Case Study: Welcome to University
R Junkies
24 Ekim 2017
Introduction
In this case study we are going to explore university entrance examinations (YGS/LYS) dataset from 2017. Dataset consists of undergraduate programs offered in 2017. Each program offers an availability (i.e. quota). Then students get placed according to their lists and their scores. Each program is filled with the students ranked by their scores until placements are equal to availability. Student placed to a a program with the highest score forms the maximum score of that program and the last student to be placed forms the minimum score.
Load Data
# Download dataset from GitHub (do it only once)
download.file("https://mef-bda503.github.io/files/osym_data_2017_v2.RData", "osym_data_2017.RData")
# Load tidyverse package
library(tidyverse)
# Load the data
load("osym_data_2017.RData")
#Set locale
Sys.setlocale (locale="Turkish_Turkey.1254")## [1] "LC_COLLATE=Turkish_Turkey.1254;LC_CTYPE=Turkish_Turkey.1254;LC_MONETARY=Turkish_Turkey.1254;LC_NUMERIC=C;LC_TIME=Turkish_Turkey.1254"Number of University Departmens
The table below shows the number of university departments in Istanbul.
university_departments <- osym_data_2017 %>%
group_by(University_Name=university_name) %>%
filter(city=='İSTANBUL' & substr(program_id,0,1)=="2") %>%
summarise(Departments=n()) %>%
arrange(desc(Departments))
university_departments## # A tibble: 41 x 2
## University_Name Departments
## <chr> <int>
## 1 ÝSTANBUL GELÝÞÝM ÜNÝVERSÝTESÝ 212
## 2 OKAN ÜNÝVERSÝTESÝ 179
## 3 BEYKENT ÜNÝVERSÝTESÝ 176
## 4 YEDÝTEPE ÜNÝVERSÝTESÝ 172
## 5 ÝSTANBUL AYDIN ÜNÝVERSÝTESÝ 161
## 6 ÝSTANBUL MEDÝPOL ÜNÝVERSÝTESÝ 155
## 7 ÝSTANBUL AREL ÜNÝVERSÝTESÝ 141
## 8 BAHÇEÞEHÝR ÜNÝVERSÝTESÝ 128
## 9 MALTEPE ÜNÝVERSÝTESÝ 123
## 10 ÝSTANBUL BÝLGÝ ÜNÝVERSÝTESÝ 115
## # ... with 31 more rowsLet’s visualize this data in a barchart.
ggplot(university_departments, aes(x=reorder(University_Name,-Departments), y=Departments)) +
geom_bar(stat = "identity", aes(fill=university_departments$University_Name=='MEF ÜNİVERSİTESİ')) +
labs(title="# of University Departments in Istanbul",x="University",y="# of Deparments",fill="") +
theme (axis.text.x=element_text (angle=-90,vjust=0.5, hjust=0)) +
scale_fill_manual(values = c('#707070', 'red'),guide=FALSE)
Maximum Scores of Universities in Istanbul
maximum_scores <- osym_data_2017 %>%
select(University_Name=university_name, max_score, city, program_id) %>%
filter(city=='İSTANBUL' & substr(program_id,0,1)=="2") %>%
group_by(University_Name) %>%
summarise(Max_Score=max(max_score)) %>%
arrange(desc(Max_Score))
maximum_scores## # A tibble: 41 x 2
## University_Name Max_Score
## <chr> <dbl>
## 1 KOÇ ÜNÝVERSÝTESÝ 569.1112
## 2 ÝSTANBUL MEDÝPOL ÜNÝVERSÝTESÝ 559.4780
## 3 ACIBADEM MEHMET ALÝ AYDINLAR ÜNÝVERSÝTESÝ 542.3482
## 4 SABANCI ÜNÝVERSÝTESÝ 538.7725
## 5 YEDÝTEPE ÜNÝVERSÝTESÝ 531.3691
## 6 BAHÇEÞEHÝR ÜNÝVERSÝTESÝ 530.4845
## 7 BÝRUNÝ ÜNÝVERSÝTESÝ 528.6418
## 8 ÝSTANBUL AYDIN ÜNÝVERSÝTESÝ 525.5809
## 9 ÝSTANBUL 29 MAYIS ÜNÝVERSÝTESÝ 522.9686
## 10 ÖZYEÐÝN ÜNÝVERSÝTESÝ 522.6578
## # ... with 31 more rowsLet’s visualize this data in a barchart.
ggplot(maximum_scores, aes(x=reorder(University_Name,-Max_Score), y=Max_Score)) +
geom_bar(stat = "identity", aes(fill=University_Name=='MEF ÜNİVERSİTESİ')) +
labs(title="Maximum Score of Each University",x="University",y="Maximum score",fill="") +
theme (axis.text.x=element_text (angle=-90,vjust=0.5,hjust=0)) +
scale_fill_manual(values = c('#707070', 'red'),guide=FALSE)
University Department Quotas in Istanbul
department_quota <- osym_data_2017 %>%
select(University_Name=university_name,general_quota,city,program_id) %>%
filter(city=='İSTANBUL' & substr(program_id,0,1)=="2") %>%
group_by(University_Name) %>%
summarise(General_Quota=sum(as.integer(general_quota))) %>%
arrange(desc(General_Quota))
department_quota## # A tibble: 41 x 2
## University_Name General_Quota
## <chr> <int>
## 1 ÝSTANBUL MEDÝPOL ÜNÝVERSÝTESÝ 4495
## 2 BEYKENT ÜNÝVERSÝTESÝ 3974
## 3 ÝSTANBUL GELÝÞÝM ÜNÝVERSÝTESÝ 3950
## 4 ÝSTANBUL AYDIN ÜNÝVERSÝTESÝ 3625
## 5 YEDÝTEPE ÜNÝVERSÝTESÝ 3569
## 6 BAHÇEÞEHÝR ÜNÝVERSÝTESÝ 2843
## 7 OKAN ÜNÝVERSÝTESÝ 2506
## 8 ÝSTANBUL KÜLTÜR ÜNÝVERSÝTESÝ 2284
## 9 ÜSKÜDAR ÜNÝVERSÝTESÝ 2150
## 10 MALTEPE ÜNÝVERSÝTESÝ 2107
## # ... with 31 more rowsLet’s visualize this data in a barchart.
ggplot(department_quota, aes(x=reorder(University_Name,-General_Quota), y=General_Quota)) +
geom_bar(stat = "identity", aes(fill=University_Name=='MEF ÜNİVERSİTESİ')) +
labs(title="University Department Quotas in Istanbul",x="University",y="Quota",fill="") +
theme (axis.text.x=element_text (angle=-90,vjust=0.5,hjust=0)) +
scale_fill_manual(values = c('#707070', 'red'),guide=FALSE)
Listing Programs in MEF by scores
maximum_scores <- osym_data_2017 %>%
select(university_name,program_name,max_score) %>%
filter(university_name=='MEF ÜNİVERSİTESİ') %>%
group_by(program_name) %>%
summarise(Max_Score=max(max_score)) %>%
arrange(desc(Max_Score))
maximum_scores## # A tibble: 44 x 2
## program_name
## <chr>
## 1 Hukuk (Tam Burslu)
## 2 Ýngilizce Öðretmenliði (Ýngilizce) (Tam Burslu)
## 3 Psikoloji (Ýngilizce) (Tam Burslu)
## 4 Ýngilizce Öðretmenliði (Ýngilizce) (%75 Burslu)
## 5 Bilgisayar Mühendisliði (Ýngilizce) (Tam Burslu)
## 6 Endüstri Mühendisliði (Ýngilizce) (Tam Burslu)
## 7 Makine Mühendisliði (Ýngilizce) (Tam Burslu)
## 8 Mimarlýk (Ýngilizce) (Tam Burslu)
## 9 Rehberlik ve Psikolojik Danýþmanlýk (Ýngilizce) (Tam Burslu)
## 10 Siyaset Bilimi ve Uluslararasý Ýliþkiler (Ýngilizce) (Tam Burslu)
## # ... with 34 more rows, and 1 more variables: Max_Score <dbl>Let’s find Score in top %10 percentile
quantile_scores <- osym_data_2017 %>%
select(city,max_score,program_id)%>%
filter(city=='İSTANBUL' & substr(program_id,0,1)=="2") %>%
summarise(Quantile_Score=quantile(max_score,c(.90))) %>%
arrange(desc(Quantile_Score))
q10 <- quantile_scores
quantile_scores## # A tibble: 1 x 1
## Quantile_Score
## <dbl>
## 1 443.8652Let’s visualize this data in a barchart.
ggplot(maximum_scores, aes(x=reorder(program_name,-Max_Score), y=Max_Score)) +
geom_bar(stat = "identity", aes(fill=Max_Score>q10[[1]])) +
labs(title="Maximum Score of Programs on MEF University",x="Programs",y="Maximum score",fill="") +
theme (axis.text.x=element_text (angle=-90,vjust=0.5,hjust=0)) +
scale_fill_manual(values = c('#707070',"red"),guide=FALSE)
Conclusion
Our findings and suggestions to MEF University Management:
We compared MEF University with other Private/Foundation universities in Istanbul.
- Less number of departmants compared to others => Should increase for more students by establishing new programs
- Best students seems that they don’t prefer MEF => Should incentivize best students
- Less department quota compared to others => Should increase for more students by increasing department quotas
- Following departments stands for the 10% percentile among other universities: Hukuk (Tam Burslu) İngilizce Öğretmenliği Psikoloji (İngilizce) Tam Burslu İngilizce Öğretmenliği (%75 Burslu) Bilgisayar Mühendisliği (İngilizce, Tam Burslu)
These departmants mostly belong to “TM”. Should increase performance in “MF”.