. 9 94 80
10 78 83
Step 2. Level of Significance: α = 0.05

Step 3. Appropriate Test Statistics: Pearson R Correlation


With the use of excel formula, the Pearson r would be r = - 0.414
And with the use of MegaStat,
Correlation Matrix

Student Chemistr
No Physics y
Student No 1.000    
Physics .024 1.000  
Chemistry .354 -.414 1.000

sample
9 size
± .666 critical value .05 (two-tail)
± .798 critical value .01 (two-tail)

Step 4. Decision: Reject Ho.

Step 5. Conclusion: There is a significant relationship between Physics and


Chemistry score of the mathematics students’ examination. It
further implies that there is a negative correlation between them,
meaning if one of them decreases, the other increases.

b. Using the data in number 1 convert the scores to rank and find the relationship
between them using the spearman rank correlation. Test the significance of ρ at
the 0.05 level.

Step 1.
Null Hypothesis: There is no significant association between Physics and
Chemistry score of the mathematics students’ examination.

Alternative Hypothesis: There is a significant association between Physics and


Chemistry score of the mathematics students’ examination.

Step 2. Level of Significance: α = 0.05

Step 3. Appropriate Test Statistics: Spearman Rank Correlation


Spearman Coefficient of Rank Correlation

Chemistr
Student No Physics y
Student No 1.000    
Physics -.006 1.000  
Chemistry .397 -.445 1.000

10 sample size

± .632 critical value .05 (two-tail)


± .765 critical value .01 (two-tail)
Step 4. Decision: Reject Ho.

Step 5. Conclusion: There is a significant association between Physics and


Chemistry score of the mathematics students’ examination. It
further implies that there is a negative correlation between them
meaning as if one of them decreases, the other increases.

c. Using the data in number 1 compute the linear regression equation and predict
the score in chemistry if the score of physics is 80.

Step 1.
Null Hypothesis: There is no significant prediction of Chemistry score by Physics
score.

Alternative Hypothesis: There is a significant prediction of Chemistry score by


Physics score.

Step 2. Level of Significance: α = 0.05

Step 3. Appropriate Test Statistics: Linear Regression

Regression Analysis

r² 0.167 n 10
r -0.409 k 1
Dep. Chemistr
Std. Error 4.500 Var. y

ANOVA
table
Source SS df MS F p-value
Regressio
n 32.5317 1 32.5317 1.61 .2406
Residual 161.9683 8 20.2460
Total 194.5000 9      

Regression output confidence interval


coefficien std. t 95% 95%
variables ts error (df=8) p-value lower upper
63.358 139.144
Intercept 101.2513 16.4322 6.162 .0003 6 1
Physics -0.2465 0.1944 -1.268 .2406 -0.6948 0.2019

Predicted values for: Chemistry


95% Confidence 95% Prediction
    Interval Interval  
Predicte Levera
Physics d lower upper lower upper ge
80 81.535 77.752 85.318 70.491 92.579 0.133

Step 4. Decision: Reject Ho.

Step 5. Conclusion: There is a significant prediction of Chemistry score by


Physics score.

d. A study was conducted to determine the compatibility of Microsoft and Linux


operating system to the college computer students in PIT. The data above show the
result of the survey. Using the 5% level of significance, test if the two operating
systems do not differ in their responses.
Operating High Moderate Low
System Compatibility Compatibility Compatibility
Microsoft 96 90 27
Linux 56 84 67

Step 1.
Null Hypothesis: There is no significant relationship between Microsoft and Linux
operating system.

Alternative Hypothesis: There is a significant relationship between Microsoft and


Linux operating system.

Step 2. Level of Significance: α = 0.05

Step 3. Appropriate Test Statistics: Chi-square

Chi-square Contingency Table Test for Independence


High Moderate Low
Compatibility Compatibility Compatibility Total
Microsof
t 96 90 27 213
Linux 56 84 67 207

Total 152 174 94 420

27.67 chi-square
2 Df
9.79E-07 p-value
             

Step 4. Decision: Reject Ho.

Step 5. Conclusion: There is a significant relationship between Microsoft and


Linux operating system. It further implies that both of them are
moderately compatible.

Sections
Students
e. Four teachers taught college Basic Statistics to A B C D
four sections of college students. The final
grades of a random sample of 8 students per 1 86 75 89 78
section are shown right. Is there a significant
difference in the final grades given by the 4 2 80 84 79 82
teachers at 0.05 level? 3 85 83 86 84
4 91 86 92 81
Step 1.
Null Hypothesis: The difference between 5 87 79 78 84
means of the final grades
6 92 83 82 87
of the students given by
four teachers are not 7 86 81 81 83
statistically significant.
8 90 88 90 79
Alternative Hypothesis: The difference
between some means of the final grades of the students given by
four teachers are statistically significant.

Step 2. Level of Significance: α = 0.05

Step 3. Appropriate Test Statistics: One-way ANOVA


One factor ANOVA

Mean n Std. Dev  


54.765 4.5 8 2.45 Sections
54.765 87.1 8 3.87 A
54.765 82.4 8 4.07 B
54.765 84.6 8 5.34 C
54.765 82.3 8 2.92 D

68.2 40 32.50 Total

ANOVA
table
p-
Source SS df MS F value
50,253.8 12,563.45 3.00E-
Treatment 0 4 0 12.02 06
36,585.7
Error 0 35 1,045.306
86,839.5
Total 0 39      

Post hoc analysis


Tukey simultaneous comparison t-values (d.f. = 35)
Sections D B C A
4.5 82.3 82.4 84.6 87.1
Sections 4.5          
D 82.3 4.81        
B 82.4 4.82 0.01      
C 84.6 4.96 0.15 0.14    
A 87.1 5.11 0.30 0.29 0.15  

critical values for experiment wise error rate:


0.05 2.88
0.01 3.53

p-values for pairwise t-tests


Sections D B C A
4.5 82.3 82.4 84.6Tablets87.1
Sections 4.5       A  B C  D
2.85E- 5.2 9.1 3.2 7.1
D 82.3 05        
4.7 7.1 5.8 6.6
2.78E-
B 82.4 05 .9939   8.1 8.2
  2.2 9.3
 
1.83E- 6.2 6.0 3.1 4.2
C 84.6 05 .8840 .8901    
3.0 9.1 7.2 7.6
1.15E-
A 87.1 05 .7648 .7706 .8780  

Step 4. Decision: Reject Ho.

Step 5. Conclusion: The difference between some means of the final grades of
the students given by four teachers are statistically significant.

f. The data in the following table represent the number of hours of relief provided by 4
different brands of headache tablets administered to 20 subjects experiencing fevers
of 380 C or more. Perform the Kruskal-Wallis analysis of variance and test the
hypothesis at the 0.05 level of significance that the mean number of hours of relief
provided by the tablets is the same for all 4 brands. Discuss the results.

Step 1.
Null Hypothesis: The difference between medians of the number of hours of
relief among the four brands of headache tablets are not
statistically significant.

Alternative Hypothesis: The difference between some medians of the number of


hours of relief among the four brands of headache tablets are
statistically significant.

Step 2. Level of Significance: α = 0.05

Step 3. Appropriate Test Statistics: Kruskal – Wallis


Kruskal-Wallis Test

Avg.
Median n Rank  
5.20 5 8.20 A
8.20 5 15.10 B
3.20 5 6.00 C
7.10 5 12.70 D
6.40 20 Total

7.374 H (corrected for ties)


3 d.f.
.0609 p-value

multiple comparison values for avg.


ranks
9.87 (.05) 11.76 (.01)

Step 4. Decision: Accept/do not reject Ho.

Step 5. Conclusion: The difference between medians of the number of hours of


relief among the four brands of headache tablets are not
statistically significant.

You might also like