Arquivos de periódicos de maio 2020

09 de maio de 2020

I Am Biased & An Aberration

Statistical problems with my observations

Until the government imposed lock-downs this year I travelled north during winter to escape the cold weather. As I make a lot of observations compared to others, some interesting things start to become apparent. One such artifact is seasonality records for the Scarlet Percher dragonfly Diplacodes haematodes show a bulge during August and September. This is a fairly common species in New South Wales and Queensland and, with mature males being bright red and perching beside water habitat, also readily photographed. Until this past summer nearly half of the records were from me and the chart pictured here was even more distorted. Anybody looking at that would think there is a strange peak emergence at the end of winter but actually its just a freak emerging from the south. As more users come on board this curve will gradually be flattened but highlights the problem of me. 😊

I am somewhat of a pariah - I'm not doing what everybody else is doing and I'm doing a lot of it. This includes local observations too. For example, a common mushroom here is the attractive, bright blue Pixie's Parasol Mycena interrupta but over 90% of the records on iNaturalist are mine.

But other biases are less apparent. I am more inclined to photograph something I know and ignore stuff I expect won't be identified. Historically I've been ignoring small herbs and bryophytes as unidentifiable and therefore not worth recording. For example, so far this year I have recorded the common moss Cyathophorum bulbosum 20 times, which represents over half the observations here. Its not like it suddenly appeared, I've just not been able to recognize it previously and thought I wouldn't be able to have it identified anyway. Another example is Indian Weed or Eastern St Paul's-Wort Sigesbeckia orientalis: half the Australian records are from me and half of those are from this year.

So by being one of the more active observers I am introducing statistical biases. I hope you don't mind. 😎

Publicado em 09 de maio de 2020, 11:41 PM por reiner reiner | 4 comentários | Deixar um comentário