Classifications and rankings of U.S. universities for the years 2017--2020.
data(qswur_usa)Format
A tibble of 13 variables on 612 cases:
- year
year of rankings
- institution
institution of higher learning
- size
size category of institution
- focus
subject range of institution
- res
research intensity of institution
- age
age classification of institution
- status
status of institution
- rk_academic
rank by academic reputation
- rk_employer
rank by employer reputation
- rk_ratio
rank by faculty--student ratio
- rk_citations
rank by citations per faculty
- rk_intl_faculty
rank by international faculty ratio
- rk_intl_students
rank by international student ratio
Source
Quacquarelli Symonds (2021).
Details
Ranking data were obtained from the public QS website.
References
Quacquarelli Symonds (2021) "University Rankings". TopUniversities.com https://www.topuniversities.com/university-rankings.
Examples
# subset QS data to rank variables
head(qswur_usa)
#> # A tibble: 6 × 13
#> year institu…¹ size focus res age status rk_ac…² rk_em…³ rk_ra…⁴ rk_ci…⁵
#> <int> <chr> <fct> <fct> <fct> <int> <chr> <int> <int> <int> <int>
#> 1 2017 MASSACHU… M CO VH 5 B 6 4 12 10
#> 2 2017 STANFORD… L FC VH 5 A 5 5 20 14
#> 3 2017 HARVARD … L FC VH 5 B 1 1 42 8
#> 4 2017 CALIFORN… S CO VH 5 B 23 90 3 4
#> 5 2017 UNIVERSI… L FC VH 5 B 13 47 54 49
#> 6 2017 PRINCETO… M CO VH 5 B 10 32 115 3
#> # … with 2 more variables: rk_intl_faculty <int>, rk_intl_students <int>, and
#> # abbreviated variable names ¹institution, ²rk_academic, ³rk_employer,
#> # ⁴rk_ratio, ⁵rk_citations
qs_ranks <- subset(
qswur_usa,
complete.cases(qswur_usa),
select = 8:13
)
# calculate Kendall correlation matrix
qs_cor <- cor(qs_ranks, method = "kendall")
# calculate eigendecomposition
qs_eigen <- eigen_ord(qs_cor)
# view correlations as cosines of biplot vectors
biplot(x = qs_eigen$vectors, y = qs_eigen$vectors, col = c(NA, "black"))
