Re: misusing stats: examples
On 15 May 2000 07:31:17 -0700, [EMAIL PROTECTED] (Michael Granaas) wrote: snip The misinterpretation of results by the popular press has become a core topic for me in recent years. While some of the misinterpretations may be harmless (I doubt that eating extra fiber would hurt you unless it lulls you into a false sense of security about about your health). On the other hand some misinterpretations lead to all kinds of mischief. In recent weeks the press locally has jumped on the report that women earn about $.73 for every $1.00 that a man earns. This is being reported locally as the pay difference FOR THE SAME JOB! But the data are talking about the large aggregate (on average, if you will) not about folks within the same jobs. - well, where did you see this? The $.73 is a bit dated, but I am afraid that from the original reporting that I have seen, they are right and you are wrong, as to the intentions. I thought it was more like $.79 or $.83, across all industries and occupations, nowadays, but there is still a gap in the U.S., which is less than in many countries. It's nearly vanished in a few occupations, if narrow ones - For instance, all U.S. Senators get paid the same. There has been more than one such report. The statistical matching and control has often done pretty well, and with imagination. There's been a gap. In about 1970, when I first entered a workforce, it would be true that male college professors, after 10 years of tenure in an English department, would expect to have a distinct and definite income edge over their similarly qualified female counterparts. There would have been, then, in academia, one of the best work places for equality, at least a 15% difference -- so far as I think I recollect. What gets harder to figure out is whether the tenured man should be compared to a *tenured* female, or should he be compared to the female who was denied tenure solely because she is female? My own sister got extremely pissed off, about 1972, when the insurance agency where she worked automatically recruited a *male* as "management trainee" -- younger, stupid-er, higher paid, with no better background -- instead of considering, at all, the females who were underemployed as secretaries in their own offices. The more extreme comparisons today do try to "control for" the unfairness in the background; and that can be controversial, too. How much penalty should there be, for dropping out to have a child? There is additional difficulty in trying to compare and contrast jobs that are "traditionally male" versus "traditionally female" and which (among "male" jobs) may still have high barriers for entry. The last time that I saw a dollar-earned comparison, in was from a scoffer who hyperbolized, mis-cited, invented arguments, and generally insulted the statistics profession -- as if none of the studies, ever done by anybody, had ever controlled for anything. This was in the local newspaper. I keep pretty flexible in my expectations for the local newspaper. -- Rich Ulrich, [EMAIL PROTECTED] http://www.pitt.edu/~wpilib/index.html === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===
Re: misusing stats: examples
On Fri, 12 May 2000, Rich Ulrich wrote: snip Or, there are bad news reports, that don't really say what the study said. more snipping So: Here is another aspect of error -- what is reported in a journal, as opposed to what is claimed in a newspaper. The misinterpretation of results by the popular press has become a core topic for me in recent years. While some of the misinterpretations may be harmless (I doubt that eating extra fiber would hurt you unless it lulls you into a false sense of security about about your health). On the other hand some misinterpretations lead to all kinds of mischief. In recent weeks the press locally has jumped on the report that women earn about $.73 for every $1.00 that a man earns. This is being reported locally as the pay difference FOR THE SAME JOB! But the data are talking about the large aggregate (on average, if you will) not about folks within the same jobs. The public concern about this discrepency can lead to the passage of unnecessary legislation and a fair amount of public acrimony. I would agree that the misinterpretation of otherwise legitimate results is a major topic for discussion. Much more so than incorrect use of statistical procedures. MG *** Michael M. Granaas Associate Professor[EMAIL PROTECTED] Department of Psychology University of South Dakota Phone: (605) 677-5295 Vermillion, SD 57069 FAX: (605) 677-6604 *** All views expressed are those of the author and do not necessarily reflect those of the University of South Dakota, or the South Dakota Board of Regents. === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===
RE: misusing stats: examples
At 10:30 AM 5/15/00 -0500, Simon, Steve, PhD wrote: There have been a lot of interesting comments in this thread. Let me just add my two cents. Anyway, what I tell them is that nine times out of ten, the mistake was not in how the data was analyzed, but in how it was collected. After all, if you collect the wrong data, it doesn't matter how sophisticated the analysis is, does it? many MANY years ago ... early in the 80s ... i presented a little paper at a little conference at a little school titled: THE OVERRATED IMPORTANCE OF STATISTICS IN RESEARCH ... where i took a little 2 group experimental/control group design ... and listed some steps in this process such as: defining TARGET population taking a sample FROM that population subdivision OF sample into exp and cont groups IMPLEMENTATION of the treatment use of RELIABLE measures using APPROPRIATE analyses engaging in REASONABLE interpretations of the data and other things ... and tried to show that in the entire scheme of things ... that the process of using appropriate statistical analysis was the LEAST important of the batch ... and in fact, rather paled in importance when compared to the mess one can easily get into when one or more of the OTHER 'steps' has some flaw ... sometimes potentially fatal recovery from some INappropriate analysis is accomplished rather easily but, recovery from mistakes in the other areas is usually almost impossible to do ... i sometimes have a relook at this paper ... especially when we tend to get so hung up in trivial details of various analysis methods ... and especially when we try to make a big deal out of something like a p value ... for some specific test ... when i bet a quarter to a penny that there are so many other problems within the context of 'typical' studies as to make such p value squabbles rather silly ... Dennis Roberts, EdPsy, Penn State University 208 Cedar Bldg., University Park PA 16802 Email: [EMAIL PROTECTED], AC 814-863-2401, FAX 814-863-1002 WWW: http://roberts.ed.psu.edu/users/droberts/drober~1.htm FRAMES: http://roberts.ed.psu.edu/users/droberts/drframe.htm === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===
Re: misusing stats: examples
On 15 May 2000 08:58:41 -0700, [EMAIL PROTECTED] (Simon, Steve, PhD) wrote: ... "Here's a draft of what I have written." (review of article for Steve's Web site). On-line reference given for article. Thornley, Ben, and Adams, Clive "Content and quality of 2000 controlled trials in schizophrenia over 50 years" British Medical Journal 1998; 317: 1181-1184. Overview of research studies - Studies published between 1948 and 1997. - Patients with schizophrenia and other non-affective psychoses. Variety of interventions - Drugs (e.g., anti-psychotics and anti-depressants) - Therapy (e.g., individual, group, and family) - Miscellaneous (e.g., electroconvulsive treatments) Four difficulties 1. Types of patients - The ideal study would be community based. - Only 14% of actual studies were community based. 2. Number of patients - The ideal study should include at least 300 patients. - The average number was only 65 patients. - Only 3% of studies met the target of 300 or more patients. 3. Length of the studies - The ideal study should last at least six months. - More than half of the studies lasted six weeks or less. - Only 19% of the studies met the target of six months or more duration. 4. Measurement - The ideal studies should concentrate on a small number of standard measures. - These 2000 studies employed 640 different measures. - There were 369 measures that were used once and never used again. Conclusions Much of the work in schizophrenia failed to meet appropriate research standards. Too many of the studies... - examined the wrong patients, - studied too few patients, - ended too soon, - used fragmentary measurements. Research in schizophrenia leaves much room for improvement. my reaction to the article : Okay, I have been involved in research with schizophrenic patients since 1970. And I have been scornful about meta-analyses for about all the studies I have read with soft criteria, and this one deserves scorn, too. And it does not even try to average an outcome measure; it displays how badly one can draw conclusions just based on "lumping." A big problem is always the selection of studies. Here is a "meta-analysis" that reviews studies, over a 50 year period, which did not, hardly ever, try to be "controlled studies." Most had small N, followed patients for under 6 weeks (instead of over 6 months), and were not "blind" or double-blind. The big conclusion and criticism is that these studies had small N, short followup, and were not blind, etc. hmmm. This is, approximately, "all studies"? Why does he thing that long, expensive studies should predominate? (Would that not be an inversion of nature?) What, pray tell, determines the fitting mix of large studies and small studies? There is never, ever, *any* virtue in the smaller study, or shorter study? There is only one kind of allowable study? It would be more useful, I think, to take the set of studies that did *pretend* to be control studies. How big were they? What were their questions, and outcome measures? How many achieved useful results? I think that what the BJM published was a poor imitation of science. And, what *should* they say about the 95% that did not pretend to be controlled studies? -- especially if the N is small, time is short, etc., these must be totally, wholly worthless studies which have no justification for being published? -- unless these authors are overlooking some alternate ends I have worked on big studies. The study that I started working on in 1970 was drug versus placebo (plus a factor for social treatment), two years of followup with 374 outpatients, across 3 clinics. Note, this is the midpoint of the time period of these authors. But before we published a few years later, no one knew that drug would beat placebo! And it would keep on beating it, even after 6 months, and after a year! The N, by the way, was *far* larger than we needed for the original question, but it was large enough that we were able to spin off an important extra study: now that drug *did* (amazingly, unexpectedly) appear to be useful for two years, what would happen if we followed patients longer, when we took away their meds, after those couple of successful years? This article in the British Journal of Medicine is (IMHO) what Americans sometimes call "a hatchet job." Now I see that it may have helped to inspire and justify the non-funding of studies "because they don't have enough power, not having 300 subjects." I have read a hint of that before, and I thought that it was just malicious, bureaucratic double-talk, from someone opposed to spending money. I did not realize that the committee might consider themselves on the scientific-cutting edge, having read the BJM. Of course, the non-funding of studies with N *over* 300 is ever-justified because the studies would be too difficult, and/or would cost too
Re: misusing stats: examples
Gene Gallagher wrote: I have recently seen examples of the thrip fallacy in the op-ed pages of the Boston Globe. Massachusetts has implemented state-wide standardized testing and has increased state funding for school districts with low test scores. Statistical analysis reveals that Five or six socioeconomic factors (parents educational level, annual salary, % two-parent households, etc) account for over 90% of the variance in town-to-town K-12 standardized test scores. The implication is that only 10% of the variance in mean test scores COULD be due to differences in curriculum, teacher quality, or financing for the school (Take that Teacher's Unions!). Some might conclude that spending money on schools teachers since only 10% of the town-to-town variance in these scores could be due to factors outside the home. This fallacy fails to consider that a high median income and other socioeconomic factors often are strongly associated with a better tax base, lower class sizes, better trained teachers, more innovative curriculum etc. This fallacy should have a name, but I don't know it. I point my students to Wright's path analysis and structural modeling approaches (LISREL, and AMOS) to show alternatives to the misleading inference based on an R^2 in a multiple regression equation. There is an additional fallacy here (I think). As I understand it they used town-town means to infer a small effect of other factors on children's education. This is an example of the ecological fallacy. The town mean scores allow no firm inference about the effect of any factor on individual children (they could be similar in magnitude, different in magnitude or even in different directions). Thom === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===
Re: misusing stats: examples
- Original Message - From: Gene Gallagher Here is an error that is subtle, but very common. The statistical test (multiple regression) was applied perfectly, but the statistical inference was wrong. My first reference to this type of error is in the classic, but highly controversial, ecology treatise by Andrewartha Birch (1954): The distribution and abundance of animals, p. 580. These Australian ecologists wanted to show that animal populations aren't controlled by density-dependent factors like competition or predation. They regress 14 years of thrip (an insect) abundance vs weather variables. They considered weather a density-independent factor (mortality from a storm or a hot day isn't directly related to animal density). They conclude, "...altogether, 78 per cent of the variance in thrip maximal abundance was explained by four quantities which were calculated entirely from meteorological records. This left virtually no chance of finding any other systematic cause for variation, because 22 per cent is a rather small residium to be left as due to random sampling errors. All the variation in maximal numbers from year to year may therefore be attributed to causes that are not related to density: not only did we not find a "density-dependent factor," but we also showed that there was no room for one." Also, as both weather and population data tend to be autocorrelated, simple regression would tend to overestimate the significance of correlation (though not the correlation itself), by imputing more degrees of freedom than genuinely present. -Robert Dawson === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===
Re: misusing stats: examples
In article [EMAIL PROTECTED], Uplandcrow wrote: I teach research methods for social science at a small liberal arts college. The level of math in the class is low, I use Richard Black's "Doing Quantitative Research in the Soc. Sci." and excerpts from Gujarati's "Basic Econometrics." SNIP I am looking for examples of articles that use a stat procedure incorrectly. Here is an error that is subtle, but very common. The statistical test (multiple regression) was applied perfectly, but the statistical inference was wrong. My first reference to this type of error is in the classic, but highly controversial, ecology treatise by Andrewartha Birch (1954): The distribution and abundance of animals, p. 580. These Australian ecologists wanted to show that animal populations aren't controlled by density-dependent factors like competition or predation. They regress 14 years of thrip (an insect) abundance vs weather variables. They considered weather a density-independent factor (mortality from a storm or a hot day isn't directly related to animal density). They conclude, "...altogether, 78 per cent of the variance in thrip maximal abundance was explained by four quantities which were calculated entirely from meteorological records. This left virtually no chance of finding any other systematic cause for variation, because 22 per cent is a rather small residium to be left as due to random sampling errors. All the variation in maximal numbers from year to year may therefore be attributed to causes that are not related to density: not only did we not find a "density-dependent factor," but we also showed that there was no room for one." The logical/statistical flaw in the Australian thrip story was published in Smith, F.E. (1961) Density dependence in the Australian thrips. Ecology 42: 403-407. Since weather accounted for such a high proportion of the variance in the data (78%), AB assumed other factors could not be important. This is a fallacy. Smith argues that some density-dependent factors, unmeasured but probably correlated with weather, must be acting to control abundances. I have recently seen examples of the thrip fallacy in the op-ed pages of the Boston Globe. Massachusetts has implemented state-wide standardized testing and has increased state funding for school districts with low test scores. Statistical analysis reveals that Five or six socioeconomic factors (parents educational level, annual salary, % two-parent households, etc) account for over 90% of the variance in town-to-town K-12 standardized test scores. The implication is that only 10% of the variance in mean test scores COULD be due to differences in curriculum, teacher quality, or financing for the school (Take that Teacher's Unions!). Some might conclude that spending money on schools teachers since only 10% of the town-to-town variance in these scores could be due to factors outside the home. This fallacy fails to consider that a high median income and other socioeconomic factors often are strongly associated with a better tax base, lower class sizes, better trained teachers, more innovative curriculum etc. This fallacy should have a name, but I don't know it. I point my students to Wright's path analysis and structural modeling approaches (LISREL, and AMOS) to show alternatives to the misleading inference based on an R^2 in a multiple regression equation. One could experimentally demonstrate this fallacy by transferring students from affluent communities to communities whose schools have dismal standardized test scores. Somehow, I don't think the parents would accept the statistical argument that the their children's mean scores could decline at most 10% since 90% of the variance was due to socio-economic variables. -- Eugene D. Gallagher ECOS, UMASS/Boston Sent via Deja.com http://www.deja.com/ Before you buy. === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===
Re: misusing stats: examples
Ralph Johnson and J. anthony Blair are the authors of _Logical Self-Defense_. (I know, I did a year of grad study in Informal Logic with them.) At 8:13 PM -0400 5/9/00, Donald F. Burrill wrote: On Tue, 9 May 2000, Jerry Winegarden wrote, in reply to Uplandcrow's request: Uplandcrow wrote: SNIP I am looking for examples of articles that use a stat procedure incorrectly. A "MUST READ" for this class: "How To Lie with Statistics". Best little book in the world! Many wonderful practical examples of the misuse of statistics. With a heavy political season upon us and all the wonderful ads with the very compelling graphs, every citizen should be required to read this little book! :-) Also very good for every citizen: "Logical Self-Defense", by a couple of faculty members at the University of Windsor (Ontario, Canada, across the river from Detroit) whose names elude me. Uses lots of examples culled from the public press of errors in logical thinking. Some of the errors are essentially statistical; but even the ones that aren't ought to be in every teaching statistician's armamentarium of bad examples. -- Don. Donald F. Burrill [EMAIL PROTECTED] 348 Hyde Hall, Plymouth State College, [EMAIL PROTECTED] MSC #29, Plymouth, NH 03264 603-535-2597 184 Nashua Road, Bedford, NH 03110 603-471-7128 === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ === Jill Binker Fathom Dynamic Statistics Software KCP Technologies, an affiliate of Key College Publishing and Key Curriculum Press 1150 65th St Emeryville, CA 94608 1-800-995-MATH (6284) [EMAIL PROTECTED] http://www.keypress.com __ === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===
Re: misusing stats: examples
On Tue, 9 May 2000, Jerry Winegarden wrote, in reply to Uplandcrow's request: Uplandcrow wrote: SNIP I am looking for examples of articles that use a stat procedure incorrectly. A "MUST READ" for this class: "How To Lie with Statistics". Best little book in the world! Many wonderful practical examples of the misuse of statistics. With a heavy political season upon us and all the wonderful ads with the very compelling graphs, every citizen should be required to read this little book! :-) Also very good for every citizen: "Logical Self-Defense", by a couple of faculty members at the University of Windsor (Ontario, Canada, across the river from Detroit) whose names elude me. Uses lots of examples culled from the public press of errors in logical thinking. Some of the errors are essentially statistical; but even the ones that aren't ought to be in every teaching statistician's armamentarium of bad examples. -- Don. Donald F. Burrill [EMAIL PROTECTED] 348 Hyde Hall, Plymouth State College, [EMAIL PROTECTED] MSC #29, Plymouth, NH 03264 603-535-2597 184 Nashua Road, Bedford, NH 03110 603-471-7128 === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===
Re: misusing stats: examples
Found references to two of Hurlbert's papers: Hurlbert, S. H. 1984. Pseudoreplication and the design of ecological field experiments. Ecological Monographs 54:187-211. Hurlbert, S. H. 1990. Spatial distribution of the montane unicorn. Oikos. 58: 257-271. "V. Partridge" wrote: Look up papers by Stuart Hurlburt, who points up commonly-made errors in ecological research. Two of note are his paper on pseudoreplication (in Ecological Monographs, circa 1989, I think) and "The Spatial Distribution of the Montane Unicorn" (I don't recall the journal). V. Partridge Uplandcrow wrote: I teach research methods for social science at a small liberal arts college. The level of math in the class is low, I use Richard Black's "Doing Quantitative Research in the Soc. Sci." and excerpts from Gujarati's "Basic Econometrics." (FYI, if you have not seen Black's text yet, take a look. It is a wonderful teaching textbook, best I've seen) I am looking for examples of articles that use a stat procedure incorrectly. For example, I have one artivle from a business journal that conducts OLS but does not present any F or t tests or even standard errors. Yet the authors make inferences about their subject based on their results (essentially on R^2). In short, if you know of assessable articles which (in your view) misuse a particular method (especially descriptive states, ANOVA, OLS, logit, and probit) I'd be interested in the reference. Perhaps there is a web site you know of that deals with this? I am not out to denegrate anyone's research, merely to point out (common?) mistakes as a way to teach my students to be careful in their research. Thanks === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===
Re: misusing stats: examples
Look up papers by Stuart Hurlburt, who points up commonly-made errors in ecological research. Two of note are his paper on pseudoreplication (in Ecological Monographs, circa 1989, I think) and "The Spatial Distribution of the Montane Unicorn" (I don't recall the journal). V. Partridge Uplandcrow wrote: I teach research methods for social science at a small liberal arts college. The level of math in the class is low, I use Richard Black's "Doing Quantitative Research in the Soc. Sci." and excerpts from Gujarati's "Basic Econometrics." (FYI, if you have not seen Black's text yet, take a look. It is a wonderful teaching textbook, best I've seen) I am looking for examples of articles that use a stat procedure incorrectly. For example, I have one artivle from a business journal that conducts OLS but does not present any F or t tests or even standard errors. Yet the authors make inferences about their subject based on their results (essentially on R^2). In short, if you know of assessable articles which (in your view) misuse a particular method (especially descriptive states, ANOVA, OLS, logit, and probit) I'd be interested in the reference. Perhaps there is a web site you know of that deals with this? I am not out to denegrate anyone's research, merely to point out (common?) mistakes as a way to teach my students to be careful in their research. Thanks === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===
Re: misusing stats: examples
would you please compile responses and re-post them? Uplandcrow wrote: I teach research methods for social science at a small liberal arts college. The level of math in the class is low, I use Richard Black's "Doing Quantitative Research in the Soc. Sci." and excerpts from Gujarati's "Basic Econometrics." (FYI, if you have not seen Black's text yet, take a look. It is a wonderful teaching textbook, best I've seen) I am looking for examples of articles that use a stat procedure incorrectly. For example, I have one artivle from a business journal that conducts OLS but does not present any F or t tests or even standard errors. Yet the authors make inferences about their subject based on their results (essentially on R^2). In short, if you know of assessable articles which (in your view) misuse a particular method (especially descriptive states, ANOVA, OLS, logit, and probit) I'd be interested in the reference. Perhaps there is a web site you know of that deals with this? I am not out to denegrate anyone's research, merely to point out (common?) mistakes as a way to teach my students to be careful in their research. Thanks === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===
Re: misusing stats: examples
would you please compile responses and re-post them? Yes, I plan to. I've gotten 4 or 5 good sugestions and I want to look at the articles. Then I will post the citations and a summary. Cheers, Jon === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===
Re: misusing stats: examples
Uplandcrow wrote: I am looking for examples of articles that use a stat procedure incorrectly. A literature search of important journals in the subject area in which your students major might show that common problems have been addressed. For example, the articles below address problems that can arise with the application of multivariate models to clinical data. Concato J et al, Ann Intern Med 1993;118:201-10. Simon R Altman DG, Br J Cancer 1994;69:979-85. For example, I have one artivle from a business journal that conducts OLS but does not present any F or t tests or even standard errors. Yet the authors make inferences about their subject based on their results (essentially on R^2). Of course, there are situations when the above approach could be appropriate. -- Michael === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===