Forgot about Dok Will Hunting
tholdren will probably try to chime in, which would be
I digress. I ain't no data scientist nor really that proficient at higher level math (since it bored the out me), but I feel I have a keen common sense and a good logical "eye." Battling these Covid truthers has got me a bit more into data (like real data, beyond looking at basketball reference and advanced stats), but I'm still a novice. That said, I wanted to examine lockdown efficacy in the primary EU countries + UK.
I created this scatter plot in hopes to reveal a correlation. What you're looking at here is how hard (stringency) a country locked down on the first day of their reported death. Insofar, that is all what this plot is looking at (it's not examining stringency of lockdown after the first death was recorded). Note: Don't be thrown off by the bottom range. Higher means less stringent. I was lazy and just clipped the template off OurWorldInData.
I'm curious where a data person might draw their trend line. I know Splits is a data guy, but he ain't around. I know BaselineBum is post-grad level math, so I'm throwing out the bat signal.
Forgot about Dok Will Hunting
Added some data viz and fixed the bottom numbers so they aren't inverted.
Relationship between the two looks more parabolic than direct tbh, I wouldn’t draw a trend line.
BwahahahahhahahahahahHHh
It would be interesting to see this plot for US states. Perhaps, the top 15 or so states with largest populations.
Right on. Would you say there's a correlation between lockdown measures and death mitigation? I've been told by truthers that the correlation is low to non-existent. To me, the correlation seems clear, but I know data interpretation is not always straightforward.
Literally as bad as using raw data case count. Actually worse
When did they join the EU?
beahahahahhahahahahahaha
Just using eyeballs to determine correlation.
Shut this gossip down
No. Countries that had an initial "lockdown index" of 40 or above all fared better than countries that were late to lockdown. Explain the block of countries is the far right block that all had indexes under 20 and suffered the worst?
so much salt
There's nothing to explain. You're using undefined terms with your data.
Additionally, as I stated before you are using an eyeball instead of math to determine a correlation. And you are using one piece of data.
This is as worthless as you thinking case counts raw matter, without figuring out age or detail (ab/pcr/date of onset) etc. It's worthless.
I'm not. The terms are clearly defined. How lockdown measures "on the day of first reported death" relate to final death toll.
Scatterplots aren't supposed to be granular and are useful for seeing if there's correlation between two variables, like say age and obesity.
We're not worried about age stratification, how deaths are counted, obesity rates of these countries, we're looking at if there's any relationship between the two defined variables of death toll and initial lockdown stringency. We can see a correlation (countries lighter on measures more deaths).Scatter plots are used to observe relationships between variables.
^And Will Hunting brought up a key observation in how the trajectory is parabolic. It seems countries with a 25 or greater lockdown index can manage their death toll pretty well, but once you cross that 20 threshold, the death toll accelerates. This would vibe with the exponential nature of the virus. Going into that 20 index which might mean reopening places like bars, theaters, and churches which causes the death toll to take off.
You are. You do not take into context the cases reported before lockdown, how cases are determined, how deaths are counted, how deaths are determined, or even mitigation pre lockdown, etc.
You cant determine correlation without math. Your scatteplot is as worthless as current raw data you use to describe the horror of the pandemic because it uses no context. This is as dumb as you saying case count means xxxxx
/thread
Again, scatterplots are not supposed to granular, dude. If you're going to do a scatterplot on how obesity relates to life expectancy, you're not going to consider the genetic history of each individual, how stressful their job it, if they are smokers or not. You just want to see if there's a relationship between being fat and how long you live. I don't need to get that granular here. I plotted 25 different countries and the relationship seems pretty obvious. Also, scatterplots are meant to be visual.
What do we see here? As stringency lessens, deaths rise, ergo positive correlation.Correlations may be positive (rising),
https://www.statisticshowto.com/line-of-best-fit/
https://www.statisticshowto.com/prob...t-chart/#excel
Don't need the math so much as a quick "how-to" for excel, which does it for you, it seems.Adding a Trendline
Step 1: Click the “Layout” tab.
Step 2: Click “Trendline” and then click “More trendline Options.”
Step 3: Click the “Show equation on chart box” and then click “Close.”
Example 2: Create a scatter plot in Microsoft Excel plotting the following data from a study investigating the relationship between height and weight of pre-diabetic patients:
Height (inches): 72, 71,70,67,65,64,64,63,62,60
Weight (lb): 180, 178,190,150,145,132,170,120,143,98
Select chart itself.
A "chart design" menu should appear along top.
Select "add chart element" (upper right in excel 2016), and select "linear", and it will add it automatically. "least squares regression" in statistics-speak. Can probably find the equation somewhere.
nvm
Last edited by Trainwreck2100; 07-28-2020 at 01:36 PM.
Bwahahahahahhahahahahahajahah
The only thing you have correct is scattetplots visual. Worst math sense I have seen.
Well besides the time you said states don't include probable deaths.
That was also ignorant
There are currently 1 users browsing this thread. (0 members and 1 guests)