R-squared
A statistical measure of the proportion of variation in an outcome that is explained by a regression model; a value of 1 means complete explanation
As defined in Melissa Dell
Other definitions
A statistical measure of how much of the variation in an outcome (e.g., growth rates) is explained by the explanatory variables (e.g., institutions).
Peter Henry on Growth, Development, and Policy 07/27/2009
Where it comes up · 10
- Melissa Dell2 claims
- Peter Henry on Growth, Development, and Policy 07/27/20091 claim
- Finding Undiscovered Talent1 claim
- Nate Silver1 claim
- Lee Ohanian on the Great Recession and the Labor Market1 claim
- Patrick Collison on Innovation and Scientific Progress1 claim
- Raj Chetty on Economic Mobility1 claim
- Tyler Cowen & Daniel Gross - Identifying Talent - [Invest Like the Best, EP. 277]1 claim
and 2 more
Most important claims · 10 of 11Search all claims for “R-squared” →
- Across countries, GDP per capita and energy consumption per capita are tightly correlated (R-squared likely above 0.8), so energy use and production is effectively the ultimate proxy for human prosperity and economic activity.From SpaceX to Founders Fund to Solving America's Nuclear Fuel Problem
- Differences in marginal tax rates that affect the return to work explain a large share of why total hours worked per working-age population in Germany, France, and Northern Europe fell from above U.S. levels in the 1950s-60s to about 20-25% below the U.S. today; in regressions, adding unionization, employment protection, and other labor variables barely changes the tax coefficient's magnitude or t-statistic or the R-squared.Lee Ohanian on the Great Recession and the Labor Market
- 'Zip code is destiny' is false: the R-squared or explanatory power of neighborhoods for individual outcomes is only about 5-10%, meaning there is large variation within neighborhoods and many other factors (family, education, individual inspiration) matter far more than where you grow up; neighborhood is just one significant level of variation, not a determinant of fate.Raj Chetty on Economic Mobility
- The real debate over institutions is not whether property rights help growth (they obviously do) but one of magnitude—how much of the variation in growth rates institutional factors actually explain—and pinning down such point estimates over 300 years is essentially impossible because you cannot control for everything else.Peter Henry on Growth, Development, and Policy 07/27/2009
- Psychometric effect sizes are extremely light; even IQ's celebrated relationship to earnings, with an R-squared around 0.6 as Daniel recalls, is less decisive than it sounds because it still leaves much variance unexplained.Finding Undiscovered Talent
- The scatterplot of GDP per capita against self-reported wellbeing across the world has an R-squared of about 0.68, an astonishingly strong correlation that is rare in human psychology and sociology, which makes the central case for economic growth extremely strong despite social downsides.Patrick Collison on Innovation and Scientific Progress
- Because the world is fundamentally complex and income gaps between poor and rich places are enormous (some sub-Saharan African places are 40 times poorer than the US), a historical variable that explains 10 percent of the variation (partial R-squared of 0.1) represents very strong persistence, while leaving massive room for other theories of development to also matter.Melissa Dell
- Comparing very different countries like South Korea and the Philippines is uninformative about whether the historical state matters because they differ in too many ways; comparing localities within South Vietnam that share recent modern history but differ in deep history isolates whether past history can still matter without claiming an R-squared of one.Melissa Dell
- New York is extremely efficiently priced at the neighborhood level: a FiveThirtyEight analysis for New York Magazine found cost accounts for an R-squared of about 0.93 with their quality index, meaning it is really hard to improve your lot by neighborhood choice.Nate Silver
- Among personality frameworks, the Big Five Aspect Scale is the least bad, and although the Big 16 may have higher explanatory power in statistical terms, five concepts are usable for teams while sixteen are not.Tyler Cowen & Daniel Gross - Identifying Talent - [Invest Like the Best, EP. 277]