We took the data from Our World In Data to conduct a preliminary data exploration of covid cases for each country. The correlation matrix provides interesting incites into which features have an effect on covid cases and deaths. The multi-billion-dollar question is "do vaccinations stop the spread of covid?"
The data science provided is the first step in our major exploration. The maths involved may be faulty due to missing values and bad data. For our final project, extensive data cleaning is needed. The tables and visualisations shown here should be considered as an indication as opposed to being the final calculations.
Further research is needed into the efficaciousness of vaccines for particular ages, demographics and various 'vulnerable' categories.
The above correlation matrix is a bit hard to read and does not make any sense without understanding what the numbers and columns mean. Any intersecting box with a coefficient higher than 0.7 is considered to have a high correlation, meaning that one feature increases the likelihood of another. Less than 0.70 has no interest to us.
If we look along the column "new_cases_smoothed_per_million", we do not see anything that sticks out as having an effect on covid cases. Based on this chart alone, no amount of covid vaccinations or booster shots is slowing down the spread of the virus.
If we look at 'new_deaths_smoothed_per_million', the number of people in ICU(Intensive Care Units) beds, and people in hospital in general, has a big impact on the amount deaths due to covid.
New covid cases vs people fully vaccinated
As said previously, in-depth data cleaning is needed to make sure all the mathematical calculations are correct. As it currently stands, it looks like vaccine programs are definitely not stopping the spread of covid.
The covid vaccines also do not seem to have much effect on how many die due to covid. This might be due to the overconsumption of vaccines by the healthy population. More research needs to be done on the efficacy of vaccines on various demographics and vulnerable groups. We dont even know if mixing various brands of vaccines is a good idea.
Note that we do not have any data on long-term side effects which could make the situation in the ICU units worse over the next seven years. According to this study, resources would be better spent improving ICU unit hospitals in general.
Coronavirus Pandemic (COVID-19). Our World in Data.
GitHub - Covid data science.
Over 800 people died after catching Covid-19 in Irish hospitals, figures show.