UserFriendly Strip Comments


	The Daily Static UF Archives Register UF Membership Ad Free Site Postcards Community Geekfinder UFie Gear Advertise on UF Forum Rules & FAQ Username Password Create a New Account

Back to UserFriendly Strip Comments Index

ATTN: Statistics Ufies	by vectorz	2006-09-18 04:04:33
It really depends on the experiment,	by Jeff_uk	2006-09-18 04:27:34
Alternatively you could use the interquartile mean	by Jeff_uk	2006-09-18 04:33:31
Thanks & furter info.	by vectorz	2006-11-19 12:55:59
What about grouping?	by anymouse	2006-09-18 05:53:34
They probably do	by vectorz	2006-09-18 06:59:18
Confounding factors!	by mathwaster	2006-09-18 07:23:43
If you have 8 repeats of the same experiment they ought to be spread across the plate (and all other experiments should be as widely spaced as possible) to try to remove 'location on plate' as a confounding factor. To display the data I can recommend boxplots (basically plot mean, interquartile range, maximum and minimum and a few outliers). Ths should help. Mathematically speaking, if you want to use a single statistic you should be looking for various (beta)-robust statistics to use. The median is the ultimate version of this, but as JeffUK pointed out, it loses a large ammount of 'spread' information. Perhaps a better choice is the alpha-trimmed mean, where you ignore the top and bottom alpha percent and find the mean of the rest. However if you have 2 out of 6 (or 8) trials going 'wrong' there is not much statistics can do to help you unless you can confidently assign them to 'junk' and believe the rest. In which case, finding the mean of the rest is OK. This procedure is liable to error due to people being too confident...
[ Reply ]
Thanks.	by vectorz	2006-09-18 07:50:32

[Todays Cartoon Discussion] [News Index]

Come get yer ARS (Account Registration System) Source Code here!
All images, characters, content and text are copyrighted and trademarks of J.D. Frazer except where other ownership applies. Don't do bad things, we have lawyers.
UserFriendly.Org and its operators are not liable for comments or content posted by its visitors, and will cheerfully assist the lawful authorities in hunting down script-kiddies, spammers and other net scum. And if you're really bad, we'll call your mom. (We're not kidding, we've done it before.)