<h3>I have been looking online for some data that can demonstrate whether there is any correlation between covid19 vaccinations and covid cases and deaths for some time now. For some reason i could not find anyone who has shown this data one way or another. Nevertheless, it seemed to me a mystery to me why looking at a few examples that many countries with very high vaccination rates still have high covid rates.. and conversely many countries with low vaccination rates have quite low relative rates. </h3>

I like to know the truth of things, and i also was not comfortable coming to any conclusions based on potentially cherry picking a few countries and trying to say anything from that.

Statistics are BLOODY hard to get right, and you truly have to be objective and look for the truth, rather than to prove a point when you use them. Therefore yesterday i spent the entire day researching for myself using data from https://ourworldindata.org/ to figure out for myself what the truth is about this. I don't trust anyone else's graphs anyways, i want to do proper due diligence on this very important topic!

### Choosing the right data
It is a challenge to choose the right data, and there are many many options to choose from when trying to correlate vaccine use with covid rates. Do you look at cases, deaths, hospitalisations, people who had one shot, or two, or are fully vaccinated? How many countries should you compare?

So, in the end i decided to focus on cases and deaths, separately.  In my opinion deaths may well be a better marker than cases since deaths are always recorded, whereas cases depends on the numbers of people who were tested. I also decided to compare EVERY country in the world (more or less) to provide the best set range that i could. 

Finding correlations is a delicate science, and correlating different countries over the same time frame of one year seemed to me to be the best way to do this correlative check. That is because it removes many variables such as seasonal changes etc. Some people were trying to deny my data on Facebook yesterday by simply showing me how covid cases have decreased in the UK alone since vaccinations started. That is a very poor way to measure any correlation since it doesn't take into account natural variation, seasonal changes and more. 

I would like to say that consider myself a good scientist. That is to say, i am impartial and have no attachment to the outcome one way or another. I am willing to accept the conclusions whichever way it goes. Id also like to mention that i have a strong background in Biology and medicinal microbiology. I have also studied statistics, and learned about how to determine whether data is statistically relevant or not using systems like Chi-Square. 

### The Data
COVID CASES % 
![cases.gif](https://files.peakd.com/file/peakd-hive/eco-alex/23zSNPBj8k85jDbF9b8AiKjYxojSgaa69iYd2XjFCJhNeJsWs5tL5o8CH7Ncp6AeRh4rF.gif)

COVID DEATHS PER MILLION PEOPLE
![deaths.gif](https://files.peakd.com/file/peakd-hive/eco-alex/23xytHMjqYDayJeFjDBtvQN2p81J2LYbZKXd3iCg5iiNU9GVgr4tWiPJBirGSGxcfXFmQ.gif)

The blue bars are the % vaccinated people from high to low for each country in the world. The orange bars are the % cases for the same country over the last year.
What one would expect to see, if vaccines are working, is the orange bars should go from low to high. This means low cases where there high vaccinations and high number of cases where the vaccines are low. 

I find the data that I have presented to be very interesting indeed. In my opinion, based on this data, i come to the conclusion that THERE IS NO CORRELATION BETWEEN BEING VACCINATED AND COVID CASES OR DEATHS. In fact, in terms of cases i can see a clear inverse correlation for cases, whereby the countries with the most cases seem to also have higher vaccination rates. 

The data for deaths shoes less correlation for an inverse relationship, BUT it clearly shows that there is still no correlation between being vaccinated and the number of deaths. According to this data, vaccinations do not seem to be doing very much. 

### Interesting Data Points
I would like to share these interesting data points relating to income.

**High Income**
67% of people are vaccinated and there is a 327% increase in cases and 1500% increase in deaths

**Upper Middle Income**
65% of people are vaccinated and there is a 633% increase in cases and 871% increase in 
deaths

**Low Middle Income**
27% of people are vaccinated and there is a 15% increase in cases and 347% increase in 
deaths

**Low Income**
3% of people are vaccinated and there is a 1.2% increase in cases and 56% increase in 
deaths

These numbers are startling! The difference between them is way beyond anything that you can contest to be chance, and the figures follow a clear logical progression. It is clear, those with more income are having more vaccinations and are also suffering much more from deaths and cases that lower income people. If we take the two extremes we can see a 1500% increase in deaths for the high income people, and yet just 56% increase for low income. That is a 10x difference. Not only that but the vaccination rates are just tiny at just 3% for low income people, compared to 67% for high income people.

I don't think i really need to draw any conclusions from this data, as it is self evident to me. For once, the low income and poorest people seem to be in the best situation! 

### Raw Data
If you would like to see the raw data that i have compiled for each country you can see it below. It is hard to see it all in the graph as there are so many countries!

![raw.gif](https://files.peakd.com/file/peakd-hive/eco-alex/23tw8AehKDmrACLVponiJDU9fayy2kaqBMcuNLo9G4iyphSsLm5Zz4aR2HZWCBobTPiH6.gif)

### Do you want to see more?
If you think that this work is interesting please let me know in the comments. I am happy to create more of these statistics for a wide range of other factors. I am interested to look at fertility for example!

If you understand statistics and have any feedback please also let me know. If you believe my data is somehow flawed i want to know about it. If you think there is a better way to measure this correlation please tell me about it. I am seeking the truth only!