Phil Henshaw wrote: > I'm very pleased you also see the curve as reflecting a pattern in the > phenomenon of public trust I said it was reasonable to posit that. Also reasonable to posit other things to find the model that explains the most while assuming the least, e.g. that `trust' is something all people seek, and thus some kind of universal explanation for human inclinations. > I think you're right on to observe that > the decay pattern started *prior* to any of the big mistakes that later > confirmed it. One thing that might produce that pattern is excessive > original expectations, that we may have had 'irrational exuberance' for > things we thought we could do about 9/11 and as a community, we held on > to the desire to retaliate even as our ability to believe it possible > kept getting steadily harder and harder... > Actually I meant the period prior to 9/11. There appears to be a downward trend there too.
A contrasting hypothesis being roughly "Things go bad for typical presidents in this democracy, it's just a question of how fast." Put another way, that intensity of presidential criticism scales with time, automatically as a general collective protection mechanism. As cumulative deaths in Iraq are necessarily only increasing, and popularity is going down steadily, it's no surprise to see a correlation. (And they say this.) I find the volatility and change plots in the paper more convincing correlates of Iraq-related events, e.g. figure 6 and 8. And really not a big difference in Table 2 comparing change in log gas price vs. cumulative U.S. deaths in Iraq / 1000. It would be interesting to see what they find should they look at local trend models, as they indicate they want to (page 20). Marcus ============================================================ FRIAM Applied Complexity Group listserv Meets Fridays 9a-11:30 at cafe at St. John's College lectures, archives, unsubscribe, maps at http://www.friam.org
