discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/281585940
CITATIONS
READS
978
1 author:
Therese Yamuna Mahesh
Amal Jyothi College of Engineering
10 PUBLICATIONS 1 CITATION
SEE PROFILE
All content following this page was uploaded by Therese Yamuna Mahesh on 08 September 2015.
The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
Department of Computer Science and Engineering, Bharath University, Chennai, Tamil Nadu, India
Department of Basic Science and Humanities, Amal Jyothi College of Engineering, Kanjirapally, Kerala, India
Department of Computer Science and Engineering, R.M.K. College of Engineering, Chennai, Tamil Nadu, India
E-mail: Yamuna.mahesh@gmail.com, yalenathomas@amaljyothi.ac.in, hodcse@rmk.ac.in
Abstract
One of the requirements of the National Board of Accreditation (NBA) for accrediting a program in an
academic institution in India is the analysis of the attainment of program outcomes by the students. This
paper gives a method for analysing the extent to which the students attain the program outcomes on
completing the program. The method illustrated analyses the attainment level for a single student as well
as a batch. The initial calculations are done using excel and the same database is used to train a neural
network to get the required output. Data set of individual students or batch of students can be given to the
trained network to get the attainment level of program outcomes. The analysis can be done by using the
marks secured by the students or by defining rubrics for the respective courses.
Keywords: Evaluation methodologies, public spaces and computing, country-specific developments,
teaching/learning strategies, applications in subject areas
method of evaluation using the marks scored by
INTRODUCTION
The basic terms to be understood in the
program
educational
objectives,
graduate
attainment
of
Course
etc. [1]
Outcomes,
Program
Communication (PO1).
Engineering (PO2).
complex
problems
in
Electronics
and
(PO4).
(GA3).
Ethics (GA7).
(PO6).
Communication (GA9).
follows:
problems (PO12).
student.
expectation)
are
defined.
An
8) is shown in Table 1.
of
certain
courses
like
project
work,
Table 1: Rubrics for (PO-8) to work as an Individual and also to Promote Team Work.
Measurements
Criteria(CR)
Research
and
Information
Gather
Needs
Improvement
Does not collect
any relevant
information; no
useful
suggestions to
address team's
needs
Taking Responsibility
Valuing
Members
Other
Team
Can do Better
Satisfactory
Collects
information
when prodded;
tries to offer
some ideas, but
not well
developed, and
not clearly
expressed, to
meet team's
needs
Collects basic,
useful
information
related to the
project;
occasionally
offers useful
ideas to meet
the team's needs
Performs
assigned tasks
but needs many
reminders;
attends meetings
regularly but
generally does
not say anything
constructive;
sometimes
expects others to
do his/her work
Usually does
much of the
talking; does not
pay much
attention when
others talk, and
often assumes
their ideas will
not work; no
personal attacks
and put-downs
but sometimes
patronizing;
when others get
through to him,
works
reasonably well
with them
Performs all
assigned tasks;
attends
meetings
regularly and
usually
participates
effectively;
generally
reliable
Generally
listens to others'
points of view;
always uses
appropriate and
respectful
language; tries
to make a
definite effort to
understand
others' ideas;
Exceeds
Expectation
Collects and
presents to the
team a great
deal of relevant
information;
offers welldeveloped and
clearly
expressed ideas
directly related
to the group's
purpose
Performs all
tasks very
effectively;
attends all
meetings and
participates
enthusiastically;
very reliable
Always listens
to others and
their ideas;
helps them
develop their
ideas while
giving them full
credit; always
helps the team
reach a fair
decision.
systems.
structures.
shown below.
CO[4] Understand
DFT
and
convolution
concepts.
defined.
below.
correlation
.
with
Program Outcome
Course
9 10 11
Outcome
CO (4)
CO (5)
CO (1)
CO (2)
CO (3)
the
PO.
It can be seen that the CO of the above course contributes to PO2, PO10 and PO11.
under PO1.
Calculation of PO
An example of PO attainment calculation is
as shown below:
to a single course.
Exceeds
Expectat
ion
<75%
Calculation of CO
CR1, CR2 and CR3 are the three criteria defined
the rubrics.
Table 5: Calculation of CO(k) and the Total CO Attainment Value for a Course Based on Rubrics
Assessment.
Name
C C
R R
1 2
C
R
3
CO
1
1
2
2
1
3
3
1
2
0
0
1
2
2
1
3
3
1
2
0
0
3
6
6
3
9
9
3
6
0
0
a
b
c
d
e
f
g
h
i
j
1
2
2
1
3
3
1
2
0
0
CO1
nor
mali
zed
0.33
0.67
0.67
0.33
1.00
1.00
0.33
0.67
0.00
0.00
C C C
R R R
1 2 3
2
3
2
2
3
1
2
3
3
1
2
3
2
2
3
1
2
3
3
1
2
3
2
2
3
1
2
3
3
1
CO
2
6
9
6
6
9
3
6
9
9
3
CO2
nor
mali
zed
0.67
1.00
0.67
0.67
1.00
0.33
0.67
1.00
1.00
0.33
C
R
1
C
R
2
C
R
3
CO
3
2
3
3
2
3
3
2
3
3
2
2
3
3
2
3
3
2
3
3
2
2
3
3
2
3
3
2
3
3
2
6
9
9
6
9
9
6
9
9
6
CO3
nor
mali
zed
0.67
1.00
1.00
0.67
1.00
1.00
0.67
1.00
1.00
0.67
CO
15
24
21
15
27
21
15
24
18
9
CO
nor
mali
zed
0.56
0.89
0.78
0.56
1.00
0.78
0.56
0.89
0.67
0.33
CO1=0.5
CO2=0.73
CO3=0.87
CO=0.7
departments.
The graphical representation of the above attainment level can also be represented as shown in Figure 1
below:
Table 6: Calculation of CO(k) and the Total CO Attainment Value for a Course based on Marks Secured.
The calculation of CO and PO can be easily implemented using neural networks [2]. Normalisation is not
required in the case of neural Network based evaluation. The weights are assigned for CO and PO
calculation according to the table shown below. Prior knowledge of the PO under which the CO is grouped
is essential for proper classification. The below example shows all the COs (CO1, CO2, CO3) being
grouped under a single PO. The classification is done using multilayer perceptron. The weights assigned for
classification are shown in the Table 7 below.
Table 7: Calculation of CO(k) and the Total CO Attainment Value for a Course based on Marks Secured.
Measureme
nt
Basic Score
Score for
CO(k)=Basic
score *3
Score for PO
under which
the CO is
grouped
10
Needs
Improv
ement
0
Can Do
Better
Satisfactor
y
Exceeds
Expectation
12
16
20
Results of the classification using neural networks (multilayer perceptron) are shown in Figure 2 and Figure
3 shown below:
11
12
Fig. 3: Graphical Classification for CO1, CO2, CO3, CO Values and PO.
Test Data
The trained neural network is used for testing. The
13
CONCLUSION
levels.
14
REFERENCES
2.
3.
Shunmuganathan, Measurement of
2015; 5(2).
15