Comp 328:Research Methods For Computer Science Question Paper
Comp 328:Research Methods For Computer Science
Course:Bachelor Of Computer Science
Institution: Kabarak University question papers
Exam Year:2012
UNIVERSITY EXAMINATIONS
2011/2012 ACADEMIC YEAR
FOR THE DEGREE OF BACHELOR OF COMPUTER SCIENCE
COMP 328: RESEARCH METHODS FOR COMPUTER SCIENCE
DAY: WEDNESDAY DATE: 21/03/2012
TIME: 9.00 – 11.00 A.M. STREA
M: Y3S2
INSTRUCTIONS:
1. Answer question ONE and any other TWO questions
2. Show all your working and be neat
QUESTION ONE (30 marks)
a)
Outline research procedure cycle one has to follow
for a credible study as taught to you in this course
[5 marks]
b)
Assume you have conducted a research for your academic dissertation/project, briefly outline what should
appear in each chapter of the university research booklet
[5 marks]
c)
Write short notes on simple random sampling design,
stratified random sampling, cluster sampling, multi-stage sampling and quota sampling including their merits and demerits.
[6 marks]
d)
Distinguish
between Karl Pearson’s correlation coefficient and
Spearman’s rank correlation coefficient,
showing clearly when one selected over the other
[4 marks]
e)
Differentiate between simple correlation and linear
regression as taught in this course
[6 marks]
f)
What is a one-way and two-way ANOVA; differentiate
them
[4 marks]
Page
2
of
4
QUESTION TWO (20 MARKS)
a) The table below shows results of a survey in which 250 respondents were categorized according to level of
education and attitude towards student demonstration at a certain college. Test the hypothesis that the two
criteria of classification are independent. Let
05.0
=
a
{Note d.f. of 4, 5, 6 for
2
?
at 5% level are 7.81,
9.49, 11.07 and 12.59 respectively}
[7 marks]
Education
Attitude
Against
Neutral
For
Total
Less than high school
40
25
5
70
High school
40
20
5
65
Some college
30
15
30
75
College graduate
15
15
10
40
Total
125
75
50
250
b) You are using sample observations from two normal populations with equal variances to evaluate the
difference of their means. You have obtained six observations from the first population and twelve
observations from the second. The means of the populations are denoted
1
µ
and
2
µ
respectively. You
determined the pooled standard deviation as,
0.2
=
pooled
S
. Calculate the length of the 99% confidence
interval for
2
1
µ
µ
-
. {note
921
.2
)
16(
,
12.2
)
16(
01.0
05.0
=
=
t
t
}
[6 marks]
c)You are to perform a one factor analysis of variance to evaluate variations in the average number of
hospital
stays for certain occupational groups. You are give
n:
i
Occupational group
Observed number of hospital
stays
1
Professionals
0, 2, 1
2
Manufacturing
0, 2,
2, 4
3
Construction
3, 4, 2
4
Services
0,
0, 1, 1, 3
The average number of hospital stays for occupational group
i
is denoted by
i
µ
. Calculate the length of
the 95% confidence interval for
1
3
µ
µ
-
[7 marks]
QUESTION THREE (20 MARKS)
a)
State a linear regression equation and the assumptions we make for a valid analyses.
[3 marks]
b)
Construct a linear regression analysis as well as estimating model parameters from the data below.
[4 marks]
Test the hypotheses concerning intercept and slope
as well as construct a 95% confidence interval
[6 marks]
{Use
}
262
.2
)9(
,
306
.2
)8(
,
365
.2
)7(
025
.0
025
.0
025
.0
=
=
=
t
t
t
c)
Estimate the value of Y which should correspond on
average to X=10 and its possible maximum value
[2 marks]
Page
3
of
4
d)
Calculate the Pearson’s correlation coefficients and test the appropriate hypotheses.
[5 marks]
QUESTION FOUR (20 MARKS)
a) You are given the following information about a
one-factor analysis of variance: i) The analysis i
s based on 5
samples of six observations each. (ii) The total sum of squares about the mean is 246. (iii) the F-rat
io is 4.0.
Determine
2
p
S
{where p=pooled}
[6 marks]
b) You are given the following student scores on a
computer examination given at four different unive
rsities:
[9 marks]
University
Test scores
A
50
65
73
82
95
B
60
62
64
C
75
80
92
93
D
80
82
84
86
Using Scheffe’s procedure, determine the length of
longest symmetric 95% confidence interval for the difference
between pairs of means
c) Assume the average cost of laptops owned by KABU staff is Ksh 50,000.00 and standard deviation of K
sh
4,000.00. If repeated samples of size 30 are drawn,
what proportion would yield a mean greater than Ks
h52,000.00?
{use
Z=2.734 is associated with area of 0.4968
}
[5 marks]
QUESTION FIVE (20 MARKS)
a)
We need to quantify the number of students at KABU with computers and hence estimate the additional
computers the university could purchase to supplement this expensive resource. How will you go about
planning for the study by answering the following questions:
i)
How many categories of students do KABU have? Name them.
[2 marks]
ii)
What is the sampling frame? Where would you generally get the sampling frame for KABU
students?
[2 marks]
iii)
Suppose you are told that in the three campuses of KABU there are three categories of
students. Describe how you will go about sampling a
sample of 420 students so that you
achieve gender balance among other minorities. What
sampling design would you use?
[4 marks]
X: 1 2 3 4 5 6 7 8 9
Y: 2 4 1 3 6 8 7 5 9
Page
4
of
4
b)
A new computer company wishes to relocate to Kenya
and would wish to hire you as its sales person. You
are expected to write a brief in the form of concept note describing the level of IT situation in Kenya which
could make the new company make informed decision a
s well as hire you. Write a brief of about 150-200
words describing Kenya’s level of IT distribution.
You are free to suggest a short study to elucidate
this
information, describing methodologies you would use
, given you access some facilitation, to collect ac
curate
information for you prospecting employer.
[
6 marks]
c)
You are required to introduce a new computer company in Central Kenya including and Nairobi city. You
need to do a feasibility study about the marketability of the technology. Explain clearly how you will
go about
the study by telling us what indicators you will consider; what secondary and primary data you will collect;
how you will go about constructing a sampling frame
if it does not exist and finally organize a credible
sample survey.
[6 marks]
More Question Papers