Comprehensive Examination: Statistics Engr. George C.
Serot, MBA, MAS
July 20, 2023 Facilitator
------------------------------------------------------------------------------------------------------------------------------------------
Instructions:
1. Answer all the questions to the best of your knowledge.
2. All interpretations must be direct to the point.
3. Use data analysis or Mega stat in all your computations.
4. Discussion, computation and justification must be summarized in Word Document
by way of a folder with your examination number as file name.
_____________________________________________________________________________________
Questions
1. Four Pupils (A, B, C & D) obtained the following Grades in 11 subjects In Santa Barbara Elementary School as
of March 2021 as follows:
A – 83, 94, 88, 90, 85, 91, 95, 81, 89, 93, 80
B - 90, 87, 94, 84, 96, 80, 89, 95, 90, 92, 75
C - 80, 90, 97, 84, 88, 93, 91, 96, 89, 92, 77
D – 79, 95, 88, 83, 90, 84, 91, 89, 85, 93, 78
Pupil A Pupil B Pupil C Pupil D
Mean 88.09091 Mean 88.36364 Mean 88.81818 Mean 86.81818
Standard 1.569334 Standard 1.959929 Standard 1.891696 Standard 1.661424
Error Error Error Error
Median 89 Median 90 Median 90 Median 88
Mode #N/A Mode 90 Mode #N/A Mode #N/A
Standard 5.204893 Standard 6.50035 Standard 6.274045 Standard 5.510321
Deviation Deviation Deviation Deviation
Sample 27.09091 Sample 42.25455 Sample 39.36364 Sample 30.36364
Variance Variance Variance Variance
Kurtosis -1.2338 Kurtosis 0.319625 Kurtosis -0.18312 Kurtosis -0.89674
Skewness -0.31282 Skewness -0.92711 Skewness -0.69939 Skewness -0.25298
Range 15 Range 21 Range 20 Range 17
Minimum 80 Minimum 75 Minimum 77 Minimum 78
Maximum 95 Maximum 96 Maximum 97 Maximum 95
Sum 969 Sum 972 Sum 977 Sum 955
Count 11 Count 11 Count 11 Count 11
a. Who performs better and why? Justify your answer…
Answer. Based on the result above, if we are going to consider the standard deviation, the one who performs
better in class is the Student A for having standard deviation of 5.2 which means that it is less dispersed
compared to the other pupils. Aside from that he/she is only the pupils with no grade 79 below. Therefore,
Pupil A is the one who performs better in class.
a. Do the Grades of Pupils A & D significantly differ?
t-Test: Two-Sample Assuming Unequal Variances
Variable 1 Variable 2
Mean 88.09091 86.81818
Variance 27.09091 30.36364
Observations 11 11
Hypothesized Mean Difference 0
Df 20
t Stat 0.55689
P(T<=t) one-tail 0.291891
t Critical one-tail 1.724718
P(T<=t) two-tail 0.583781
t Critical two-tail 2.085963
Answer. Yes, pupil A and Pupil D have so many differences including its mean. Pupil A has
a mean of 88.09091 while Pupil D has a mean of 86.81818 which means that Pupil A perform
better in class compare to Pupil D. Their next differences are when it comes to their variance,
Pupil A has a variance of 27.09091 while Pupil D has a variance of 30.36364. This means
that the grades of Pupil A are less dispersed and are closely related to one another than the
grades of Pupil D.
b. Do the Grades of Pupils B & C significantly relate?
c. t-Test: Two-Sample Assuming Unequal Variances
Variable 1 Variable 2
Mean 88.36364 88.81818
Variance 42.25455 39.36364
Observations 11 11
Hypothesized Mean Difference 0
df 20
t Stat -0.16687
P(T<=t) one-tail 0.434574
t Critical one-tail 1.724718
P(T<=t) two-tail 0.869147
t Critical two-tail 2.085963
Answer. Yes, their grades are significantly related considering that their variance is almost close
from each other. Although its mean has a little difference but they are close or almost related that is
why their grades are significantly related.
2. Given the following scores of 66 Pupils in National Achievement Test as of July
2020
120 132 111 121 122 144 133 130 120 132 131
120 110 111 121 141 151 126 157 124 175 151
140 141 111 100 100 100 100 121 101 103 140
151 140 121 101 144 155 154 150 130 133 144
120 140 144 156 165 120 130 140 144 114 155
101 111 100 100 100 100 130 144 156 155 150
Construct the frequency distribution for the given set of data with a number of class
interval of 7. Describe some of the results.
Class f <cf >cf Class Class Boundaries Relative
Interval Mark Frequency
Lower Upper
Boundary Boundary
100- 110 13 13 66 105 99.5 110.5 19.70 %
111- 121 14 27 65 116 110.5 121.5 21.21%
122- 132 10 37 58 127 121.5 132.5 15.15%
133- 143 9 46 46 138 132.5 143.5 13.64%
144 - 154 12 58 37 149 143.5 154.5 18.18%
155- 165 7 65 27 160 154.5 165.5 10.61%
166- 176 1 66 13 171 165.5 176.5 1.52%
N = 66
Answer. The frequency distribution table displays the scores of the 66 pupils in National
Achievement Test. The scores range from 100 to 175. The table organized into interval which is
10.7142857142 or 11. It also shows that only 1 got the score of 166-176 and many pupils got the
score of 111 – 121.
3. From a study with ten patients with hypertriglyceridemia, data on cholesterol levels and triglyceride
levels were recorded and was found out to be linearly correlated significantly at 0.05 level of
significance, Suppose the study will be replicated here in the Philippines and we have the following
data on cholesterol and triglyceride levels of 15 Filipinos with hypertriglyceridemia.
Cholesterol Level- 9.51, 9.02, 9.70, 8.94, 9.56, 4.45, 4.54, 5.69, 4.46, 4.59, 6.62, 6.84, 7.37, 7.60, 7.10
Triglyceride Level- 8.41, 8.17, 7.93, 7.87, 8.27, 3.39, 3.94, 4.10, 3.82, 3.60, 4.86, 4.22, 4.16, 5.00, 4.27
A. Would you say that cholesterol and triglyceride levels are also linearly
correlated here in the Philippines?
SUMMARY OUTPUT
Regression Statistics
Multiple R 0.908657
R Square 0.825657
Adjusted R 0.812246
Square
Standard
Error 0.863978
Observations 15
ANOVA
Significanc
df SS MS F eF
Regression 1 45.95633 45.95633 61.56581 2.76E-06
Residual 13 9.703962 0.746459
Total 14 55.66029
Coefficient Standard Upper Lower Upper
s Error t Stat P-value Lower 95% 95% 95.0% 95.0%
Intercept -0.95176 0.847964 -1.1224 0.281994 -2.78367 0.880157 -2.78367 0.880157
X Variable 1 0.90922 0.115877 7.846389 2.76E-06 0.658882 1.159558 0.658882 1.159558
Answer. Yes. Considering the above data where I used regression, cholesterol and
triglyceride levels here in the Philippines are also linearly correlated. We can see in the data
that it has obtained a multiple R of 0.908657 which means that there is a very high
correlation between the two.
b) If an individual with hypertriglyceridemia is known to have a cholesterol level of
8.0 (mmol/l), then at what level is his/her triglyceride level is expected to be?
Answer. We solved the level of his/her triglyceride level by using the formula Y= a+bx
Y= a+bx
= -0.951758253 + 0.909219771 (8.0)
Y= 6.322
If an individual has hypertriglyceridemia and has a cholesterol level of 8.0 then his/her triglyceride
level is 6.322
4. Discuss at least two important learnings in Descriptive Statistics and another two
in Inferential Statistics. How are these learnings applied in your present work?
Answer. Descriptive Statistics involves analysis, interpretation, and presentation of data. An
example of which is the measures of Central Tendency and Measures of Variability. While Inferential
Statistics goes beyond Descriptive Statistics, it uses simple data to make inferences or draw
conclusions about a larger population. Examples are Regression Analysis and Hypothesis Testing.
These are very helpful in my career especially that we usually reported the MPS of our learners every
end of the quarter in order to track their progress. Aside from that I also recorded their performance
using the standard deviation. Every quarter they are going to count the starts they have earned and I
am going to give them points using the standard deviation since some of them earned more than 100
stars. Last month, I was assigned as the Proctor of the Regional Summative Assessment (RSA). Before
conducting RSA, we first perform random sampling since only 20 pupils are going to take the said
exam out of the 80 plus students that we have. Statistics really helped me in my chosen career.
5. Enumerate, illustrate and discuss the different properties of a normal curve.
Answer. A normal curve, also known as Gaussian distribution is a type of distribution that is
symmetrical around its center which means that the left half of the curve is the same or mirror to the
right half. Another is that its distinct pattern which is the bell-shaped pattern. Its next property is that
it is unimodal which means it has a single peak. Lastly is that, the SD remains constant throughout the
entire curve. The wider the curve, the larger the SD, and vice versa.
Based on the illustration above, the curve is symmetrical with each other and it also has a bell-shaped
and has a single peak or unimodal.
6. One of the major considerations in parametric statistic is normality of the data.
What are the different parameters to validate that the data is normally distributed.
Explain
Answer. It is normally distributed if the variance of the data is equal, if its curve is a bell-shaped
curve. If its skewness is equal to 0. Another thing to consider is that if its kurtosis is 0.625 and the
significance of the Shapiro-Wilk Test, Kolmogorov-Smirnov Test, or other appropriate normality test
just like for example the p value is greater than 0.05.
7. Differentiate Descriptive from Inferential Statistics
Answer. Descriptive and Inferential Statistics are two branches of statistics with different purpose
and method. Descriptive Statistics focuses only on summarizing and presenting data in a way that it is
easy to understand and easy to interpret. Its goal is to describe the data base without making
generalizations about the larger population. For examples is the Central Tendency which is the mean,
median and mode and the measures of variability, the variance and standard deviation. Its data can
be represented by bar graphs, histogram and many other.
While on the other hand, Inferential Statistics focuses on drawing conclusions and making
predictions about a larger population based on the sample gathered. Examples are the regression
analysis and hypothesis testing. Questions like significant relationships between two variables was
ask in this type of statistics. Though they are different from each other but they both played an
essential role in statistics.
8. There are cases where the three measures of Central Tendency are not considered
appropriate. What instances where one is superior over the others
Answer. The mean which is the average is more powerful over the median and the mode if it is
normally distributed and if there were no outliers. Then, the median is more powerful than the two if
the data is skewed or if it contains outliers since it represents the middle value. Aside from that the
mean is also powerful if it is an ordinal data. As for the mode, it is superior over the other if it is a
multimodal data.
END
