--- In FairfieldLife@yahoogroups.com, Peter <[EMAIL PROTECTED]> wrote: > > > > --- sparaig <[EMAIL PROTECTED]> wrote: > > > > > > The ME theory is a large numbers theory. The crime > > rate of Fairfield > > could be skewed by having a single pickpocket living > > among the > > sidhas. For that matter, the traffic accident rate > > during the Taste > > of Utopia course could go up because the town's > > adult population more > > than doubled (tripled) during that time. To > > accurately measure the > > predicted claims of the ME, you would need a much > > larger statistical > > sample population for them to effect than Fairfield, > > IA, like > > Washington, DC. > > Another post hoc explanation. The ME theory as now > constructed does not take sample size into > consideration. What is the critical sample size and > why?
Sample size is not so much a key component of theory -- its a more practical consideration. Sample size determines both the size of the error term one wants to deal with, and the cost of the project. But its an steep slope exponential relationship. Halving the error term requires a 2^2 increase in sample size, and thus project costs. Your point is more applicable to overall research design: how many and what types of ME projects, in what types of cities, over what periods of time, and what corresponding physical (weather, etc) and social data (education, income, age, employment levels, etc.) that can be collected that is relevant to the analysis of exogenous explanatory variables etc. The reason for each research design decision is driven by the need for observing as much variation in the dependent variable (crime), driven driven by many diverse explatory variables. This diversity allows the teasing out, via powerful statiscal methods -- such as multiple regression -- which parts of the variation in the dependent variable (crime) are dirven by the independnet variable of interest (ME) and which are driven by physical, regioanl and social factors. Observing the effects on only one city, during one period of time, is roughly analogous to testing a new medicine on one patient. It may provide some insights, but not be conclusive. ------------------------ Yahoo! Groups Sponsor --------------------~--> Join modern day disciples reach the disfigured and poor with hope and healing http://us.click.yahoo.com/lMct6A/Vp3LAA/i1hLAA/UlWolB/TM --------------------------------------------------------------------~-> To subscribe, send a message to: [EMAIL PROTECTED] Or go to: http://groups.yahoo.com/group/FairfieldLife/ and click 'Join This Group!' Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/FairfieldLife/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
