Criteria
Indicator
Code
Weight
Quality of
Education
Alumni of an institution winning Nobel Prizes and
Fields Medals
Alumni 10%
Quality of Faculty
Staff of an institution winning Nobel Prizes and
Fields Medals
Award 20%
Highly Cited Researchers
HiCi
20%
Research Output
Papers published in Nature and Science*
N&S
20%
Papers indexed in Science Citation Index-Expanded
and Social Science Citation Index
PUB
20%
Per Capita
Performance
Per capita academic performance of an institution
PCP
10%
According to the objectives and analytical tools that have been used in the problem formulation, other
datasets have been added.
Checking the integrity of the extracted and sampled data on the individual datasets & cleansing data
(data cleansing)
Missing values
. In datasets, some missing values have been detected in “times data” and expenditure on
education. The missing values have been filled by calculating the median of other data.
Integrating data
Integration of multiple databases, I will create new data taken form shanghai data and Cwurdata in
order to similarity among them. According to that table then will find most influential factors,
Transforming the data and creating new variables
I will transform after integrating and data cleansing steps and will make new variables.
Removing independent variables with low or no predictive power
On the process of data cleansing I removed some column of “times data” (student_staff ratio, male-
famale ratio, income,) “CWURdata” (broad impact) which are no value for my future analysis.
Irrelevant links, comments, descriptions also removed as there is no need for them.
V.
Data mining
Sampling is no need in my analysis just comparing data source is more important.
I have done many types of analysis. Since my dataset is a bit larger, some types of analysis in separate
sheets. The main purpose of these analysis that I have done is to show relationship between variables
and datasets.
First of all, my main data is Central World University rank data which shows different rank of university
accordance of different factors. I try to find firstly how many universities do countries have. Data show
universities rank between 2012 and 2015 years. I want to know how number of universities were
changed over these years
This bar chart show that USA had many universities in 2012 compared to other countries and it followed
by United Kingdom, Japan and Switzerland.
In 2015 the Indicators of data stayed almost same as 2012, however, some countries university became
a fewer, some countries became more. One thing we can understand from this two chart is that USA,
UK, Japan Switzerland countries have more appropriate universities that included in Top 100 world rank
universities across years.
Next, I identify the quality of education among top 10 universities and how they are changed over years.
My analysis shows that Harvard, Stanford, Massachusetts, and Cambridge have high quality on
education compared with other universities. Any eager students who want to get high quality education
these universities are assumed most appropriate.
Nevertheless, we could not rely on only quality of education, there should be one more factor that help
to choose right university for any degree. According to CWUR data I collected top 10 universities to
show rank of alumni employment. Employment is one of the big factor that effects on every student’s
education part.
Back in 2012, Universities’ rank that mentioned above was lower on alumni employment. By 2015, these
indicators have changed significantly.
Harvard university and Massachusetts Institute of Technology have changed its rank from ninth place to
first place, and from seventeenth place to eleventh place respectively. The rank of Stanford university
on alumni employment has changed significantly from eleventh place to second place. Behind these
indicators laid great efforts of universities.
Now let’s look at overall picture of Top 15 universities rank in the world.
Harvard, Stanford, Massachusetts, Oxford universities have maintained their dominance during these
years.
I did some regression analysis in order to find out which factors are more important in identifying world
rank universities not relying on some factors alone. In my data, there are many factors effect on world
rank of universities for example, quality of education, alumni employment, quality of faculty,
publications, influence, citations, patents. I highlighted important parts with yellow and made small
table which show importance of factors in percentage. I have found with the use of a regression analysis
that quality of faculty is a number one factor for identifying world rank. The second purpose of
regression analysis is to find out the relationship between dependent and independent variables. From
regression it’s shown that I have 7 independent variables and result of dependent variable is world rank
and highlighted them as X (x1,x2 ..) and Y respectively. For that I do multiple regression and through this
we can conclude that in the future this relationship between variables will continue as well. Without
these factors world rank is meaningless concept
There is another data Times Higher Education World University Ranking which show many top
universities around the world as CWUR dataset.
According to these data, the percentage of International Student is display that what kind of Universities
are welcome for foreigners.
According to chart, Imperial College of London are more welcome to international students compared
with Harvard university. The lower percentage of International students the harder to get into them, so
students who planned to study abroad should care to develop their strength more.
Moreover, I did line graph to show changes how top Universities world rank as CWUR data. Harvard,
Stanford, Massachusetts Universities are once again dominating the world rank.
Another analysis shows that all universities we assumed are described all data sources differently.
For example, in Cwur and Shanghai data Harvard university is placed in the first rank, but on the Times
data it takes second place. Massachusetts university, Stanford university, Cambridge Oxford and other
universities take different place depending on different factors on sites.
SHanghai rankings uses measures that reflect elements of academic quality, including how many of an
institution’s alumni have won a Nobel prize and how many faculties have won Nobel prizes as a result of
the work done while at the university (in order to prevent rich universities from “buying” Nobel prize
winners). The importance of research outputs is measured by examining where and how often faculty
publish in certain key indicator journals. This highly quantitative methodology produces a ranked list
that represents some very impressive educational and research outcomes, but from a narrow
perspective. The Cwur and Times rankings are more broadly based and include more diverse indicators,
measurements of student numbers, diversity of faculty and students, etc., but are significantly
influenced by an opinion poll/global survey of faculty and other researchers around the world that
focuses on what they know about research strengths of other institutions.
VI.
Conclusion
In this project, Appropriate information and criteria are tried to give for every student’s problem who
wish to pursue higher education in foreign universities but do not have any idea regarding which options
he/she have with respect to his/her overall profile and also have not quite exposed to the procedure
how students get admitted at foreign universities. However, every student should choose universities
based on their criteria whether they want to get high quality of education or take patent for invention or
other expectation. Just relying on rank does not describe whole description of universities or countries.
This project just conclude some important factors which help to identify real rank of universities based
on 3 famous rank system. Final decision depend on student decision and his/her preference who are
going to continue education degree.
Do'stlaringiz bilan baham: |