Country

Indicator

Value type

Values present

Persistence to last grade of primary, total (% of cohort)

Source: worldbank.org, 03.09.2025

Year: 2020

Country	Value	Value change, %	Rank
Aruba	85.5	-3.04%	76
Albania	91.9	+2.8%	70
Andorra	66.1	-1.89%	89
Argentina	96.5	+2.22%	52
Armenia	99.6	+0.257%	12
Austria	99.7	+0.0944%	9
Azerbaijan	94.2	-1.24%	66
Benin	54.1	+39.6%	94
Burkina Faso	62.1	+14%	90
Bulgaria	97.5	+1.78%	43
Bosnia & Herzegovina	95	+1.51%	64
Belarus	98.3	+0.00728%	37
Belize	96.3	+6.78%	53
Barbados	94.8	+1.55%	65
Switzerland	99.4	-0.179%	18
Chile	99.6	+0.367%	14
China	96.8	-0.0624%	49
Côte d’Ivoire	70.1	-10.2%	87
Congo - Kinshasa	73.3	+62%	84
Colombia	97.3	+1.81%	47
Cape Verde	96.8	+9.06%	50
Costa Rica	97.4	+4.34%	45
Cuba	97.3	+1.36%	48
Cyprus	98.1	+0.863%	39
Czechia	99.5	-0.0781%	16
Djibouti	89.5	+1.36%	71
Denmark	99.7	+0.0095%	10
Dominican Republic	56.2	-34%	93
Algeria	89	-2.83%	72
Ecuador	98.5	+2.11%	32
Egypt	99.8	+0.111%	5
Spain	99.9	+0.105%	1
Estonia	99.7	-0.0532%	8
Finland	99.8	+0.266%	4
United Kingdom	99.3	-0.522%	21
Georgia	99.2	+0.903%	25
Gambia	78.2	-1.86%	82
Greece	99.5	+0.00624%	17
Guatemala	93.7	+22.5%	68
Honduras	72.7	-5.26%	85
Croatia	98.5	+0.0211%	34
Hungary	98.9	+0.0998%	28
India	99.6	+1.15%	11
Israel	98.8	-0.189%	29
Italy	99.2	-0.705%	26
Jordan	97.4	+4.02%	44
Kyrgyzstan	97.7	-0.0273%	42
Kiribati	83	+5.14%	80
South Korea	99.7	+0.551%	6
Kuwait	85.7	-8.74%	75
Lebanon	86.1	+1.52%	74
Liechtenstein	96.3	+6.81%	54
Lithuania	99.5	+0.576%	15
Luxembourg	78.4	-5.54%	81
Latvia	98.5	+1.59%	33
Macao SAR China	99.2	+0.149%	24
Morocco	92	-3.24%	69
Monaco	93.8	+14.6%	67
Moldova	95.7	+2.72%	58
Mexico	97.9	+1.47%	40
Marshall Islands	83.7	+14.1%	79
North Macedonia	95	+0.38%	63
Malta	96	-2.07%	56
Montenegro	98.4	-0.631%	35
Mauritius	95.6	+0.156%	59
Malaysia	97.7	+0.711%	41
Norway	99.2	+0.0833%	23
Oman	96.5	-2.72%	51
Peru	95.5	+2.54%	61
Portugal	99.3	+0.246%	20
Palestinian Territories	98.7	-0.339%	31
Qatar	84.8	-10%	77
Romania	95.8	+0.685%	57
Senegal	61.5	-11.4%	91
Singapore	99.7	+0.134%	7
El Salvador	84.6	+2.96%	78
San Marino	96.2	-1.78%	55
Serbia	98.3	+0.808%	36
Suriname	70.1	-20%	86
Slovakia	99.6	+0.379%	13
Slovenia	99.4	-0.198%	19
Sweden	99.8	+0.111%	3
Seychelles	95	-0.513%	62
Syria	68.4	-26.6%	88
Togo	59.6	+12.8%	92
Turkmenistan	99.3		22
Tunisia	97.3	+3.32%	46
Turkey	99.8	-0.0297%	2
Ukraine	99.1	+0.725%	27
Uruguay	98.7	-0.195%	30
United States	87.1	-3.81%	73
Uzbekistan	98.1	-1.63%	38
Vanuatu	77.1	-9.03%	83
South Africa	95.5	-0.108%	60

Python
R

                    
# Install missing packages
import sys
import subprocess

def install(package):
subprocess.check_call([sys.executable, "-m", "pip", "install", package])

# Required packages
for package in ['wbdata', 'country_converter']:
try:
__import__(package)
except ImportError:
install(package)

# Import libraries
import wbdata
import country_converter as coco
from datetime import datetime

# Define World Bank indicator code
dataset_code = 'SE.PRM.PRSL.ZS'

# Download data from World Bank API
data = wbdata.get_dataframe({dataset_code: 'value'},
date=(datetime(1960, 1, 1), datetime.today()),
parse_dates=True,
keep_levels=True).reset_index()

# Extract year
data['year'] = data['date'].dt.year

# Convert country names to ISO codes using country_converter
cc = coco.CountryConverter()
data['iso2c'] = cc.convert(names=data['country'], to='ISO2', not_found=None)
data['iso3c'] = cc.convert(names=data['country'], to='ISO3', not_found=None)

# Filter out rows where ISO codes could not be matched — likely not real countries
data = data[data['iso2c'].notna() & data['iso3c'].notna()]

# Sort for calculation
data = data.sort_values(['iso3c', 'year'])

# Calculate YoY absolute and percent change
data['value_change'] = data.groupby('iso3c')['value'].diff()
data['value_change_percent'] = data.groupby('iso3c')['value'].pct_change() * 100

# Calculate ranks (higher GDP per capita = better rank)
data['rank'] = data.groupby('year')['value'].rank(ascending=False, method='dense')

# Calculate rank change from previous year
data['rank_change'] = data.groupby('iso3c')['rank'].diff()

# Select desired columns
final_df = data[['country', 'iso2c', 'iso3c', 'year', 'value',
'value_change', 'value_change_percent', 'rank', 'rank_change']].copy()

# Optional: Add labels as metadata (could be useful for export or UI)
column_labels = {
'country': 'Country name',
'iso2c': 'ISO 2-letter country code',
'iso3c': 'ISO 3-letter country code',
'year': 'Year',
'value': 'GDP per capita (current US$)',
'value_change': 'Year-over-Year change in value',
'value_change_percent': 'Year-over-Year percent change in value',
'rank': 'Country rank by GDP per capita (higher = richer)',
'rank_change': 'Change in rank from previous year'
}

# Display first few rows
print(final_df.head(10))

# Optional: Save to CSV
#final_df.to_csv("gdp_per_capita_cleaned.csv", index=False)

                    
# Check and install required packages
required_packages <- c("WDI", "countrycode", "dplyr")

for (pkg in required_packages) {
  if (!requireNamespace(pkg, quietly = TRUE)) {
    install.packages(pkg)
  }
}

# Load the necessary libraries
library(WDI)
library(dplyr)
library(countrycode)

# Define the dataset code (World Bank indicator code)
dataset_code <- 'SE.PRM.PRSL.ZS'

# Download data using WDI package
dat <- WDI(indicator = dataset_code)

# Filter only countries using 'is_country' from countrycode
# This uses iso2c to identify whether the entry is a recognized country
dat <- dat %>%
  filter(countrycode(iso2c, origin = 'iso2c', destination = 'country.name', warn = FALSE) %in%
           countrycode::codelist$country.name.en)

# Ensure dataset is ordered by country and year
dat <- dat %>%
  arrange(iso3c, year)

# Rename the dataset_code column to "value" for easier manipulation
dat <- dat %>%
  rename(value = !!dataset_code)

# Calculate year-over-year (YoY) change and percentage change
dat <- dat %>%
  group_by(iso3c) %>%
  mutate(
    value_change = value - lag(value),                              # Absolute change from previous year
    value_change_percent = 100 * (value - lag(value)) / lag(value) # Percent change from previous year
  ) %>%
  ungroup()

# Calculate rank by year (higher value => higher rank)
dat <- dat %>%
  group_by(year) %>%
  mutate(rank = dense_rank(desc(value))) %>% # Rank countries by descending value
  ungroup()

# Calculate rank change (positive = moved up, negative = moved down)
dat <- dat %>%
  group_by(iso3c) %>%
  mutate(rank_change = rank - lag(rank)) %>% # Change in rank compared to previous year
  ungroup()

# Select and reorder final columns
final_data <- dat %>%
  select(
    country,
    iso2c,
    iso3c,
    year,
    value,
    value_change,
    value_change_percent,
    rank,
    rank_change
  )

# Add labels (variable descriptions)
attr(final_data$country, "label") <- "Country name"
attr(final_data$iso2c, "label") <- "ISO 2-letter country code"
attr(final_data$iso3c, "label") <- "ISO 3-letter country code"
attr(final_data$year, "label") <- "Year"
attr(final_data$value, "label") <- "GDP per capita (current US$)"
attr(final_data$value_change, "label") <- "Year-over-Year change in value"
attr(final_data$value_change_percent, "label") <- "Year-over-Year percent change in value"
attr(final_data$rank, "label") <- "Country rank by GDP per capita (higher = richer)"
attr(final_data$rank_change, "label") <- "Change in rank from previous year"

# Print the first few rows of the final dataset
print(head(final_data, 10))

Compare

Persistence to last grade of primary, total (% of cohort)

Year: 2020

Acquire full dataset with Python or R from source