Thanks Chris .. this is a terrific article and many undergrads can plow through it. I have gone back and forth on the p versus CI (which is simply rearranging the math) versus effect size issue and have come to the conclusion that we have to keep our options open and not use one rule to evaluate research findings. In my stat class - after doing my lecture on how an IQ difference of 1 point can be significantly different when N = 5000 per group - I sometimes talk about the 1988 (?) study of aspirin and Myocardial infarction in JAMA or NEJM (I am home away from my notes) which found a .8% reduction in MI from a sample of 11,000 placebo controls (risk = 1.7%) and about 11,000 who took aspirin (risk = .9%). The chi square is p < .001 but the effect size is tiny, but even that 1% drop is important when the stakes are high and you are one of the roughly 100 who was spared a MI. that's when I introduce "relative risk" thinking: .9 versus 1.7 means the chance of a MI is cut in half. That type of comparison is especially important when dealing with low base rate diseases. And thanks Jim for the divorce example ...
JK ========================== John W. Kulig, Ph.D. Professor of Psychology Coordinator, Psychology Honors Plymouth State University Plymouth NH 03264 ========================== ----- Original Message ----- From: "Jim Clark" <[email protected]> To: "Teaching in the Psychological Sciences (TIPS)" <[email protected]> Sent: Wednesday, February 12, 2014 6:46:59 PM Subject: RE: [tips] Scientific method: Statistical errors : Nature News & Comment Hi Interesting article, although I need to think more about it. One obvious weakness is the old canard about effect size being a better indicator of importance than p value. The author uses the example of a divorce rate change being tiny: "meeting online nudged the divorce rate from 7.67% down to 5.96%." One source indicates that there are about 2,000,000 marriages in the USA per year. 7.67% is 153,400 divorces, 5.96% is 119,200 divorces, for a reduction of 34,200 divorces or 22.3% fewer divorces every year. Not exactly what I would call a "tiny" difference. Take care Jim Jim Clark Professor & Chair of Psychology 204-786-9757 4L41A -----Original Message----- From: Christopher Green [mailto:[email protected]] Sent: Wednesday, February 12, 2014 4:43 PM To: Teaching in the Psychological Sciences (TIPS) Subject: [tips] Scientific method: Statistical errors : Nature News & Comment An interesting article about the problems of p-values that might even be understandable to undergraduates. http://www.nature.com/news/scientific-method-statistical-errors-1.14700 Chris ....... Christopher D Green Department of Psychology York University Toronto, ON M6C 1G4 [email protected] http://www.yorku.ca/christo --- You are currently subscribed to tips as: [email protected]. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13251.645f86b5cec4da0a56ffea7a891720c9&n=T&l=tips&o=34162 or send a blank email to leave-34162-13251.645f86b5cec4da0a56ffea7a89172...@fsulist.frostburg.edu --- You are currently subscribed to tips as: [email protected]. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13338.f659d005276678c0696b7f6beda66454&n=T&l=tips&o=34164 or send a blank email to leave-34164-13338.f659d005276678c0696b7f6beda66...@fsulist.frostburg.edu --- You are currently subscribed to tips as: [email protected]. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=34165 or send a blank email to leave-34165-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
