Handwashing is one of the most recommended things to avoid corona. how effective is this? Here is the correlation analysis I made based on countries using Python.

First of all, let me clarify that I am not a medical professional. I made a statistical analysis only in the light of the available data and I recommend you to go to more specialized sources for precise medical information.

The Data

I collected the datasets from different sources. Handwashing statistics are based on research by the organization Worldwide Independent Network of Market Research (WIN). This research shows ratios of hand-washing after going to the toilet among countries. Corona Dataset is taken from the Johns Hopkins University Center for Systems Science and Engineering (JHU CSSE)’s repository. Finally, I retrieve Countries’ population dataset from worldbank.org in order to find ratios per population.

I first started by combining these datasets on a single Pandas Dataframe. I calculated the rate of corona cases per 1 million people among the population. The data I use to compare was this ratio and handwashing percentage.

Spearman Correlation

We find the Spearman Correlation Coefficient at -0.29. This indicates a moderate correlation between the handwashing rate and the appearance of Corona cases. This result is significant because the probability of this result occurring by chance is very low. (p value is < 0.05)

There is negative relation which means a decrease in handwashing may cause an increase in Corona cases.

Visualization on the world map?

In this part, I compare the visualization of corona cases and percentages of people who do not wash their hands. Similar trends can be seen a little here. For example, we see that the handwashing rate in China is quite low together with high corona rate.


Darker colours indicate lower handwashing percentages in the first graph and higher corona rate among the population in the second graph.

Case per 1.000.000 population
Percentage of not Washing Hands


In European graphics, we can observe that colours darken at similar points.

Case per 1.000.000 population
Percentage of not Washing Hands


From the correlation analysis and these visualizations, we can observe that there is a moderate correlation between handwashing statistics and corona observation rates.

Caution: This research has been done personally on available datasets and may not have medical validity.

Source Code