ST.

XAVIER’S COLLEGE
KOLKATA
(AUTONOMOUS)

1st SEMESTER EXAMINATION

JANUARY – FEBRUARY 2021
B.A./B.Sc.
Wednesday, January 20, 2021
HSTCR1012T 12:00 Noon to 2:30 PM
2 hours 30 minutes
DESCRIPTIVE STATISTICS
Full Marks : 50

PLEASE READ THESE INSTRUCTIONS BEFORE YOU START WRITING:

• Of the questions attempted, the answers to only the first required number of questions (as stipulated
in the question paper) will be evaluated. So please do not attempt extra questions.

• Only HAND WRITTEN answer scripts (using black or blue ink) on A4 size sheets will be
considered.
• Answer in your own words as far as practicable.
• Save the scanned pages to a single PDF file and name the document accurately i.e. Roll No_Paper
Code.PDF (example: 147_HPHCR2032T).
• Students have to write legibly their answers on A4 size sheets, scanning them and sending the entire
answer script as one PDF file to the following email address within 30 minutes of the completion
of the examination: [email protected]
• In the subject field of your email, please write “Answer Script – Roll No, Paper Code” (example:
“Answer Script – 147, HPHCR2032T”).
• The scanned answer scripts should have enough clarity to enable evaluation.
• On top of each page the following information should be entered by the student: Name, Roll
Number, Paper Code , Date, and Page Number
• No multiple submissions would be allowed.

The marks are given in brackets [ ] at the end of each question or part question.

The question paper consists of 4 pages.

2021 1
 Of the questions attempted, the answers to only the first required number of
questions (as stipulated in the question paper) will be evaluated.
So, PLEASE DO NOT ATTEMPT EXTRA QUESTIONS.

Answer ANY TWO questions from Question Nos. 1 to 4 and

ANY TWO from Question Nos. 5 to 8.

1. (a) Random samples of 20 bobbins were selected from each of the five factories. The
diameters of the bobbins were measured and the students were asked to report the
overall variation in the measurement. By mistake one student used the formula
1 5 2
∑
5 i =1
Si but got the same answer as obtained by using the correct formula. Here Si2

denotes the variance in measurement corresponding to the ith (i=1,2,..,5) factory.

Assuming that the student made no calculation error how can you interpret the
finding?
(b) Paired observations on two variables (x,y) are provided for two separate categories.
To assess the overall strength of association between x and y, one computes r1 and r2
and observes that both are close to zero, where ri (i=1,2) denotes the correlation
coefficient of the ith (i=1,) set. She concludes that x and y are not linearly related. Is
she justified in her statement? [2.5+2.5]

2. (a) Daily temperature for the months of November, December and January for the year
2019 are provided for Kolkata and Darjeeling. Suggest with justification an
appropriate measure of dispersion for comparing the variability in temperature
between the two places.
(b) State true or false with justification (no proof needed) “For a variable x, mean=8,
median=4 and variance=3.24” [3+2]

3. A variable x takes only two values x1 and x2 with equal frequencies. Compute Mean
deviation about the arithmetic mean. [5]

4. At the beginning of the Semester in Health dynamics, the measurement on the Forced
Vital Capacity (FVC) (the volume of air a person can expel from their lungs) was taken for
some college students. The FVC in litters for 25 females are given on the basis of which
the five points are calculated as
x(1)=2.80; Q1=3.40; Q2=3.60; Q3=4.20; x(n)=4.20.
The symbols above have their usual meaning. Give a rough sketch of an appropriate
diagrammatic presentation from the above information. Interpret the findings. [5]

5. (a) Suppose a researcher is interested in studying the factors influencing academic

performance of undergraduate students in the age group 19-22. Mention four factors
that you may think can affect the academic performance. Classify the factors
according to the different scales of measurement (All the four factors should not be
measured using the same scale).

2021 2
 (b) Describe the situations where you would recommend the use of (i) a multiple bar
diagram (ii) a pie diagram. Give one example in each case.
R2
(c) Interpret the cases (i) g 2 = −2 (ii) S 2 = , where g2 , S and R denote respectively
2n
the moment measure of kurtosis, the standard deviation and the range, calculated
from a set of n observations.
(d) Let x be a variable assuming the value i with frequency fi and cumulative frequency of
less than type Fi (i=1,2,..,k). Obtain an expression of arithmetic mean in terms of
cumulative frequencies of less than type. [5+5+5+5]

6. (a) The relationship between dose (x) and response (Y) is known to be of the form
x
Y= . Convert it to a linear model and estimate the model parameters by least
β 0 + β1 x
square technique, based on n pairs of observations on (x,y). What does the ratio
V ( fitted Y )
indicate?
V (observed y )
(b) Prove or disprove: Standard deviation =2.8, Mean absolute deviation about arithmetic
mean =3.2 [12+8]

7. (a) Consider the following response-predictor relationship: Y = α x + β x2 . Based on n

pairs of observations on (x,y) and using Cramer’s rule derive the least square
estimates of the model parameters. Obtain a simplified expression of V(fitted Y).
V ( fitted Y )
Hence compute the ratio and interpret.
V (observed y )
(b) Prove or disprove: “For a variable x with range 6, the variance cannot be more than
9”. [12+8]

8. Data on length of stay in hospital (y) after been tested COVID positive and the patients’
viral load (x) are available from p randomly selected hospitals of Kolkata. From each
hospital, records of m randomly selected patients are noted. Moreover data on various
infrastructural facilities and overall hygiene maintenance (z) are also recorded for each
hospital.
(a) Based on the data on z, two judges rank the p hospitals. Assuming that there could be
ties in their ranking, interpret the cases: rR = −1, rR = +1, rR = 0 , where rR denotes
Spearman’s rank correlation coefficient.
(b) Using the data on y, a researcher obtained “between hospital variation” as 0.20.
Interpret the finding.
(c) Suppose another researcher classifies the viral loads into k mutually exclusive
categories. Corresponding to each category she notes the length of stay in hospital.
Let yij denote the duration of stay in hospital by the jth person corresponding to ith
k
viral load (j=1,2,..,ni; i=1,2,..,k). Note that ∑n
i =1
i = mp .

(i) Fit a linear regression equation to the above data. Give expression of the slope
and intercept parameters.
2021 3
 (ii) Interpret the finding Sw =0 , where Sw denotes “variation within viral loads”.
(iii) Interpret the findings: (i) e yx2 = 1, r 2 = 0 (ii) eyx2 = 0.5, r 2 = 0.5 (iii) eyx2 = 1, r 2 = 1
2
where eyx and r2 denotes respectively the correlation ratio and the square of
correlation coefficient between y and x . [3+2+(10+2+3)]

***********************

2021 4