The Correlates of International Tourism Arrivals

A data set on the correlates of international tourism arrivals, from around 2019.

Usage

itarriv

Format

A data frame with 149 observations on the following 8 variables.

country

a character vector for the English country name

iso3c

a character vector for the three-character ISO code

ln_itarriv

international tourism, number of arrivals (log-transformed)

ln_airtranspc

air transport, passengers carried (log-transformed)

relprice

the price-level ratio of the PPP conversion factor (GDP) to the market exchange rate

ln_gdppc

GDP per capita in constant 2015 USD (log-transformed)

pvest

political stability and absence of violence/terrorism (estimate)

fcas

a dummy variable indicating whether the state is a "Fragile and conflict affected situation"

Details

Data generated by use of the WDI package on 25 September 2025. Underlying sources are multiple. These are the UN World Tourism Organization (tourism arrivals), the International Civil Aviation Organization (air transport), World Development Indicators (relative price), and Worldwide Governance Indicators (political stability). GDP per capita data draw from various sources too indirectly specified by the World Bank Group to say here. The dummy variable for fragile and conflict-affected situations comes from the wb_groups data frame in this package and are current as of the 2025 fiscal year.

Tourism arrivals are for the 2019 fiscal year. Relative price, GDP per capita, political violence, and air transport passengers carry forward whatever is the most current observation (with 2014 as the oldest year). The most recent observation possible for these indicators is 2018. I teach students how to do this on my blog.

Relative price is theoretically unbounded at the right, but bound at the left at 0. It will never be observed, but 0 represents a conceptual extreme in which a country is infinitely cheaper than the United States. 1 is the United States. Higher values than 1 indicate a country's currency is relatively more expensive compared to the U.S. dollar.

Mali apparently had 0 air transport passengers in the most recent year available, per the procedures I used here. I excluded Mali from this analysis to avoid the discomfort of the undefined log of 0 and the discomfort of having to be super technical with what a logarithmic transformation of x + 1 is actually communicating. See Chen and Roth (2024) on this point.

Some of these indicators are no longer supported by the World Bank's API. Use only for the intended purpose of doing teaching students about the simple linear model with an application that might make H. Peter Gray proud if he were still with us.

References

Chen, Jiafeng, and Jonathan Roth. 2024. "Logs with Zeros? Some Problems and Solutions." The Quarterly Journal of Economics 139(2): 891–936.

Miller, Steven V. 2024. "Make Simple Cross-Sectional Data with World Bank Data (from {WDI}). URL: https://svmiller.com/blog/2024/10/make-simple-cross-sectional-world-bank-data-wdi/