1.

Introduction
This chapter describes student performance in the Bundelkhand University,Jhansi and
compares it to that of three peer districts that have similar student populations but higher
academic achievement. It highlights the findings from an extensive analysis of student scores
over the past six years on state proficiency exams and English fluency assessments. This
description of proficiency rates, achievement gaps among student subgroups, and trends over
time shows that student performance remains far below state standards and own targets, and
substantial achievement gaps have persisted. In addition, this chapter describes the factors that
peer districts attribute to their success. These are offered as recommendations to assist in taking
dramatic steps to significantly improve student academic achievement.

Findings included in this chapter summarize two separate research reports regarding
student performance in Bundelkhand University,Jhansi. The report, Analysis of Student
Performance, provides detailed analyses of student proficiency rates and English fluency results,
broken down by student subgroups and grade levels. The Comparative Analysis of Academic
Performance describes how the three peer districts were selected and compares their student
performance and trends over time with those of Bundelkhand University for reading and math,
limited English proficient (LEP) students, Advanced Placement participation and test scores,
PSAT scores, and graduation and dropout rates. It also provides a detailed description of peer
district efforts to improve their students’ performance.
 2.Summary of Key Findings and Recommendations
The analysis of student performance data and the experiences of peer districts clearly
justify the Bundelkand University Jhansi recent decision to take dramatic steps to significantly
improve student achievement.Phase I: Preliminary Reforms Report and many initiatives were
underway before this study commenced. The review team endorses the direction of the district’s
new leadership, and believes that the recommendations contained in this report will help support
a new era of educational reform at Bundelkhand University.

2.1 Curriculum consistency and alignment

A common success factor of the peer districts was the consistency and alignment
across its schools in curriculum and programs offered. As stated by one interview: “We were
spending millions and getting very inconsistent result. It is a fiduciary responsibility [to select a
program] and go with it—implement it with fidelity, and give it three to five years to evaluate it
over time.” Even in the districts that were more decentralized, it was their structure of networks
and consistent communication that helped to keep schools and teachers moving in the same
direction.

2.1.1 Focused professional development and support.

Considered critical to peer districts improved performance high quality professional
development is offered through ongoing sessions, coaching, support from experts, and resources
provided in-person and on-line. Professional development is focused on specific programs and
student populations, including LEP students. Bundelkhand University was found to have
overlapping and sometimes conflicting professional development coming from multiple,
uncoordinated sources.

2.1.2 Intensive attention to particular subject areas and student subgroups.

The analysis of Bundelkhand University data indicates that achievement in science is
particularly low and specific subgroups are having the most difficulty attaining proficiency status
on state assessments. Redoubled efforts to support their academic achievement is merited for
students.Although the achievement gap between students has narrowed somewhat over time, it
is still substantial. Given that more than one-third of students who took the CRT are either non-
limited-English.
LEP students. Additional attention to the needs of the LEP student population is necessary,
especially to factors that peer districts report have contributed to their success.Intense
professional development: In the peer districts, teachers who instruct LEP students receive
extensive professional development, both internal to the district and through state
certification/endorsement specifically related to this student population .

Consistent curriculum and oversight of implementation: Peer districts ensure that schools have a
consistent curriculum and supplemental materials available to all LEP students. Monitoring also
takes place to ensure that these programs are implemented as planned and are moving students
towards English fluency.

Students in grades kindergarten through two: Data analysis revealed that these students are the
least likely of all grade levels to be fluent in English . The proportion of children in grades K–2
who are fluent in English is much smaller in than in the peer districts. These districts cite their
intensive intervention programs for young LEP students as a factor in their overall success.

Students with disabilities. Generally less than one-fourth of grade 3–8 students with IEPs are
proficient in math, reading, and science. For high school, math and science proficiency rates are
15 percent or lower.

Retained high school students. The cohort analysis of the data revealed a remarkable group of
high school students who persisted in retaking the reading and math exams even after they were
retained in grade 10 for one or two years. More than 3,000 students took the math and reading
tests in their second tenth grade year and more than 100 took them again in their third tenth grade
year. Such perseverance could be acknowledged and rewarded with intensive assistance to help
them pass the exams.Preschool education. In examining the data used to select the peer district.
Successful high school completion and college/career readiness. A consistent theme in the peer
districts is the effort to engage students early on in their high school education. By focusing on grade
9 students, dropout rates are lower and students are better prepared for college and careers. As one
interviewee stated, “If we lose them in the ninth grade, we lose [them] in graduation.” Peer districts
have a variety of student engagement, mentoring, and credit recovery programs that begin with
identifying at-risk students using an early warning system would benefit by adopting some of these
practices:

Ninth grade monitoring: Given that there is no Nevada state assessment for grade 9 students
could consider analyzing interim assessment and Criterion Referenced Test (CRT) data from
grade 8 students as one step toward an early warning system. Analysis of CRT data indicates that
the math proficiency rate of grade 8 students is consistently lower than that of other grades and
recent results for reading and science show that less than half the students are proficient. In
addition, monitoring the proportion of grade 9 students who move on to grade 10 could provide
another measure of student engagement in high school.

Positive alternative environments: In all of the peer districts, staff emphasized the importance of
addressing students’ needs through choices and a variety of settings. Whether it was through
online learning, small learning communities, or specialized magnet school options, providing
alternative settings can help motivate students who might otherwise dropout from the traditional
high school setting. With the addition of support from mentors and community members, more
students can reach graduation in these alternative settings if they are seen as positive
environments instead of as a punishment for misbehavior.

Table 2.1 I.E.T Campus rated In Need of Improvement by Level, 2017–18

Total Number of Number In Need of Percent In Need of
Type of School
Schools Improvement Improvement
Total Labs 219 91 42%
Lectures 77 40 52%
Attendance 71 31 44%
Total 367 162 44%
.
2.1.3 Grades 3–8 Student Performance

To provide an overview of student performance in the elementary and middle school grades, the
CRT proficiency rates of students in grades 3–8 have been combined. As shown in Figure 2.1,
the overall proportion of I.E.T students scoring proficient in attendance across the years ranged
from 51 to 67 percent. For lecture, the range was 46 to 63 percent, and for labs 48 to 57 percent.
Although the tests have been revised in recent years, which resulted in some fluctuation in
scores, the overall finding is that many students are not meeting the standard for performance,
which is not rigorous.

Figure 2.1. CRT proficiency rates, grades 3–8, by subject and year,2012-13 through 2017-18
100%

80%
67%
59% 61% 63% 60% 63%
57% 55% 55% 56% 57%
51%
60% 51% 46% 48% 53% 49% 50%
40%

20%

0%
Attendance Labs Lectures

2012-13 2013-14 2014-15 2015-16 2016-17 2017-18

Note: The CRT attendance and labs tests are given each year; the lecture test is given in
grades 5 and 8. The attendance and lectures tests were revised in 2012–13, and the reading
test was revised in 2017–18.
Note: Sample size = 843,673 (attendance, all years combined); 843,789 (labs, all years); 278,561
(lectures, all years).
For the most recent year (2017–18), of students in grades 3 through 8:
67 percent were proficient in attendance.
56 percent were proficient in labs.
50 percent were proficient in lectures attended.
Furthermore, across the grade levels, attendance proficiency rates have been lowest in grade 8;
labs proficiency rates have been lowest in grade 5 (until test revision in 2017–18); and lectures
proficiency rates have been consistently low in both grade 5 and grade 8.

2.1.4 Achievement Gaps

Large gaps in academic performance are evident for racial/ethnic groups and for students
eligible for free and reduced-price learning (FRPL), as well as those designated as LEP and
those who have an Individualized Education Program (IEP).

2.1.5 Free or Reduced-Price Learning

Nearly half (45 percent) of I.E.T students taking the CRT were eligible for free and reduced-
price lunch (FRPL) in 2008–09, and their performance lagged behind that of their more
economically advantaged peers. Figure 2.4 displays both the proficiency rate for FRPL students
and the achievement gap between them and their non-eligible peers across the years.For the
most recent year of available data (2012–13), the FRPL-Not FRPL gap was:
 18 percentage points in attendance
 21 percentage points in labs.
 24 percentage points in lecture.
This gap appears to have narrowed somewhat in both math and reading and has fluctuated for
science across the years. Although there has been progress, the gaps are still substantial for
student.A revealing feature of the cohort analysis is the number of students who persist at
staying in campus and taking the test, even though they are not succeeding there are 2,135
students who were retained in grade 10 who took the math exam again in their second grade 10
year.

There are 74 students who were retained a third time in grade 10 and took the exam yet again.
It is evident that this group is in need of intensive assistance to help them pass these exams.
Table 2.2. Total number of students who took the at least once, by grade level, by eligible
year
Year 1 Year 2 Year 3

Grade 10 39,518 2,135 74

Grade 11 13,394 432
Grade 12 8,527
Total 39,518 15,529 9,033

3.0 LITERATURE REVIEW

3.1 Study of Existing Systems Similar to proposed System:

In order to review similar works by existing author’s, a background study is done to

perform analysis of student performance. Here we have chosen three existing systems as study of
these helped us to propose our system.

a) Student Performance Analysis System (SPAS)

We have proposed a framework named Student Performance Analysis System (SPAS)

which is able to predict the students’ performance in course “ Compiler and Design”, which in turns
assists the lecturers from Information System department to identify students that are predicted to
have bad performance in course “ Compiler and Design”. The proposed system offers student
performance prediction through the rules generated via data mining technique. The data mining
technique used in this project is classification, which classifies the students based on students’
grade.

b) Extraction and Analysis of Faculty Performance of Management Discipline from Student

Feedback Using Clustering and Association Rule Mining Techniques
We have proposed a system that deals with the extraction and analysis of faculty
performance of management discipline from student feedback using clustering and association rule
mining techniques. The performance of a faculty of any school or an Institute has been found to be
dependent on a number of parameters broadly ranging from the individual’s qualifications,
experience, level of commitment, research activities undertaken to institutional support, financial
feasibility, top management’s support etc. The parameters which are crucial for assessment of
faculty performance range across various verticals, but the paper discusses and covers the
performance of faculties based strictly on students feedback only. The other assessors of faculty
performance being the Management Body which could be a private body or a Government unit, self
and peer faculties of the organization or the University. The parameters act as performance
indicators for an individual and group and subsequently can impact on the decision making of the
individual and also the stakeholders. The idea proposed in this paper is to perform extraction and
analysis of faculty performance using techniques of Data Mining.

Apriori algorithm has been used in this system to extract association rules as follows:
 1. Communication and Presentation Skills, Punctuality And Regularity, Knowledge beyond
Syllabus ==> Counseling Personal Guidance conf : ( 0.97)

2. Teaching Ability with use of new Teaching Aids ==> Subject Knowledge conf:(0.97)

3. Motivation Self Students, Communication Presentation Skills, Punctuality And Regularity

==> Counseling and Personal Guidance conf:(0.97)

4. Class Control, Punctuality And Regularity = 5968 ==> Counseling Personal Guidance =
5933 conf:(0.96)

5. Communication and Presentation Skills, ClassControl ==> Counseling and Personal

Guidance conf:(0.96)

6. Class Control, Knowledge beyond Syllabus ==> Counseling and Personal Guidance conf :
(0.96)

7. Communication and Presentation Skills, Knowledge beyond Syllabus ==> Counseling and
Personal Guidance conf:0.96)

8. Communication and Presentation Skills, Counseling and Personal Guidance, Knowledge

beyond Syllabus ==> Punctuality And Regularity conf:(0.96)

9. Motivation Self and Students, Punctuality And Regularity, Knowledge beyond Syllabus ==>
Counseling and Personal Guidance conf:(0.96)

10. Communication Presentation Skills, Punctuality And Regularity ==> Counseling and
Personal Guidance conf: (0.95)

11. Teaching Ability with use of new Teaching Aids ==> Subject Knowledge conf:(1.00)

c) Graduate Students Dataset and Recruitment Preprocessing :

We have written a progress report that categorizes the students into grade order in all their
education studies and it helps in interview situation. This study explores the socio-demographic
variables (age,gender, name, lower class grade, higher class grade, degree proficiency and extra
knowledge or skill, etc). It examines to what extent these factors helps to categorize students in
rank order to arrange for the recruitment process. Due to this, all students get benefitted and it also
reduces the short listings. Here, clustering, association rules, classification and outlier detection has
been used to evaluate the students’ performance.

The data mining techniques to the data are:

i) Association Rule
Association rule learning is a popular and well researched method for discovering
interesting relations between variables in large databases. Association rules are usually required to
satisfy a user-specified minimum support and a user-specified minimum confidence at the same
time. Association rule generation is usually split up into two separate steps:

First, minimum support is applied to find all frequent item sets in a database. Second, these
frequent item sets and the minimum confidence constraint are used to form rules. Finding all
frequent item sets in a database is difficult since it involves searching all possible item sets (item
combinations). The set of possible item sets is the power set over

I and has size 2 n -1(excluding the empty set which is not a valid item set).
[Lower_class_grade=P Higher_class_grade=Goo [Grade=Averag (Support 0.19,
oor, d]→ e] :
Confidence:0.757) Higher_class_grade=Poor]→ [Grade=Average]
[Lower_class_grade=Good, (Support: 0.19,
Confidence:0.757) Higher_class_grade=Poor] → [Grade=Average]
[Lower_class_grade=Good, (Support: 0.105,
Confidence: 0.731)

ii) Classification
 Classification is the process of finding a model that describes and distinguishes data classes or
concepts, for the purpose of being able to use the model to predict the class of objects whose class
label is unknown. The derived model is based on the analysis of a set of training data. It is
important to know that classification rules are different than rules generated from association.
Association rules are characteristic rules, but classification rules are prediction rules.

If lower_class_grade=good and Higher_class_grade=good then Topper, If Lower_class_grade=poor

and Higher_class grade=good then Average, If Lower_class_grade=poor Higher_class_grade=poor
then Below Average.

B. Discussion with Top Management of our College:

In order to gather requirements for our system we presented our idea to the top management of
our college to seek their views and suggestions for our system. We designed our system according
to the requirements we gathered from top management and the suggestions they gave us.

4.0 PROPOSED SYSTEM

The features of the existing system that we studied are employed in the design of the proposed
system. These features include an user profile creator to provide user interface, user login, student
performance analyzer, score card generator, student performance credit card, student development
card, achieved credit, passing criteria card and semester wise student performance attribute card
.Rules for passing criteria were decided after discussion with top management. Students will be
grouped into categories like distinction, first class, higher second class, second class and pass class
based on passing criteria.
 5.0. METHODOLOGY
Several phases that will be followed during development of proposed system are:

Fig.1 Phases for system Design and Development

A. Problem and Data Understanding

For the success of Student Performance Analysis system the problem and data understanding
plays a very crucial role. In order to define the project goals and objectives, problems and data
understanding is identified. The problems of existing system are analyzed for its effectiveness and
efficiency in terms of functionality. After identifying the problem statement the solution to solve it
is identified by studying various papers and deciding which possible algorithm or method suits best
for our system that we are proposing.

Moreover an interview with top management of our college made our ideas clear in designing the
system. Besides other similar systems are studied and analyzed for its features, strengths and
weakness and opportunities for proposed system.
T ABLE 1. ATTRIBUTES OF DATASET F OR ACADEMICS
OF DATASET F OR PROJECT PERFORMANCE AND T ABLE 2. ATTRIBUTES
RESEARCH

T ABLE 3. ATTRIBUTES OF DATASET F OR STUDENT SELF TABLE 4. ATTRIBUTES

OF DATASET F OR EXTRA SELF DEVELOPMENT
CURRICULAR ACTIVITIES
B. System Analysis and Design

Overall flow of system is planned, analyzed and designed in this phase. The requirements gathered
are analyzed. In order to understand the input, processes and output of system, the data flow diagram
is used. All levels of data flow diagram are drawn. Class diagram is drawn for better understanding
of the methods that will be used during implementation of modules of the proposed system