Azad Rasul

The Pearson Correlation will tell you how the air temperature and the disease are linearly correlated, in other words, dependent or related. Thus, a positive correlation indicates that, as one variable increases, the other also increases and a negative correlation indicates that as one variable increases, the other decreases. However, the significance of this dependency will depend on the estimate. If the correlation estimate is different from 0, it means these two variables(in your case, air temperature and disease) are dependent, and if the value is 0.5 or more, it means this dependency is significant.

The R-squared, on the other hand, measures the overall fitness of the model, telling you how the variability of the response variable (in this example Disease) has been explained by the explanatory variable (air temperature). An R-squared greater or equal to 0.5 means about 50% or more of the variability in the response variable has been explained by the explanatory variable meaning the overall fitness of your model is good.

Best regards,

Caleb.

It hardly matters. Further, if there is a seasonal impact it will likely be both temperature and humidity which matter, in combination. Multiple regression is therefore likely what you will want to explore.

I suggest a graphical residual analysis.

Cheers - Jim

R squared is more reliable as an index of model adequacy than Pearson product moment correlation coefficient. The former called the coefficient of determination indicates the extent of variability in the dependent variable explained by the independent variable. The latter is an index of the linear correlation between the two variables.

Dear Azad Rasul

A tool used by Department of Homeland Security (DLS) to calculate the decay of SARS-CoV-2 may guide you on your question.

https://www.dhs.gov/science-and-technology/news/2020/06/12/news-release-st-launches-tool-predict-decay-airborne-coronavirus