In lieu of an abstract, here is a brief excerpt of the content:

10 How Economics Stays That Way: The Textbooks and the Referees Small wonder that students have trouble [learning significance testing ]. They may be trying to think. w. edwards deming 1975, 152 The proximate cause of the unhappy situation in economics is that almost all the teachers of econometrics claim that statistical significance is the same thing as scientific significance. The econometrician David Hendry, for example, is famous for saying “test, test, test,” where the phrase means “Fisher, Fisher, Fisher,” and most statistical textbooks in any field, from advanced theoretical statistics down to the merest cookbook, recommend the same (Hendry 1980). A few get it right. Morris DeGroot, a Roosevelt University graduate (1952) and a distinguished statistician and teacher of several Nobel laureates in economics at Carnegie-Mellon University, wrote as follows in his exemplary textbook of 1975. It is extremely important . . . to distinguish between an observed value of U that is statistically significant and an actual value of the parameter . . . . In a given problem, the tail area corresponding to the observed value of U might be very small; and yet the actual value . . . might be so close to [the null] that, for practical purposes, the experimenter would not regard [it] as being [substantively] different from [the null]. (496). DeGroot does not leave the matter as a throwaway point, a single sentence in an otherwise Fisherian tract, as so many of even the minority of statistics books that so much as mention the matter do. On the contrary, he 106 goes on, “It is very likely that the t-test based on the sample of 20,000 will lead to a statistically significant value of U. . . . [The statistician] knows in advance that there is a high probability of rejecting [the null] even when the true value . . . differs only slightly from [the null]” (497). But few econometrics textbooks make the distinction between statistical and economic significance. Even the best do not give equal emphasis to economic significance, to balance the scores, sometimes hundreds, of pages devoted to explaining Fisherian significance. In the texts widely used in the 1970s and 1980s, for example, when bad practice was becoming standard, such as Jan Kmenta’s Elements of Econometrics (1971) or John Johnston’s various editions of Econometric Methods (1963, 1972, 1984), there are no mentions of the distinction. Peter Kennedy, in his A Guide to Econometrics (1985), briefly mentions that a large sample always gives “significance.” This is part of the point, but not nearly all of it, and in any case it is relegated to an endnote (62). He says nothing else on the matter. Clive Granger on Not Mentioning Economic Significance Arthur Goldberger gives the topic of “Statistical vs. Economic Significance ” a page of his A Course in Econometrics (1991), quoting a little article by McCloskey in 1985. Goldberger’s lone page has been flagged as unusual . The same Clive Granger reviewed four econometrics books in the March 1994 issue of the Journal of Economic Literature and wrote that “when the link is made [in Goldberger between economic science and the technical statistics] some important insights arise, as for example the section [well . . . the page] discussing ‘statistical and economic significance,’ a topic not mentioned in the other books” by R. Davidson and J. G. MacKinnon ; W. H. Greene; and W. E. Griffiths, R. C. Hill, and G. G. Judge (Granger 1994, 118; italics supplied). Not mentioned. That is the standard for education in econometrics and statistics at the advanced level. The three stout volumes of the Handbook of Econometrics (Griliches and Intriligator 1983–86) contain a lone mention of the point, unsurprisingly by Edward Leamer.1 In the 732 pages of the Handbook of Statistics there is one sentence by Florens and Mouchart (Maddala, Rao, and Vinod 1993, 321). Aris Spanos has in his impressive Probability Theory and Statistical Inference tried to crack the Fisher monopoly on advanced econometrics, but even Spanos, a Hendry student, How Economics Stays That Way ⱐ 107 looks at the world with a sizeless stare (1999, 681–728). His history of hypothesis testing has in any case been ignored. In the heyday of rational expectations macroeconomics, for example, its leading practitioners did not get the statistical point even approximately right (Lucas and Sargent 1981). No wonder. As in so many other fields, the econometrics of rational expectations has been entirely inconclusive . Many articles were produced, many t-tests performed, many careers smoothly advanced. No scientific findings yet. For example, in “Rational Expectations, the Real Rate of Interest, and...

Share