Corporate Big Data Analysis
Final Project
Topic: World Universities ranking analysis
Name: Maftuna
ID: 12215034
I.
Introduction/ Overview
From the medieval age university quality has been widely known by writings of students, greats
names of professors and academic reputation of universities. But since the last quarter of the
twentieth century, university ranking systems have widely observed international universities
and updated world universities ranking annually.
During recent years, university rankings have gained a considerable importance not only among
the academia but also amongst students, parents, industry and businesses. Also, I want to see
the impact of these rankings and how they influence the stakeholders, which are winners of each
of these global university rankings, from want countries or regions. Global university rankings
tend to focus more on the research area and less on teaching and learning environment. After
the results of these rankings and others, all universities whether small or large, can improve
practices that will make them stronger. For the contemporary society it is also important for a
university to be able to innovate and help industry and businesses with consultancy and
innovations.
The Academic Ranking of World Universities, also known as the Shanghai Ranking, is an equally
influential ranking. It was founded in China in 2003 and has been criticized for focusing on raw
research power and for undermining humanities and quality of instruction. The Center for World
University Rankings, is a less well known listing that comes from Saudi Arabia, it was founded in
2012. To further extend my analyses, I’ve also included two sets of supplementary data. The
purpose of this project is to analyze Top universities in the world depending on various factors,
to make relation among data’s and predict in the future what kind of universities could be top
and what kind of factors could influence to be that.
To further extension, supplementary dataset has included. It contains information about public
and private direct expenditure on education across the nations. This data comes from the
National Center for education statistics.
II.
Problem Formulation
It is really time –consuming and boring process that students go through when analyzing and choosing
right universities for themselves to continue study. The process might start with a search for a particular
interested university from which several links to different universities are returned. The student typically
visits each website to check different factors such as Programs Offered, Cost of Living, Tution fees,
Quality of Educational Degree etc. This could involve considering alternate universities from an online
catalog, program and courses availability, location options etc. After all relevant information is gathered,
the student will then enroll into university for getting himself admitted to it using a educational
qualification as a gateway. It should be noted whenever the student want to get admission into any
university through online from any source he/she does visit the many Universities websites for getting
the desirable interest fields and information about the university which will help the student in his
decision making process whether to opt for that university or not. Like this student surf lots of time in
visiting of universities websites for getting admission into desired university or affiliated program. The
student not only surfs lots of time in visiting of university sites, and affiliated programs, but also he/she
suffers from limited option to choose the programs offered.
1.
Specific objectives:
Find out which of these categories has most impact on national rank
Which of these categories the highest and least influence on the reputation of universities
Finding correlation among attributes, such as national rank of university and quality of
education, alumni employment, quality of faculty, publication etc.
Observing similarity of Shanghai ranking and CWUR ranking among universities.
Find a number of top 500 universities among countries’ public and private expenditure on
education and number of people in this countries
Exploring highest percentage of international students among countries and universities
Prediction of further top countries within the highest rank of universities.
Creating new table among other column.2
Translation into analytical methods: For the project many analytical methods have been used for
problem translation:
Correlation for finding relationship between national rank and university and quality of
education, alumni employment, quality of faculty, publication and other sub-categories
Regression analysis
2.
Translation into analytical methods.
Statistical analysis
: making histogram to show the similarity of two data set and exploring highest
percentage of international student. And making table for school and country.
Predictive modeling:
In my data the major factors will be analyzed and identified which factor has
high impact on world rank of universities.
Case: past performance of universities’ reputation
Input: circumstance of universities between 2011 and 2016
Target: performance rank, influential factors, country level during 2011-2016
Action: identifying future reputation and pattern of performance.
The methodology for this comparative analysis contains:
o
Selection of university rankings TOP-100
o
Ranking Criteria and Weights for each selected international ranking
o
Definition of Indicators
o
Data Sources
o
Top 15 best universities – similitudes and differences between the selected rankings
o
Comparative analysis, regarding the ranking criteria and weights, top universities,
statistics by regions and by countries.
III.
Data explanation
In my data I took 3 different datasets from kaggle.com The
Times Higher Education World University
Ranking
(later called Timesdata) is widely regarded as one of the most influential and widely observed
university measures. Founded in the United Kingdom in 2010, it has been criticized for its
commercialization and for undermining non-English-instructing institutions.
The
Academic Ranking of World Universities
, (later called Shanghai data) also known as the Shanghai
Ranking, is an equally influential ranking. It was founded in China in 2003 and has been criticized for
focusing on raw research power and for undermining humanities and quality of instruction.
The
Center for World University Rankings
, (later called Cwur data) is comes from Saudi Arabia, it was
founded in 2012.
Variables.
In Cwur data Universities rank is given from 2012 to 2015 with other ranks such as national rank, quality
of education, alumni employment, quality of faculty publications, influence, citations, patents, score
ranks.
In Timesdata all universities variables are given in the form of percentage in terms of teaching,
international, research, citations, international students. Even number of students also given from 2011
to 2016
In Shanghai data national rank alumni in percentages award, hici, ns, pub, pcp, are given from 2005 to
2015.
IV.
Data preparation
Extracting and sampling data from the operational database of the database or data warehouse
(kaggle.com)
Data is extracted from three official websites and sampled only data from 2011 to 2016, sorted 3
datasets. The original data has educational attainment dataset which taken from UNESCO institute for
statistics and barro-lee dataset and Shanghai ranking and CWUR ranking among universities. Among
them Shanghai and CWUR ranking datasets are taken for analyzing.
Do'stlaringiz bilan baham: |