[Catalist] Correlation between school size and illness project
Leon Harris
leon at quoll.com
Sun May 8 11:24:42 AEST 2016
Thanks you to my colleagues who responded to my request for data about
the number of sick days they experienced in 2015 and the student
population size of their school.
This is a back-of-the-envelope style pre-investigation and has a number
of flaws, but within its limitation is quite interesting.
There appears to be no correlation between school size and number of
sick days.
You can see the results of the project at
https://plot.ly/plot?fid=Harristotle:105#
Further data will be added as it becomes available.
The main limitations to validity of this are :
* small sample size
* self-selection of respondents (can't do anything about this for this
type of study, except note the possibility that it may introduce biases)
Incidentally, plot.ly is a very valuable research and teaching tool that
we possible should use more frequently.
It is very capable with big data sets, it was designed for Big Data ^tm
B.S ^(C) Bandwagon
and has excellent bindings to languages like python and R. Increasingly
modern science is using this.
I have used it in the past to upload remote temperature data, for
example so that you can see what is happening in your compost bin while
on the train to work.
I did not expect these results at all- they are quite contrary to my
initial hypothesis.
My next set of thoughts lean to the idea that the handrails on the bus
and train might be the source of my woe, or alternatively that perhaps I
just have a c rap immune system and should crawl away somewhere and die
!! :)
Cheers,
Leon
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.stawa.net/pipermail/catalist_lists.stawa.net/attachments/20160508/9db3895d/attachment.html>
More information about the Catalist
mailing list