3 ms·

No, anecdotes are unreliable. It would be better to do a systematic study, and publish the results in a scientific journal like Current Biology. Then maybe to

No, anecdotes are unreliable. It would be better to do a systematic study, and publish the results in a scientific journal like Current Biology. Then maybe to help get the word out, publish a summary on a website like Science.

jxramos 4y ago
I wonder if some DIY computer vision video set on such a planter box of native flowers could be some indication if sufficiently crowdsourced.
- Shatnerz 4y ago
  I don't see why not? Citizen science has been pretty important to fields like ornithology. I'd actually be surprised if something like this wasn't already happening somewhere at small scale at least. edit: There is even a "Butterfly Count" section on the wikipedia entry: https://en.wikipedia.org/wiki/Citizen_science https://en.wikipedia.org/wiki/Citizen_science
- snowwrestler 4y ago
  Sounds like an interesting idea!
kilgnad 4y ago
Anecdotes have a degree of unreliability and a degree of reliability. Outright dismissing an anecdote as completely baseless and invalid is highly, highly unwise and a sign that the person is not logical. First off, anecdotes have extreme speed. You can get some sort of answer much faster and at a hugely lower cost then a "study". Second, studies are not to be trusted either. There's huge incentive to doctor these things and with the replication crisis this "doctoring" has been shown to be common place. Third, even an "honest" study can be wildly off depending on the data gathered and most of the time these studies only find correlative associations rather then causative (most people don't even know how to conduct a causative experiment). A good example of this is smoking Cannabis and Tobacco. Depending on the time both were deemed healthy to consume by research via multitudes of biased research and and a lack of unbiased studies. But with a little anecdotal common sense it's easy to see that breathing burnt plants into your lungs is likely just bad overall.

jxramos 4y ago

I wonder if some DIY computer vision video set on such a planter box of native flowers could be some indication if sufficiently crowdsourced.

Shatnerz 4y ago
I don't see why not? Citizen science has been pretty important to fields like ornithology. I'd actually be surprised if something like this wasn't already happening somewhere at small scale at least. edit: There is even a "Butterfly Count" section on the wikipedia entry: https://en.wikipedia.org/wiki/Citizen_science https://en.wikipedia.org/wiki/Citizen_science
snowwrestler 4y ago
Sounds like an interesting idea!

kilgnad 4y ago

Anecdotes have a degree of unreliability and a degree of reliability. Outright dismissing an anecdote as completely baseless and invalid is highly, highly unwise and a sign that the person is not logical. First off, anecdotes have extreme speed. You can get some sort of answer much faster and at a hugely lower cost then a "study". Second, studies are not to be trusted either. There's huge incentive to doctor these things and with the replication crisis this "doctoring" has been shown to be common place. Third, even an "honest" study can be wildly off depending on the data gathered and most of the time these studies only find correlative associations rather then causative (most people don't even know how to conduct a causative experiment). A good example of this is smoking Cannabis and Tobacco. Depending on the time both were deemed healthy to consume by research via multitudes of biased research and and a lack of unbiased studies. But with a little anecdotal common sense it's easy to see that breathing burnt plants into your lungs is likely just bad overall.