East West University

Name: A.M. Mehedi Mostafiz; ID: 2020-1-91-004

Semester Final Examination; Date: September 26, 2020
Course No. EMBA-502: Business Mathematics and Statistics
Time: 90 minutes Full Marks: 30
Course Teacher: Dr. M. Amir Hossain
(Answer all questions. Figures in the right margin indicate the marks of the
question)

1. Distinguish between the following with example (any FIVE): (5×1 = 5)

(a) Strata and Cluster
Answer: The difference between stratified sampling and cluster sampling is that
with cluster sampling, we have natural groups separating your population. For example,
we might be able to divide our data into natural groupings like city blocks, voting
districts or school districts. In case of strata, these breaks may not exist, so we divide
our target population into groups.
(b) Correlation coefficient and regression coefficient
Answer:
Correlation coefficient regression coefficient
Correlation coefficient indicates the extent Regression coefficient indicates the
to which two variables move together impact of a change of unit on the
estimated variable in the known variable
There is no difference in between In case of regression coefficient both
dependent and independent variables variables are different
In order to find a numerical value In order to estimate values of random
expressing the relationship between variables on the basis of the values of a
variables fixed variables.

(c) Census and Sample survey

Answer:
Census Survey
In a census, data about all individual units In a survey, data are only collected for a
(e.g. people or households) are collected in sub-part of the population; this part is
the population called a sample.

1
 (e) Parameter and Statistic
Answer:
Parameter Statistics
A parameter is a descriptive measure A statistic is a descriptive measure
computed from an entire population of computed from a sample of data
data
Parameters are numbers that summarize Statistics are numbers that summarize data
data for an entire population. from a sample

(f) Positive and Negative Correlation

Answer:
Positive Correlation Negative Correlation
Positive correlation describes the Negative correlation describes the
relationship between two variables which relationship between two variables which
change together change in opposing directions.
A positive correlation exists when two An inverse correlation exists when two
related variables move in the same related variables move in the opposite
direction. direction.

2. Briefly answer the following questions (any 4): (4 ×2.5 = 10)

(a) What is a scatter diagram? Explain how you can draw the scatter diagram. How scatter
diagram is used to know the presence and type of relationship between variables?
Answer:
 A scatter diagram is a chart that portrays the relationship between the two variables
of interest.
 In order to draw the scatter diagram we have to place the dependent variable on
Y–axis and independent variable on X–axis
 Scatter diagrams are used to investigate the possible relationship between two
variables that both relate to the same event. If the scatter points are close and upward
we can say that the variables have relation and its positive. If the points are close but
downwards we can say that the variables have negative correlation. And if the points
are very scattered we can say that the variables have zero correlation.

2
(b) Define dependent and independent variables with real life example in connection with
regression analysis.
Answer:
 Dependent Variable: The variable that is being predicted or estimated.
 Independent Variable: The variable that provides the basis for estimation. It is the
predictor variable.
 Example: If we spend more time in studying we will get the marks more. So in this
case time of studying is a independent variable and the mark/ result is dependent
variable
(c) Why should we go for sample survey rather than census? Give a practical example
where sample survey is the only way to collect information.
Answer:
 Reason to go for sample survey rather than census-
1. Reduces cost
2. Reduces time
3. Quick result
 The sampling of accounting records for audits and the use of a computer to sample a
large database are sample surveys that use direct observation of the sampled units to
collect the data. A goal in the design of sample surveys is to obtain a sample that is
representative of the population so that precise inferences can be made. Sampling
error is the difference between a population parameter and a sample statistic used to
estimate it.

(d) What are the different probability sampling methods you are familiar with? Do you
think probability sampling is always better than non-probability sampling? Justify.
Answer:
 Different types of probability sampling methods
1. Simple random sampling
2. Systematic sampling
3. Stratified Sampling
4. Cluster Sampling
 Researchers prefer probabilistic or random sampling methods over nonprobabilistic
ones, and consider them to be more accurate and rigorous. However Non-probability
sampling is most useful for exploratory studies like a pilot survey Researchers use this
method in studies where it is impossible to draw random probability sampling due to
time or cost considerations.

3
 So we can say that most of the time probability sampling is better than non-
probability sampling but not always. If done correctly, it can produce similar if not the
same quality of results.

3. In a school there are 55 students, you are asked to determine the average age of the
students of that school based on a random sample of 6 students. If you want to select
the sample on systematic random sampling basis, discuss clearly how will you
proceed to select the sample. (3)
Answer: Given, N =55, n =6, Intervals N/n = 9.17 = 9
First we will make sampling frame,
1 10 19 28 37 46
2 11 20 29 38 47
3 12 21 30 39 48
4 13 22 31 40 49
5 14 23 32 41 50
6 15 24 33 42 51
7 16 25 34 43 52
8 17 26 35 44 53
9 18 27 36 45 54

1st sample = 2
2nd sample = 2+9 = 11
3rd sample=11+9= 20
4th sample = 20+9=29
5th sample= 29+9 = 38
6th sample= 38+9= 47

4
4. A multinational company operating in Bangladesh maintains their office in six
divisional head quarters. They have 300 employees in three different categories
(Executive, Officer, and support staff). The compensation package is same in all six
offices. Administration wanted to go for a primary inquiry on the satisfaction level of
the employees working in the company. They decided to carry out a research on 60
randomly selected employees to check on the satisfaction level of the employees. There
are 30 executives, 180 officers, and 90 support staffs. They used a continuous rating
scale from 0 to 100 to measure the satisfaction level.

(i) What is the population here? (1)

(ii) What is the sampling unit? (1)
(iii) What would be the sampling frame here? (1)
(iv) What sampling technique the authority should follow to select the sample?
Justify the answer. .

Answer: (i) Population = 300

(ii) Sampling unit = 300
(iii) Sampling frame = 300
(iv) Stratified Random sampling technique should be followed to select the sample.
Because population is first divided into subgroups, called strata, individuals having
homogeneous criteria of the variable of interest belongs to same strata. Classification of
the individuals into different strata based on the characteristic is known as
Stratification. After stratification a sample is selected from each stratum.

5. The manager of a motor company wanted to (has asked you to help her) determine the
relationship between vehicle weight and miles per gallon (mpg) for the company’s
fleet. She has given you the following information about vehicle weights, in 100’s of
pounds, and mpg for highway driving.
Weight: 29 35 28 44 25 34 30 33 28 24
Mpg : 31 27 29 25 31 29 28 29 28 33

Using above data answer following questions:

(a) Identify the dependent and independent variables 1
(b) Draw a scatter diagram and comment on the relationship between two variables. 1
(c) Determine the coefficient of correlation and interpret. 2
(d) Develop the associated regression equation 2
(e) What would be the ‘mpg’ of a vehicle having weight 3500 pounds 1

5
 Answer: (a) Independent variable: Weight of vehicle
Dependent variables: Mpg
(b)

(c)
Mpg Weight x -xbar y-ybar (x-xbar)x(x-xbar) (y-ybar)x(y-ybar) (x-xbar)x(y-ybar)
31 29 -2 2 4 4 -4
27 35 4 -2 16 4 -8
29 28 -3 0 9 0 0
25 44 13 -4 169 16 -52
31 25 -6 2 36 4 -12
29 34 3 0 9 0 0
28 30 -1 -1 1 1 1
29 33 2 0 4 0 0
28 28 -3 -1 9 1 3
33 24 -7 4 49 16 -28
306 46 -100

X bar = 31, Y bar = 29

Co efficient of correlation, r = -100/ 118.6423 = -0.843

6
 So, Strong Negative correlation between vehicle weight & Miles per gallon has been
observed.
So, (-0.843)2 = 71.04% of the variation in mpg can be explained by the variation in weights
of the vehicle
(d) b = sum{(x-xbar)x(y-ybar)}/ sum (x-xbar)x(x-xbar) = -100/ 306 = -0.327
a= y-bx = 29-(-0.327x 31) = 39.13

So, regression equation, y= 39.13 -0.327X

(e) The the ‘mpg’ of a vehicle having weight 3500 pounds = 39.13 – 0.327 x 35 = 27.685