0
$\begingroup$

I have data that represents the measurements of the concentration of PM particles in the air at different times (every hour of every day from 2018 onwards), and I am supposed to test whether the rules for the lockdown of COVID have any statistical significance. The data is split into three sets (measurements from March 2020-Dec 2020 aka period 1 - strict lockdown rules; March '21 - Dec '21 - looser measures; March '22 - Dec '22 - lockdown is lifted). From the visualization and the descriptive statistics, one can infer that the mean concentration is higher the looser the rules were (meaning that there is a correlation between PM particles and more traffic and more people). But what confuses me is the analytical statistics bit and the choice of test. The data is not normally distributed, which leads me to believe Friedman's test would be most appropriate. Any help would be appreciated!

Improve this question
$\endgroup$
1
  • 1
    $\begingroup$ Have you looked into mixed models for longitudinal data? $\endgroup$
    – Mathemagician777
    Commented Jan 3 at 14:25

0

Reset to default