----- Original Message ----- From: "John D. Giorgis" <[EMAIL PROTECTED]> To: "Killer Bs Discussion" <[EMAIL PROTECTED]> Sent: Friday, January 16, 2004 8:15 PM Subject: Re: Double Standards on Regional Bigotry
> At 08:16 PM 1/15/2004 -0800 Doug Pensinger wrote: > >I'll also state for the record that, while we all make mistakes, Dan's > >data is usually pretty solid, and I can remember a few instances when he > >was mistaken and owned up to it. I don't think "fast-and-loose" describes > >his use of facts at all. > > While this may be true in physics, I haven't found that to be so in > economics. In particular, Dan has been fond of declaring that "economics > is not a sciece" - largely on his critique that economics cannot produce > useful prediction. I did not say that. I said that its predictive ability and its ability to falsify theories are not sufficient for it to be a science. I said that one cannot precict very simple things that one would expect to be able to predict in a science. But, that does not mean that one can not use emperical data at all. The idea that science = the ability to use emperical data is not mine. It appears to be yours. The definition of science is also not "if I can make predictions, I have a science." People can make predictions in sporting events; but that doesn't mean football is a science. I'd be happy to engage in a general discussion on what makes a science a science, and why fields such as ecconomics, sociology, social work and history are not sciences, even though they share some techniques with science. Like many definitions, this does not involve a hard and fast set of rules. But, there are reasons that the "social sciences" are different from sciences. Some characterization of science 1) Theories are falsifiable by comparing detailed predictions with observations 2) Experiments (or at least pseudo-experiments)* can be done to test the theories. 3) There is either a single paradigm or a singular shift from an old paradigm to a new paradigm. 4) Old theories are special cases of the new theory 5) Theory of complex phenomenon are explained in terms of theories of simple phenomenon. One can outline the linkage, in principal at least. 6) The outcome of experiments is not influenced by human desires or persuasions...it is independant of the free will of people. 7) Professionals do not make retorical arguements to advance their case. * One may argue that there are no experiments in conmology. Well that's only a quarer true. First, cosmology ties in very well with experimental particle physics. Second, one can propose things to look for in the cosmos to test theories. >Yet, this never seems to stop him from using economic > data to make predictions that "Republicans are horrible for the country in > that the produce X effect." Well, you can't have it both ways. You > can't continually insist that economis is not a science Which I said >and does not produce useful predictions, Which I did not say. Its not that there is no predictive ability in ecconomics. Both of us can come with example of how general things can be predicted. Specific predictions from basic theory do not exist. Some very simple things can be near impossible to predict....mostly true because ecconomics involves human behavior. >and then insist that it predicts that Republicans are just awful. I just pointed to 80 years worth of data that indicate that the US ecconomy fairs worse under Republicans than Democrats. I fully realize that this is not a scientific proof and that the actual chain of causality is such that I cannot scientifically prove my assertation. Even though the data are several sigma from random chance, there can always be an explanation for every difference. So, I have not rigorously proven my assertaiton, I've mearly given credence to its versimilitude. > In addition, many of these predictions play fast and loose with the data. > One of the classic ones is to assign the economic peformance of the country > in 1980, when Jimmy Carter was President, to Ronald Reagan - never minding > that Ronald Reagan's economic policies really couldn't have taken much > effect until 1982 at a minimum. Ah, what I did was compare a number of dates to 1980. The difference in GDP, for example, between the '92 value and the '80 values tell what happened during the Reagan/Bush I era. Now, strictly speaking, there are a few weeks of Carter thrown in those years, but that's in the noise. Now, lets look at your arguement that the effects of Reagan would have at least a 2 year time delay. In the sciences, any assertation such as that would be accopanied by a real proof. Not a decent retorical arguement. Not an appeal to common sense. A proof. Now, since I don't consider ecconomics a science, I won't consider the lack of such a proof to indicate that one can immediately dismiss your arguement. I would agree that there are delays in the results of some actions, and there are close to immediate effects with other actions. > This is an egregious and repeated error, which I just simply find to be inexplicable. OK, John , I'd like multiple cites on that. I'd like examples of how I compared 1980 data to earlier time frames and attributed that to Reagan. I can remember once grouping things by decade and saying each decade was mostly X or Y. But, as I did further analysis I worked to match up years with presidencies. Further, you give very little in the way of analyis yourself. The main arguement you seem to be making is that numbers don't matter because they can always be argued away. The problem with this type of arguement is that we, then, should never consider correlations because there can always be a good retorical arguement to ignore them. Finally, Eric, IIRC, tried putting in a 2-year time delay and there results I obtained were changed only moderately. > Likewise, he never connects > his analysis to policies, such as, for insistance taking account of the > fact that George H. W. Bush raised taxes He had to because his "read my lips" promise to keep Reagan's policies intact were impossible to sustain. >or that the recession of '81-'82, > which was produced in large part by Paul Volcker, is widely considered to > have been absolutely necessary for the long-term health of the economy. The recession still would have happened, but Reagan made sure that the ill effects were concentrated in the lower classes more than usual. This was accomplished by the increase in taxes on lower income, the decrease of taxes on upper incomes, the decrease in benefits to the working poor, and the championing of union busting. Dan M. _______________________________________________ http://www.mccmedia.com/mailman/listinfo/brin-l
