Forums / General Discussion Forums / Ad Infinitum / Epsilon Aurigae is not varying
Epsilon Aurigae is not varying
I though that title would get your attention..............
Seriously, I didn't want to highjack the Science is not pretty thread, but I would like to make a point about using selected data to 'prove' that the scientists have got it wrong.
If you plot the CitienSky observations for eps Aur for the past 800 days, you can see observations showing it to be at mag 3.4 for the whole of this period. These are, of course, the correct values, all the others are from people who have bought into the notion that it is a variable star and are determined to make the results fit their own pet theories about it being a variable star. Gosh, some people even make money out of so called variable stars...
;-)
A well done post, David. I see your point. It's always important to carefully study the whole picture, and prevent yourself from any temptation of taking a limited view that might prove preconditioned biases. If you ever entertain yourself with pseudo-science topics (i.e., UFOs, etc.) you can quickly see how the researchers of these topics survive (thrive?) off "limited perspective" analysis. It "proves" their point to ignore the greater amount of evidence that disproves their claims in favour of the smaller amount of incidental (likely mistaken or incorrect) evidence that fuels their biases.
" the first time I read your post I missed the satire..."Put it down to my quirky British humour!!:-) :-)David.
Hi Ryan,Yes, I do 'entertain' my self with pseudo-science topics, and it is often the lack of - or complete disregard for - evidence that seems to me to be at the heart of these different views. It is rare for there to be an agreed body of evidence that is interpreted differently, rather, the different views are based on different bodies of evidence. David.
Good Points!




David,Good example of how to use limited or selected data to prove a point, although the first time I read your post I missed the satire... it wasn't until I hit the smile at the end that I caught on.Extending on what David was doing a little, I have attached a plot of the last 30 or so years of photometric data on eps Aur that I have acquired as part of my Ph.D. work (see the file below), which is included on my 2010 Washington, DC AAS Poster. Not only can you see the eclipse of 1983 (to the far left), but you can also see the beginnings of the current eclipse (to the far right, data current to Dec. 12, 2009). In between, you can see what we keep calling the "out of eclipse variations" (often abbreviated OOE variations) which have complicated the analysis of the system for some time. There are spikes (in Q4 of 1987), drops (Q4 1992), missing data (gaps due to small angular separation of eps Aur to the Sun during the summer, and no data in 1995), quiet periods (1997-1999), and jaggedness (all over the graph due to small systematic differences between magnitudes reported by observers).The point David was trying to drive home was that if you use a limited sample population, you can sometimes miss the big pictures or draw the wrong conclusions (was was discussed in the "Science is not pretty thread"). A lot of my time as a researcher is spent trying to make sense of noisy data while understanding what variations are real, and what are just errors in data acquisition. If you jump to conclusions too quickly, as is the case David demonstrated above, you can come to wrong answers fairly quickly.Brian