Cekoviu wrote:United Muscovite Nations wrote:That must be why the European Social Democracies are doing so well.
Statistics nerd time! To see if you're right about whether economic and political freedom cause extra issues with the coronavirus, I've created a
coronavirus intensity measurement (CIM) which controls for population and population density (the formula to calculate it is ((number of coronavirus cases)*(area in km^2))/(population^2)), and to see if it is related to economic and political freedom, I'll be using the 2020 Index of Economic Freedom and 2019 Democracy Index as measures of those respectively. As a quick note, I've excluded Iceland from analyses because its CIM is a massive outlier at around 1400; the next highest is about 50.
First, how does CIM relate to Democracy Index?
It kind of sits at a similar low spot up until the Democracy Index reaches about 7.5, which results in a sudden exponential increase. So that means we're probably looking at an exponential curve, which means to find the strength of the exponential model, we have to log-transform the CIM. If you do that and run a linear model in R, you get:
- Code: Select all
Call:
lm(formula = log(corona_score) ~ democracy_index_2019, data = mydata)
Residuals:
Min 1Q Median 3Q Max
-6.547 -1.462 0.360 1.602 5.198
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -3.87693 0.48571 -7.982 2.92e-13 ***
democracy_index_2019 0.57151 0.08215 6.957 9.01e-11 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 2.246 on 156 degrees of freedom
(38 observations deleted due to missingness)
Multiple R-squared: 0.2368, Adjusted R-squared: 0.2319
F-statistic: 48.4 on 1 and 156 DF, p-value: 9.009e-11
So there's something to it - there's a statistically significant (p << 0.05) exponential correlation between Democracy Index and CIM. However, a large amount of the variation in CIM cannot be explained by Democracy Index, as is evidenced by the small R
2.
Now let's look at economic freedom.
Same kind of thing here, so let's do the same log-transform linear model.
- Code: Select all
lm(formula = log(corona_score) ~ index_economic_freedom_2020,
data = mydata)
Residuals:
Min 1Q Median 3Q Max
-5.9421 -1.3521 0.1506 1.7299 4.5617
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -7.7149 1.0498 -7.349 9.51e-12 ***
index_economic_freedom_2020 0.1144 0.0169 6.771 2.26e-10 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 2.226 on 161 degrees of freedom
(33 observations deleted due to missingness)
Multiple R-squared: 0.2217, Adjusted R-squared: 0.2168
F-statistic: 45.85 on 1 and 161 DF, p-value: 2.264e-10
Basically the same interpretation as above.
So what does this tell us? Well, being extremely authoritarian doesn't mean you're in better shape than a moderate libertarian because you only get the sharp increase when you start becoming very free. Given the benefits of a moderately libertarian ideology compared to auth socialism, those benefits outweigh the very minor costs here. You just shouldn't be excessively liberal.