Hi Large populations (or samples) also make clear that small differences can be important, as well as significant. In the present case, the .29% increase represents several thousand additional children having autism in this sample of over 600,000.
.0047 x 662,730 = 3115 .0076 x 662,730 = 5037 Difference equals .0029 x 662,730 = 1922 kids This represents a 62% increase relative to the baseline frequency and would have significant (in the nonstatistical sense) implications for parents, school boards, social services, and so on in any community having to address this increased need. I haven't read the article, but it would be interesting to know what the function relating risk to spacing looks like. One could at least "recommend" a minimal spacing to parents. In my case, I left 36 years between my two kids (different mothers), so presumably am safe (at least from this concern ... there is still the old father - schizophrenia link to worry about). Take care Jim James M. Clark Professor of Psychology 204-786-9757 204-774-4134 Fax [email protected] >>> "Dale R. Floody" <[email protected]> 16-Jan-11 5:35:55 AM >>> Hello. I joined this listserve a few days ago, and so far it's been interesting (especially the flame wars, aka "I'm smarter than you" displays). Michale Sylvester wrote the following: " Has anyone been following the recent discovery that a child's probability of developing autism is dependent on that child's birth proximity to the older adjacent sibling?" I did take a look at the study, and there's a summary of my findings on my web page (link below) under "updates". Basically, the authors (Cheslack-Povasta, Liu & Bearman) suggest hat a number of intervening variables (including folic acid) could contribute to the link they found between spacing and diagnosis of autism. They also argue that empirical research proceeds at a "slow crawl" and that considerably more related research is called for (they even attempt to distance themselves from "junk science" like the alleged link between autism and the MMR vaccine). In addition, my closer examination of the results revealed that overall the odds of having a child diagnosed as autistic were .0047, but for closely spaced children the odds were .0076. Large sample sizes (662,730 subjects in California) can often lead to results that are statistically significant, but that are perhaps less practically/socially significant. While an increased probability of .29% is probably worth bearing in mind, it*s probably not worth losing too much sleep over. Prenatal stress and sleep loss stemming from excessive parental worrying about close spacing can perhaps be just as problematic as the spacing itself* Dale R. Floody, Ph.D. [email protected] Viterbo University Psychology Department LaCrosse, WI 54601 Professor Emeritus UW Colleges Psychology Department www.balancedpsych.com --- 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=7930 or send a blank email to leave-7930-13251.645f86b5cec4da0a56ffea7a89172...@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=7938 or send a blank email to leave-7938-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
